Thursday, April 30, 2015

EPO president reportedly threatened to resign; Dutch diplomat concedes concern over bad press

Today's EPO staff demonstration in Munich (see my previous post) turned out a lot more interesting than expected.

An estimated 700 protesters marched from one of the EPO's Munich buildings (the closest one to the Oktoberfest venue, for those who may have visited Munich on that occasion) to the Dutch consulate-general:

There was one sign that I particularly liked:

It's unfortunately true that the EPO, claiming immunity, behaves like an enclave that doesn't have to respect European/EU human rights and labor law standards. I don't think the EU institutions can turn a blind eye to this situation. Before the EPO starts granting EU-wide unitary patents, it must at least meet a certain European minimum standard in terms of checks and balances as well as access to justice.

After a 15-minute walk, the protesters arrived at the Dutch consulate-general. They were greeted by the consul-general himself, Mr. Peter Vermeij, who was an EPO vice president (in charge of patent administration and other areas of operational support) from 2007 to 2012. I took this picture of Mr. Vermeij adressing the SUEPO crowd (with the microphone):

Mr. Vermeij acknowledged that the ongoing labor conflict at the EPO and the bad press it has already resulted in have (among others) the member states of the European Patent Organisation concerned. He said that the EPO should usually just go about its work and there should be no noise about it. This official concession of concern over negative publicity was interesting.

Mr. Vermeij invited a few staff representatives to his office to discuss the Dutch government's position (also on the human rights issues on which a Dutch court had sided with SUEPO) in private.

Toward the end of the demonstration, a staff representative said, citing a reliable but unnamed source, that a majority of the EPOrg's member states (at a Council meeting last month) was in favor of appointing an independent mediator to help resolve the sitation but EPO president BenoƮt Battistelli was adamantly opposed to this idea and threatened with his resignation for the event that mediation would have been imposed on him. The crowd ironically cheered.

As I've said several times before, the EPO has structural governance issues that must be addressed, though Mr. Battistelli's leadership style is certainly unpopular with staff. While I don't think Mr. Battistelli's resignation would in and of itself represent a solution, I also don't feel he's unreplaceable. The Administrative Council should have decided to take the mediation route anyway and should simply have accepted his resignation.

The EPO is clearly in crisis.

If you'd like to be updated on the smartphone patent disputes and other intellectual property matters I cover, please subscribe to my RSS feed (in the right-hand column) and/or follow me on Twitter @FOSSpatents and Google+.

Share with other professionals via LinkedIn:


Promise of union recognition insufficient to solve conflict at European Patent Office: new protests

By promising (after approximately four decades) formal recognition of EPO staff unions, the Administrative Council of the European Patent Organisation extended an olive branch, presumably due to a combination of political (with some key member states increasingly hesitant to support EPO president Battistelli) and legal (Dutch court decision) dynamics. Last week the kick-off of the "renewed social dialog" took place. According to SUEPO (Staff Union of the European Patent Office, which published a report on the meeting (PDF), "[t]he atmosphere was such that some frank exchanges about the situation in the Office could take place." But staff representatives still appear to be skeptical of whether any meaningful change will result from all of this.

It's clear that formal union recognition won't solve any problem. It can be seen as a gesture of goodwill, and indirectly it could have positive effects if the talks helped build a consensus, but for now there are no signs of the situation actually improving. In the most critical respects it seems to be "business as usual", with certain reforms being implemented at any rate.

It could be that the Administrative Council hoped staff representatives (particularly, but not only, SUEPO) would soften their stance on the actual issues because of the potential benefits to their organizations from formal recognition. Should that have been the plan, it doesn't appear to have worked out: SUEPO organized a march today from one of the EPO's Munich buildings to the Dutch consulate (PDF flyer).

Again, I don't know whether the Administrative Council overestimated the impact of the promise of formal recognition and the invitation to talks, but in any event the representatives of the EPOrg's member states should consider that they are not dealing with "your average trade union" such as in a traditional manufacturing industry, where there may sometimes be a disconnect between union leaders (and their personal interests) and most of the people they speak for. EPO examiners are very educated people who can tell the difference between window dressing and real change.

