Suzanne Michel, Google interview transcript

What were you brought in to do?

Well, I started about three years ago; and at that time I had been at the FTC [Federal Trade Commission], where I had been doing some patent policy-type work. When I came in, the job was not completely well defined, as in: “Suzanne, you will do x, y, z.” But the idea was that I would work on policy initiatives, partly out of the DC office (partly out of DC because my family didn’t want to move, and so I’m really grateful for the company letting me stay here).

Was there a predecessor? How was your brief handled before?

No – the role didn’t exist in terms of what I do. That’s not to say that Google was not involved in patent policy issues; Google certainly was before I got here. It’s just that there wasn’t a dedicated person who did only that. So Michelle Lee, who at that time was essentially chief patent counsel, for instance –she had been involved in working with other companies in terms of what the tech industry was hoping to get from the AIA [America Invents Act]. Michelle and some others – Tim Porter included – would make decisions on what kind of positions to take on amicus briefs, and there were other people also working on similar such issues. But just by having one dedicated person for those issues, I’m able to coordinate with the rest of the group and take some of the burden off who is now Allen.  

How was the job presented to you and what goals were set?

It is a new role, but a lot of the other companies I work have a similar person. At Microsoft, for instance, Dave Jones has a similar job; Tina Chappell at Intel; Gail Levine at Verizon – so it wasn’t a new creation. I think the need for it was somewhat clear and there was a little bit of the right place at the right time. Google was reaching a point where, like I say, these activities were being handled by people on the patent team doing very well; but you really need a person who has more time and bandwidth to coordinate if you want to engage to more fully and it was a good time for me in my personal life to transition from government.

So I reached out and we had some good conversations about how it would be great for Google to be doing more of what it had been – not necessarily a completely different kind of activity, but be able to help make the team more effective, coordinate with other companies, things like that. But we didn’t specifically define the role at the point I came in – I think at least my thought was, “This is a really interesting opportunity and the job will be shaped as I’m in it, depending on what the company needs.”

So how does it look now?

I can give you a sense of the issues I work on. So I deal with amicus brief:  when Google files an amicus brief in court, some times we start organising, but we generally reach out to other companies and work with other companies on which positions to take and we try to file briefs with multiple companies listed. Sometimes other companies approach me and say, “Hey I’m thinking we should file in this case, have you heard about this case? Do you want to work together on it? Do you want to work on forming a position?” So that’s a big part of it. I work on comments to the patent office – any time you’re working on a project like that, there’s the coordinating with other companies; there’s the coordinating with our folks internally – it isn’t my decision, I coordinate with the right people on the patent team and talk about it. And then there’s the congressional piece – working for reform legislation by talking to people at other companies, talking to trade associations, trying to form the substantive positions so I’m more on the substance side than on the lobbying side, if that makes sense.

Since I’m on the patent team and I’m a lawyer and I report up through Allen, I deal more with substance. What positions should the company take? What positions are other companies taking? How do we want to coordinate with other companies on what works and what’s doable?

So you come in, AIA is just signed – was it clear at that point that things would be moving to a second piece of patent reform and that the company was keen to push that?

No, I would say quite the opposite. The phrase ‘patent fatigue’ was being used quite a bit – not just at Google, but everywhere. When I came in, AIA had just passed; the statute said that within one year the patent office will implement regulations to make all these things work. So it was clear the patent office was going to be asking for input and comments on what those regulations should look like. So a lot of my time in the beginning was: this is big, this is important, let’s figure out what input we want to give the patent office on how, for instance, the PTAB [Patent Trial and Appeal Board] procedures should work and all the other bits that were coming out of AIA. So that was really the focus.

The idea of new legislation, actually, if anyone suggested,  “Should we ask the Hill for legislation?” what you always heard is, “Congress has patent fatigue.” When I was at the FTC, I testified at the very beginning of AIA, which I think was 2007 and it passed in 2011. It was at least a five-year process and I think there was no expectation in September 2011 that Congress would become interested in patent legislation for quite a while.

But where did the company feel that there were gaps in the legislation?

The tech industry – not just Google – certainly had a sense that a lot of initiatives they were pushing for in legislation fell by the wayside and were not incorporated in AIA. So I think there was still a sense in the tech industry that more needed to be done; but again, the feeling in September 2011 was that there would be some time before those conversations could start happening.

But when those conversations started to happen the next year, what did you put the catalyst down to?

It was the SHIELD Act, I think.

But did that take you by surprise? You must have known the Oregon representative was working on it?

Yes – at the point at which he working on it, certainly his office was doing a lot of outreach to many companies talking about what would be workable. My understanding is that Representative Fazio was still hearing from constituents, “I have a troll problem – help.” He was hearing that and he wanted to help on that issue, and so he put together the SHIELD Act and it ended up being very much a conversation starter and an indication to a lot of people on and off the Hill that the troll problem wasn’t getting better, it was getting worse. The AIA wasn’t going to stop that problem and we needed to have another conversation about the patent troll problem, so I think it was very much the catalyst.

My other sense of an important piece of the dynamic, given that AIA had passed in September 2011, you can hear in statements of Representative Goodlatte, for instance. When he said he was going to engage on patents again, one of the points he made was that there were good ideas “left on the cutting-room floor and I want to revisit some of those now”. His concern for the situation in understanding that it was a litigation abuse problem, and that that should be addressed directly and there was really nothing about that in the AIA, was also a very important part of the dynamic of the getting legislation rolling again. 

And how quickly did you become a part of that conversation?

Like I said, Representative de Fazio’s staff was in contact with many companies and Google is a member of the Coalition for Patent Fairness, the Internet Association, a couple of other groups that are all very concerned with the troll problem and the litigation abuse problem; and so we were talking to those companies there and as a group really talking about strategy, talking about substantive positions there.

And so what became the key issues as the prospect of a second piece of reform became a more realistic goal?

Where we’ve seen the legislation go, and I think is very valuable, is to deal with the litigation problems. There are multiple causes of the troll problem; there are multiple root causes. The fact that litigation is expensive, inefficient and prone to abuse is one those problems – it’s not the whole problem, but the litigation piece of the root cause is something that Congress can address and is probably well suited to address. It also seems an issue that Representative Goodlatte was engaging on. That’s why I think you see the focus on litigation issues there, as opposed to patent quality issues that the courts and patent office can deal with.

And what particularly about litigation abuse?

My own view is that what’s needed is a strong package that will make a difference. That will help address the problem that we see a lot – and it’s especially bad for smaller companies – which is the patent troll who says, “It’s going to cost $1 million to fight me; give me $500,000”; or if you’re a smaller company, “It’s going to cost you $100,000 to answer my complaint; give me $50,000.” Those kinds of nuisance-value kind of cases are very much determined and driven by the cost of litigation. And cost is not just monetary value – it’s engineer distraction, it’s executive distraction, it’s having to turn over trade secrets in discovery… There are a lot of downsides to litigation even beyond the straight paying out of money, and so what’s needed is pieces that help address that problem directly.

I don’t know if ‘x’ absolutely has to be there – there’s no one provision that I would say absolutely has to be there; but there needs to be a strong package of several provisions to make a difference. And the provisions that were in the bill that the Senate was working on – variations of which came out of the Innovation Act from the House – are all good ideas. There maybe others that would be great to discuss and it will be interesting to see how they play out.

If you look at what happened in the courts – the Supreme Court in particular in the last term, as this debate was going on – they heard a number of issues that reform was trying to get to. From your perspective, where do you feel Congress is most effective and where are the courts most effective?

So there are different problems in the way that patents operate in the technology industry. There’s a significant patent quality problem; and then there’s litigation that is inefficient and overly expensive and can lead to that arbitrage problem. Those are two separate problems with two different sets of solutions. And so while we think it would be very helpful for Congress to address the litigation problem, it’s unlikely that Congress is going to have the means or the appetite to really take on the quality problem.