The day before yesterday SUEPO published another flyer, which explains some key underlying issues (this post continues below the document):

15-04-28 SUEPO Flyer Brave New EPO by Florian Mueller

There are three key things that this flyer explains:

  • The stated reasons for certain reform measures are based on the nonsensical notion that the EPO "competes" with the USPTO, JPO, SIPO and other non-European patent offices. If it competes with anyone, it's with national patent offices, but national patent systems control the EPO through the Administrative Council and milk it (through high renewal fees that have an almost 100% gross margin for national patent systems).

  • Increased productivity pressures on staff require and inevitable result in a lowering of patentability standards, particularly with a view to the inventive step. Ultimately, this is a very problematic development that can have negative economic effects (except for the EPO and, especially, the national patent systems controlling it).

  • While the EPO is highly profitable, with a budget surplus of €364 million in 2014 and a likely higher one in 2015, it still doesn't lower its fees. Instead, the EPO leadership argues that more (not better) patents must be granted. The question of how many patents Europe needs (or, as SUEPO asks now, how many it can tolerate) came up before. SUEPO now also pointso ut that it would be a fallacy to assume that more EPO patents mean more European innovation or growth:

    "Two-thirds of the applications filed at the EPO are not of European origin and thus are more likely to hinder European industry than benefit it. A flood of badly examined patents could affect in particular the small and medium-sized enterprises that cannot afford expensive litigation."

I agree with SUEPO on all of that. There's only one thing that SUEPO has said in connection with today's protest (in a PDF flyer published on SUEPO's website) that I disagree with:

"Last week Mr Battistelli informed us that the Dutch government will join the EPO in its attempt to overturn the judgment in the next instance ('cassation'). If so then the Dutch government makes itself complicit in violating fundamental rights."

(emphasis in original)

In my opinion, the Dutch government is in its right to express its position on the legal question of EPO immunity, and if it agrees with the EPO on this one, then that's legit, even if staff representatives don't like it. I view the Dutch government's role as, practically, an amicus curiae as not objectionable at all, but it should accept the final outcome even if the previous judgment is affirmed, meaning that a final judgment in SUEPO's favor should also be enforced.

Also, the Dutch government should play a more constructive role on the Administrative Council to ensure that EPO staff have certain human rights. In one of the related contexts the IPKat blog pointed out an interesting fact:

"People who work for the World Intellectual Property Organization (WIPO), the Office for Harmonisation in the Internal Market (OHIM) and the Community Plant Variety Office (CPVO), the Benelux Office for Intellectual Property (BOIP) and the good folk whose job it is in the European Commission to make life difficult for us by thinking up new IP policies, must all have the occasional health issue too, and presumably WIPO, OHIM, the CPVO, BOIP and the Commission must have schemes that govern the health and welfare of their own employees -- none of whom, so far as Merpel is aware, have publicly complained about the health provisions that apply to them."

The EPO situation indeed appears to be unique.

I'd also like to point to this IPKat post on a subsequently-withdrawn European Commission statement (a pretty bad propaganda piece) on the proposed fees for the single European patent. I'll talk about this issue on some other occasion. I know there's a lot of unhappiness about this one in industry and in the legal community.

If you'd like to be updated on the smartphone patent disputes and other intellectual property matters I cover, please subscribe to my RSS feed (in the right-hand column) and/or follow me on Twitter @FOSSpatents and Google+.

Share with other professionals via LinkedIn:


Tuesday, April 21, 2015

Ericsson's pseudo-sale of patents to Unwired Planet and the rampant problem of privateering

In the first half of this decade, the biggest and most divisive issue in the information and communications technology (ICT) was FRAND: fair, reasonable and non-discriminatory licensing of standard-essential patents (SEPs). That issue hasn't gone away completely. Courts and regulators provided some clarification (in Microsoft v. Motorola, for example). Still, companies will continue to disagree on what constitutes a FRAND offer, on the proper royalty base, on the extent to which standard-setting organizations (SSOs) such as the IEEE should provide guidance, and on circumstances that may or may not warrant an injunction over SEPs. But all of this has a lower profile now.