And there are two quality problems – one is that there are bad patents out there and too many very bad patents; and then the other quality problem is that you want to improve the quality of what’s coming out right now from the patent office. So what the courts are dealing with more is the quality side than the litigation side – or at least, what the Supreme Court has been dealing with in terms of the Alice decision and the Nautilus decision are both about quality: both identifying patents that shouldn’t have issued in the first place but are already out there, and then setting standards that increase the quality of patents coming out of the patent office.

But you have to realise that a patent troll can take a patent and you may be, ‘I think you’re totally going to lose on that patent,’ but it’s still pretty valuable for extracting a nuisance settlement. And you’ve got to deal with the litigation piece, not just the quality piece, to have an overall better functioning patent system. So we very much do need both: we need Congress to address the litigation issues; we need the courts to address the proper standard for issuing patents going forward and helping to define what kind of patents are bad-quality patents.

What do you mean by ‘bad-quality’ patents?

There are many aspects to that question – you can break that down quite a bit. Again, I think this is a tech industry problem – and a software industry problem in particular. I want to be clear this is not an attack on the patent system writ large – it’s not an attack at all, but it’s certainly not an attack on the patent system writ large; that the problems Google experiences with the patent system, and that all tech companies experience and tech start-ups are experiencing, have  some commonalities that are very different from patents in the biotech and pharma industry.

We see patents where it’s very hard to tell what the patent covers, or you read the patent and there’s a lot of uncertainty about how it will be interpreted and whether or not it will be ultimately interpreted to cover your product, and that shouldn’t be the case. A company, an engineer, a patent lawyer ought to be able to pick up a patent and say “Here’s what it covers” – that’s really not the case. We need to think about how the courts have set the standards for patentability and why are we in that situation. Patents are supposed to promote innovation and in a lot of instances they very much do. But when a patent is awarded for work that is trivial or that would have happened whether or not there was a patent system, you’ve really created a tax on someone else’s work later; and so setting the bar high enough for getting a patent, not giving patents out too easily, is really important to make sure that the system works the way it’s supposed to. So I think… the obviousness issue, where’s the bar on that? And we also see patentable subject matter – what kind of ideas should you be giving patents on in the first place? – as part of ensuring that patents are doing a good job in our industry promoting innovation rather than just being an added cost.  

So, if you look at how the Supreme Court has come in on the issue of patent quality – you could also argue that inter partes review are pushing towards greater patent quality – what are you left on in terms of what you would like action on in patent quality? Or are you happy to say the trend lines are moving in the right direction, Congress doesn’t need to step in?

Like I said, I think Congress isn’t stepping in on patent quality – none of the proposals on the table is focused on patent quality…

But it’s certainly part of the narrative. Why are we doing this? We’re doing this because there are too many broad patents out there being litigated by trolls or non-practising entities or other companies.

Right; but the point of that narrative is that there’s a lot of litigation. A lot of the litigation is focused on bad patents, so you don’t want to say, “Well that’s the price of the patent system, right?” That litigation is not a problem if they’re good patents, some people might say. What the narrative means is not, “Congress, please clean up bad patents necessarily or patent quality”; but that it is a problem that it’s possible to make money litigating and threatening litigation on patents that should never have issued. One of the reasons it’s possible to make money doing that is tied to expense and inefficiency of litigation, and so the discussion around legislation is to address that piece. That piece has really got to be handled separately from the “Let’s get rid of bad patents” piece, so I think you see the courts starting to address, “Well, what kind of ideas do we want to give patents to and what should be the standard for granting these patents?” That’s not going to do a lot for the 2 million patents already out there; they still have to be litigated one at a time and all of those dynamics around litigation exist whether the patents are probably invalid or not.