The biggest ICT patent issue in the second half of this decade is "privateering": the act of large companies feeding trolls with patents in order to maximize their patent monetization income and/or drive up their competitors' total cost of defense. Quite often this raises FRAND issues as well. Many privateering deals involve SEPs and are part of a scheme to circumvent FRAND licensing obligations.

Centuries ago, privateers were pirates who colluded with governments. I recommend two articles, a Bloomberg story from 2013 and a more recent blog post by the von Mises economic institute, that discuss the problem and start with how Queen Elizabeth I of England commissioned Francis Drake to plunder Spanish merchant vessels. Privateers had to share their booty with the governments that backed them.

Today's patent-based privateering is a rampant problem plaguing the industry. The Bloomberg article I just linked to mentioned transactions involving patents held by Alcatel-Lucent, BT, Ericsson, and Nokia--and those are just a few examples. Furthermore, privateering is one of the issues the U.S. Federal Trade Commission is investigating in connection with patent assertion entities (PAEs).

Patents are transferable assets. There are good-faith, genuine patent transfers--and there are transactions of the kind that is styled as a sale but in commercial terms comes down to an arrangement under which a patent assertion entity is essentially a licensing and litigation service provider to the operating company.

Let's start with what a genuine asset sale looks like. If a company or private person sells a used car, it ceases to have any ongoing interest in the car. Such an agreement doesn't restrict the ways in which the purchaser uses the car; it doesn't even prevent the purchaser from selling it to a third party; and the buyer has to pay a (specific) price.

Naturally, a patent transfer has certain additional aspects. If third parties have already been extended a license to a patent or a portfolio, the acquirer must know about and respect those licenses. Also, the seller may want to continue to use the patented inventions. And the purchase price may have a variable component that gives the seller some upside if the patents prove more valuable than originally expected. But apart from that, the structure of a genuine patent sale resembles that of a genuine car sale.

On the website of the Security Exchange Commission (SEC) of the United States I have found a filing--a redacted version of a "Master Sale Agreement" between Ericsson and Unwired Planet over more than 2,000 wireless patents--that is essentially a privateering case study. I looked this up since Unwired Planet's German lawsuits against Google, HTC, Huawei and Samsung over six of the related patents will go to trial in the coming months. It was much easier than I thought to google the deal terms.

I wish to point out that it's not my objective to engage in "Ericsson bashing." Just like last year, when I wrote about an Ericsson slide deck that explains the (bad) reasons for which the Swedish company doesn't extend patent licenses to chipset makers, it's about behavior that is by no means unique to Ericsson. Ericsson (or, in this case, Unwired Planet) just happens to be particularly transparent about it. Even Apple once sold patents to a non-practicing entity (which I believe Apple wouldn't do again since it has meanwhile had to defend itself against at least one privateer). The undisputed number one patent troll feeder in the world is not Ericsson but another Northern European company: Nokia has engaged in various such transactions, and if I were an antitrust regulator, I'd want to ensure that Nokia won't sell any of Alcatel-Lucent's patents to patent assertion entities after the closing of that acquisition. I'll talk some more about Nokia's privateering on another occasion.

I've read the redacted version of the Ericsson-Unwired Planet deal in detail. A couple of structural elements are clearly very different from a traditional sale.

Section 3, Purchase Price, does not state any amount Unwired Planet had to pay upfront. Instead, Ericsson receives a percentage of whatever revenue Unwired Planet will generate with those patents:

(i) 20% of the amount of Cumulative Gross Revenue, until the Cumulative Gross Revenue equals $100,000,000; plus

(ii) 50% of the amount of Cumulative Gross Revenue in excess of $100,000,000, until the Cumulative Gross Revenue equals $500,000,000; plus

(iii) 70% of the amount of Cumulative Gross Revenue in excess of $500,000,000.

The relatively low percentage Ericsson receives on the first $100 million makes it easier for Unwired Planet to recuperate its litigation expenses. Then the percentage goes up to 50%, and above $500 million (i.e., in the event of a significant licensing success), Ericsson receives the lion's share: 70%.

Percentages like that remind me of what I've read about agency deals. For example, when sports bodies commercialize the broadcasting rights related to their events through agencies, this kind of revenue sharing is normal. But this is just not the way a "sale" in the traditional sense works.