One area that the court did rule on was fee shifting, which was obviously a part of reform, and it lowered the bar or relaxed it. Are you of the mind that fee shifting is at the right level, or are you wait and see, or do you think that we still need this to be part of any discussion in Congress?

 I believe you’re going to see that a lot of the groups that want legislation also want fee-shifting provision to be part of that legislation. And the reason for this is that the Supreme Court decision was necessarily tied to the current statute. The current statute says fee shifting in an exceptional case and no court is ever going to change that – they can’t change that, that’s in the statute. So I would agree with you that the Supreme Court lowered the standard on what’s considered an exceptional case; it’s still a fairly difficult standard to meet just in the interpretation of the phrase ‘exceptional case’. A lot of the proponents of patent reform feel that a more objective and slightly lower bar would give you a better-functioning patent system, and so the drive for fee-shifting legislation will still be there. But to be clear, no one is asking for fee shifting in every case and I don’t think that proposal is even on the table.

How could you characterise it, if you took ‘exceptional’ out? ‘Exceptional’ points to a very small percentage of cases being subject to fee shifting – would you like the percentage to become double figures?

I couldn’t give you a number and we haven’t talked in terms of, “Gee, 20% would be the perfect number.” What’s important is the deterrent aspect in some ways from fee shifting in filing unreasonable suits and taking unreasonable positions and pursuing cases beyond the point at which they seem reasonable. And I think that’s certainly a standard that a lot of companies are willing to have applied to them even as defendants, even as plaintiffs. The exact language of what that looks like will be part of the political process and part of…  Again, there’s a lot of companies in a lot of different industries thinking about what the best way to phrase that is and we’ll have to see how that turns out.

Why do you think it was unsuccessful?

There are articles that were written I guess in May when Senator Leahy decided not to hold a mark-up, so I’d refer you to those articles – I don’t have a sense of that. But I will say, I think, given the history of patent litigation in the US – you know, AIA was a six-year process at least, that the bill before that was a five-year process – it would have been unprecedented to get legislation in the patent sphere in the amount of time that was spent on it. I don’t actually think of the legislative movement as being unsuccessful, but rather taking a break for the moment.

Would you anticipate a similar multi-year process to get to legislation?

I hope not… But it had been about a year, that’s pretty short. Congress had been really engaged, and people were looking at drafts and making suggestions and a bill passed the house, and then we were talking in the Senate. That whole process was only about a year and that’s really short – nobody wants this to go on for five years, but a year would have been short. And so my point being it’s been too soon to say that it failed, rather than we’re on a break. 

Senator Leahy gave perhaps the most open interview post the delay or the shelving, and he lay the blame at Senator Reid’s feet and said Reid made it clear they weren’t going to get it to the floor. Does it become easier then if the Republicans take the Senate?

You know, like I said, I’m the patent lawyer geek, so I might not be the best person to comment on that. You certainly hear that idea thrown around. On the other hand, I’ve also heard – and again, I’m the patent lawyer geek – the way the Senate works it’s not as if any one party can ram any one thing through. The Senate is not designed to work that way and so certainly both parties have a lot to say about any bill that ultimately would go through.  

But are you preparing for a second effort?

I think the full expectation, even just based on Senator Leahy’s statements, Chairman Goodlatte’s statements, there is an expectation of a second effort.

And when you look at that and you look at Goodlatte’s proposals and the areas around discovery, transparency, fee shifting, ultimate financial involvement, what are, for you, the red-line issues – the ones you feel have to be part of a package?

We have not staked out any one issue that has to be there or doesn’t have to be there. Again, it’s about an effective package – you could have six provisions, but they’re all watered down so much that none of them are effective. You could have a couple of provisions that are super-effective…

But if you could have one?

There’s a lot of room to work there; and the other point is that Google works with a lot of other companies to reach consensus through all these different organisations and we’re going to continue to do that, and so ultimately I don’t feel it’s for me to say, “This is the one thing we have to say.” We’ll be working with our tech compatriots; I think we’ll be hearing from the retail industry; so there’s some flexibility there.

But you must have a sense that there’s one thing that will be the most damaging to patent trolls or whoever’s bringing these cases against you? Or what’s most effective in solving this litigation abuse problem?

My own person view is that there is no silver bullet, which your question somewhat suggests.

Not that there’s a silver bullet, but that there are inevitably going to be some measures that are going to be more effective than others and there’s probably going to be the most effective changes…

You know, because I think it’s the political process and there are many, many stakeholders with many different perspectives, I don’t want to say there’s one particular provision that’s the most effective and has to be there. Really, the most important thing is the package, in my view. And I think that this is actually a helpful point that there are different ways to put together an effective package, and that I think creates some optimism that there’s a path forward that can accommodate interests from different stakeholders – and by that I mean the retail industry and the tech industry and the pharmaceutical industry. That’s what we have to work towards, and so an ultimatum on any one provision at this early stage seems to me not helpful and ultimately very much depends on what particular members of Congress want to do to, and obviously they have their ideas and that’s paramount.

What do you say to the claim that what you’re really most interested in is not simply weakening the position of the worst litigation abusers, but weakening the position of patent-owning plaintiffs overall?

Well, I would vigorously dispute the fact that Google has any interest in weakening the position of patent-owning plaintiffs. You know, as an owner of a very large patent portfolio and a company that’s developing some very cutting-edge technology that it wants to license out, that just wouldn’t work. I also think, from a personal perspective, what’s interesting is an appreciation of how patents are very important in certain industries or for certain kinds of inventions. There’s certainly no desire to undermine that and it would not be a good strategy to undermine that. I think the path forward is to respect and understand the full complexity of the patent system, and try to work within it to fix the problems and not to mess it up where it is working.

But there’s this sense that there’s this patent troll narrative now that has been very effectively communicated that is actually using patent troll in a very broad way, not just to describe the worst litigation abuses. President Obama in his Google hangout said “They don’t actually make anything” – you quoted that in your blog last year – and therefore when you have this broad qualification, Bessen and Meurer research as well and the figure that they have put to the cost of this, it actually puts a blanket over a far, far wider pool of patent owners than simply what we would class as the worst kind of litigation abusers. And when you are a part of that narrative – I’m not saying that you started it, but you certainly helped perpetuate it –you are actually getting at a much broader class of patent owners than, as you would class it, the worst litigation abusers.     

What the goal of litigation is and what the goal of other aspects of patent reform are is to identify the root causes of the problem and to address the root causes. You can think of patent trolls as the symptom, not the disease. It’s helpful to understand what’s actually causing this behaviour. What is it in the patent system that makes hedge funds want to invest in patent litigation? I think that’s an important question for society to think about, because wouldn’t we rather have hedge funds invest in biotech start-ups? Wouldn’t that be better for the economy overall than investing in patent litigation? And that’s what we’re facing: we’re facing a world in which litigation investment is seen as a high return on investment kind of proposition – I don’t think that’s good for anybody.

Understanding why that is is helpful in understanding the most rational solution to the problem. Some of the reasons are that it’s the cost of litigation; it’s the discovery system that allows your classic patent assertion entity to own a patent, to have essentially nothing to discover or very little discovery costs themselves, but able to impose very broad-ranging discovery requests that are expensive and distracting, and leverage that to extract a settlement. That’s the root cause. Anybody can do that – you don’t have to be a patent troll; anybody can impose those kinds of discovery costs – but it tends to be more of a problem in these asymmetric situations. So that’s why one of the proposals in the legislation was: what if the two litigants exchanged the documents that are probably going to cover 90% of the issues, the core documents – after that, pay for what you ask for? You can get in discovery whatever you’re entitled to under the federal rules of evidence but you’ve got to pay for it, just so that the request will be tied to something that’s really needed.