The agreement also comes with all sorts of restrictions on the acquirer's business, a right of first refusal on any sale of the patents to a third party, and a "poison pill" for the event of an acquisition: in the event of a change of control of Unwired Planet, Section 3.3 allows Ericsson to terminate the agreement and to demand a "Sale Payment" for its patents, which according to 3.3(c) shall "in no event [...] be less than an amount equal to (x) $1,050,000,000 minus (y) the aggregate amount of all Quarterly Payments actually received by [Ericsson before]." It can be even more based on a valuation of those patents at the relevant time.

Let's connect the dots: Ericsson is convinced that the patents transferred under that agreement are worth, at a minimum, more than $1 billion, but it "sells" them to Unwired Planet based purely on a percentage of future revenues.

If Ericsson had kept those patents, it would have licensed them to other companies as part of its portfolio. That would have been most efficient for everyone, but Ericsson doesn't want to optimize efficiency: it seeks to maximize its patent monetization income, and it apparently believes that even a substantial de facto "agency fee" it pays to Unwired Planet is more than offset by the incremental bottom-line licensing cost to the likes of Google, HTC, Huawei, and Samsung. Ericsson may even think that its portfolio is so large that 2,000 patents more or less don't make any difference for the royalties Ericsson can obtain from licensees, but it's a large enough portfolio that Unwired Planet can go out and demand some additional percentage that is then shared with Ericsson.

Schemes like that give patent licensing--and patent transfers--a bad name.

If you'd like to be updated on the smartphone patent disputes and other intellectual property matters I cover, please subscribe to my RSS feed (in the right-hand column) and/or follow me on Twitter @FOSSpatents and Google+.

Share with other professionals via LinkedIn:


Friday, April 10, 2015

The Google that has joined Via Licensing's LTE pool is the real Google--not the FRAND abuser

Yesterday's announcement that Google has agreed to make its LTE (4G) standard-essential patents, all or most of which previously belonged to Motorola, available through Via Licensing's LTE patent pool, is for all intents and purposes more meaningful than most settlements of patent disputes between industry players are. It means that Google is being Google again, and has distanced itself from the abusive conduct that gave rise to antitrust investigations in the U.S. (consent order) and Europe (decision, but no fine, which also looks like a compromise).

By "Google being Google again" I mean that the search giant and Android maker has given up on the notion that two wrongs (overwhelmingly meritless, at least impactless, patent assertions against Android on the one hand, and retaliatory abuse of FRAND-pledged standard-essential patents on the other hand) could make a right. Google temporarily used its SEPs in ways that ran counter to positions it was simultaneously taking on non-SEPs, even to the extent that Google, the parent company, submitted a public interest statement in an ITC proceeding that was amazingly inconsistent with what Motorola, the subsidiary, was saying in its own submission in connection with a parallel case. Also, Google's (ab)use of SEPs didn't really lend credibility to its positions on patent reform.

Now the "Don't Be Evil" company has apparently decided to become consistent again. In the LTE context Google is now aligned with companies that have never abused FRAND-pledged SEPs, some of which have even made significant efforts to advocate reasonable interpretations of FRAND. These are the other contributors to the Via Licensing LTE pool: AT&T, China Mobile, Clear Wireless, Deutsche Telekom, DTVG Licensing, Hewlett-Packard, KDDI, NTT DOCOMO, SK Telecom, Telecom Italia, Telefonica, and ZTE. Temporarily, Google's (Motorola's) FRAND positions were actually the same that some infamous SEP trolls and some failed businesses with an increasing or near-exclusive focus on patent licensing (and the privateers they feed with patents) tend to take.

This is the very Google that more than any other (IT) industry giant wants to make the world a better place through investments that the stock market doesn't reward in the short term but Google can afford, such as self-driving cars and "interventions that enable people to lead longer and healthier lives." More than any other large company I know, Google is truly about much more than just making money for its shareholders and other stakeholders. (Of course, this still doesn't give Google the right to violate antitrust rules in its core business, or to leverage excessive control over Android in anticompetitive ways.)