Those are the kinds of solutions that I think don’t hurt anybody who is a legitimate patent owner trying to enforce property rights or solve a dispute, but can undermine gamesmanship which is about inflicting as much expense as possible in order to extract a settlement. So it’s not targeted at one class of patent owners versus the other ,but it’s the kind of solution that really understands the underlying problem and addresses it in a very direct way.

To what extent is it a problem of quality or valuation?

That’s a separate problem. So I just talked about the way that discovery expenses can be used to extract money in a way that becomes a tax on very innovative companies and not really supporting IP rights in a meaningful way. You just asked about what I think is the third problem, which is valuation. What a patent is worth is a very difficult issue in the tech space and it has been for a long time, and I think we have a long way to go to figure it out. But it’s that classic problem of the product is probably covered by 100,000 patents – one patent in litigation, what’s that really worth? And what kind of economic tools do you need to figure that out? We don’t have those economic tools yet; courts have made some improvements, but we’re in my view a long way from being able to have a good, rational economic valuation in that very complex situation, and so that is one piece of the patent troll problem.

There’s one kind of patent troll that you sometimes hear people call the ‘lottery ticket’ troll – you know, “I want $500 million” - that’s a different troll business model than I want to extract discovery cost settlements. But we see both; different companies see those different behaviours in different proportions. And so a big part of having a rational patent system that really works for innovation would be better economics and better case law around damages absolutely; but that’s not part of legislation.

Why would a company like Google want a strong patent system when actually a lot of what you have been about is open innovation and a lot of what you’re about as a search business is driving people to ads on your search platforms and in other media – so why would you want a robust patent system?

Google wants a patent system that works for innovation in our industry – that is the goal. The innovation ecosystem that Google participates in is much bigger than your question. Google, through Google Ventures, supports start-up companies; Google makes many acquisitions of start-up companies – as do a lot of big tech companies, it’s a lot of the way that innovation is done both in the tech industry and the biotech industry – and we bring important new ideas into the company and develop them further and bring them to market through those kind of acquisitions. Google develops contact lenses that measure glucose and licenses them out – all sorts of things.

You really have to understand the full picture, I think, to understand what Google thinks about the patent system and ultimately what all of those different pieces have in common is innovation and what kind of patent system is going to support innovation well in all of these different areas – and again, both strategically and not mess up the patent system where it is working. I don’t there’s really much debate among anyone; there are problems with the way the patent system works in the tech industry, so let’s figure out a way we can fix that, but not throw the baby out with the bathwater.  

But where have those problems been a brake on innovation in the tech industry?

Right – you’re asking me to prove a negative, in some ways.

I’m just interested because a lot of people say this and yet for the last 30 years this has probably been one of the most innovative periods in human history and the patent system has supported that, and I’m just interested in evidence of it being a brake on innovation. And I appreciate that it’s hard to prove a negative.

And then people say that the patent has supported innovation in your industry and we also need to see that, and I think both questions are very important and equally valid. In terms of where we’ve seen problems, there have been studies of start-up attitudes towards the patent system and sometimes they feel as much pain from the patent system as they have benefit; and they do, particularly, have some patent troll pain now, so that’s one particularly important point.

It’s different in different places, right? There were articles about start-ups saying, “This patent case just put me under.” You look at… and this is not to say there shouldn’t be a patent system, it’s to say, “Can we do better?” It’s hard to say we couldn’t do better. I haven’t heard anyone say, “No, this is the perfect patent system for the tech industry; it can’t get any better.” Certainly, the amount of litigation, the amount of defensive patenting just to navigate the waters – those are expenses that companies face, and you see not just Google, but a lot of tech companies questioning, “Is this the best possible patent system given what it’s costing?” And no, I think we can do a little better. That’s completely different to throwing it out; but we’re just looking to fine tune the system and I can’t see any downsides to fine tuning the system in an intelligent way – it’s got to be better for innovation than not doing it, given that we are witnessing problems.