I find it hard to believe that the timing of the announcement--one day after an appellate hearing at which Google saw that it can't win the FRAND part of its Microsoft dispute, though it is winning the real war over Android royalties--is a coincidence. In that Microsoft v. Motorola case, Google's lawyers consistently argued that patent pool rates should not be used as an indicator of FRAND rates. Different standards (H.264 and WiFi) are at issue in that case, but still: Google wouldn't have wanted to undermine its anti-pool-rate argument. It certainly didn't want to give Microsoft's counsel the chance to mention its new position at the hearing, and if it had seen any realistic chance of Judge Robart's FRAND rate-setting opinion being overturned, it might have waited (possibly forever) with this move. Yes, this is speculative, but the connection is close and strong enough to support such a theory.

It's not clear whether Apple and Microsoft (which builds LTE devices as a result of the Nokia acquisition) will be able to benefit directly from Google's contribution to the Via Licensing LTE pool. With Apple, Google has a ceasefire in place, but no license deal (at least none that would have been announced). With Microsoft, it is still embroiled in litigation, though Microsoft hasn't brought any new offensive cases against Motorola in a while (at least none that would be discoverable).

Maybe Google's agreement with Via Licensing precludes Apple and Microsoft from licensing Motorola's LTE patents through that pool until comprehensive license agreements between those companies and Google are in place. In that case, Apple and Microsoft could still use the Via Licensing pool rates as pretty powerful evidence in any FRAND rate-setting dispute with Google.

Maybe Google doesn't even care if Apple and Microsoft license its LTE patents through that pool. Google could still assert older (3G) patents if necessary, as long as those haven't expired. And with the lack of success of Apple and Microsoft's patent assertions against Android devices so far, Google may not even be afraid and, therefore, may not feel it needs any leverage from LTE patents to counterbalance Apple and Microsoft's non-SEP enforcement. Apple started its enforcement against Android more than five years ago; Microsoft, more than four-and-a-half. After all this time, Google may be convinced that it doesn't need a good offense as its best defense because a defense-defense will always do the job to protect Android.

The main reason I used to criticize Google's position on patents so much was its inconsistency. It wanted to devalue non-SEPs while trying to gain undue leverage from SEPs. At first sight, one could also say that Apple and Microsoft are inconsistent because they want to bring SEP license fees down while exaggerating the value of non-SEPs. But FRAND-pledged SEPs are encumbered, and encumbrance is not a value enhancer. Consistency obviously depends on the particular arguments that are used and on whether any differences in value have a logical basis. For example, Apple is in my opinion being inconsistent by stressing its royalty base and "smallest saleable unit" point in connection with SEPs but arguing that even minor aspects of minor features (where the price of the smallest saleable unit, Android, is technically zero) make a substantial percentage of the entire value of a smartphone.

Overnight, Google has gone from "most inconsistent" to "most consistent" when it comes to patent licensing. This overnight change took years of litigation and a couple of antitrust investigations. Still it's great news, and I hope that some others, such as Apple, will soon match Google's level of consistency in this regard.

If you'd like to be updated on the smartphone patent disputes and other intellectual property matters I cover, please subscribe to my RSS feed (in the right-hand column) and/or follow me on Twitter @FOSSpatents and Google+.

Share with other professionals via LinkedIn:


Thursday, April 9, 2015

Appeals court not inclined to toss $14.5 million FRAND breach verdict against Google's Motorola

More than 4½ years after Microsoft sued Motorola for not renewing a license agreement (a renewal that would likely have been costly after Motorola stopped making Windows devices), Motorola still hasn't paid Microsoft a dime in patent royalties on its Android devices, and for now there's no reason to assume that Motorola will come under serious legal pressure in the near term to do so. What Microsoft may, however, receive before or around the 5th anniversary of the dispute is a check over $14.5 million--not in Android patent license fees, but in damages for Motorola's abuse of its (own) standard-essential patents. This would effectively increase Google's cost of acquiring Motorola by slightly more than 0.1%. This amount is certainly much less than whatever Google spent on legal fees related to Motorola's spats with Apple (where a ceasefire was agreed upon almost a year ago, with the exception of Google's continuing fight against the European rubberbanding patent) and Microsoft. I'm sure (though I don't know what Microsoft demanded from Motorola in 2010) it's also a tiny fraction of what Microsoft would have wanted Motorola to pay in royalties.

A Seattle jury's $14.5 million damages award for breach of the duty of good faith and fair dealing will likely become enforceable in a matter of months, given that the United States Court of Appeals for the Ninth Circuit appeared, at a hearing held yesterday, unlikely to grant Google's wish for a reversal or retrial. After a panel opinion, Google could theoretically ask for a rehearing and/or file a petition with the Supreme Court. The former would hardly help, and the latter wouldn't make sense for Google, which has other priorities than this matter. So, all going as expected, it will be "game over" for the FRAND breach story soon.

As a part-time intercontinental litigation watcher who prefers not to travel I appreciate the Ninth Circuit's second-to-none degree of transparency. A video recording of the appellate hearing is available on YouTube (this post continues below the video):

Google previously appealed a decision in this case, from the Western District of Washington, to the Ninth Circuit: a temporary-restraining-order-turned-preliminary-injunction that barred Motorola from enforcing a couple of German patent injunctions. Google lost in 2012 and really, really didn't want to come back. But the Federal Circuit rejected Google's attempt at appellate forum-shopping, and yesterday Motorola's counsel, Quinn Emanuel's Kathleen Sullivan, had to face the very same panel as last time: Chief Judge Thomas, Senior (literally--he's 86) Circuit Judge Wallace, and the author of the 2012 opinion, Circuit Judge Berzon. The latter played by far the most active role on yesterday's bench, suggesting to me that she'll also write the next opinion.

None of the judges expressed any opinion on the (un)reasonableness of Motorola's 2010 royalty demand from Microsoft, which technically amounted to $4 billion a year. The focus was on the procedures that led to the jury trial and the verdict. But I can't imagine that the judges wouldn't consider it ridiculous, and there's no indication that they have a problem with District Judge Robart's FRAND rate-setting opinion, which came down to about one-twentieth of a percent of Motorola's original demand.

Google complains that the discrepancy between Motorola's initial demand (they still prefer to describe it as an "offer letter" and say Microsoft also rejects 99% of all initial licensing offers it gets) and Judge Robart's rate, which was presented to the jury as law of the case (so Google was unable to challenge the underlying rationale), was tantamount to a directed verdict. My interpretation is that the judges were telling Google between the lines: "It's your problem that the only conclusion at which the jury could arrive based on these numbers was unfavorable to you. That still doesn't mean Judge Robart treated you unfairly in the process."

I'm not surprised about that. However, I didn't expect the following: throughout the Seattle proceedings, Motorola had tried to stall and derail. It was against pretty much everything Judge Robart did, except when he denied some Microsoft motions. But the appeals court apparently can't find a clear and timely objection by Motorola to the course of events that resulted in a jury verdict against the background described before. The case was bifurcated (FRAND rate first, breach damages later), and Google now says that any consent was only limited to the possibility of the court creating an actual license agreement, not to an isolated rate-setting opinion (since there are various non-monetary terms that go into a full-fledged license agreement) that was subsequently used as law of the case in a jury trial on damages.

The judges indicated to Google's counsel that they disagreed with her, and nothing she said appeared to change that. By contrast, the message between the lines to Microsoft's counsel, Sidley Austin's Carter Phillips, was like "we agree with you and just want to be sure that we didn't miss anything important." You can see on the video above how comfortable he was with the overall situation.

Mr. Phillips stressed that the way Judge Robart introduced the rate-setting opinion into the jury trial and instructed the jury on breach was as favorable to Motorola as it could have been under the circumstances (one circumstance being that the rate determination was law of the case). I think that's right. Judge Robart didn't disadvantage Motorola, and he didn't do some things he could have done, such as granting Microsoft summary judgment on breach. This made it harder, or at least more time-consuming, for Microsoft to prevail. But it also upped the ante for Google's appeal, which became very clear yesterday.

If you'd like to be updated on the smartphone patent disputes and other intellectual property matters I cover, please subscribe to my RSS feed (in the right-hand column) and/or follow me on Twitter @FOSSpatents and Google+.

Share with other professionals via LinkedIn: