The monopoly strategy behind the Google/Microsoft mobile patent wars
pluralistic.net
pluralistic.net
There is no credible economic theory of firm “power seeking”. What’s more, you don’t need such a theory to explain what’s going on here.
If you want to explain why Google paid so much for Motorola, you can do so via the various defects in intellectual property law.
All large firms are at least occasionally involved in patent litigation, frequently with patent trolls (though sometimes with large competitors too). This is familiar enough to everyone on this site. For Google this was a defensive acquisition, but a defense against frivolous litigation. There is no need to invent some theory of firm “power seeking” to make the case.
He discusses his view of economics in this twitter thread here:
https://twitter.com/doctorow/status/1450136949327425537?s=20
- First of all, I have no interest in or patience for Ayn Rand. I don't know any economists who take her remotely seriously. All of the American economists I know are on the left politically. So are most of the non-American economists I know. There are, of course, prominent exceptions to the rule.
- I don't know why we are accused of having no theory of prices? I teach that in week 2 of my principles of microeconomics class. I don't teach inflation b/c I don't teach macro, but they do have theories.
- We don't assume people have perfect information, with the possible exception of using it as a very simple benchmark case on our way to something more complicated and realistic. Incomplete/imperfect information has been a central feature of all theoretical and empirical work in the profession for more than 40 years now.
- He shares a post about concentration in meatpacking. I agree that we have allowed many markets to become too concentrated, including meatpacking! I am in favor of stronger antitrust enforcement! My entire sub-field is empirical studies of imperfectly competitive industries.
- Skipping several tweets (there is really too much here to argue with every point...) we have talked about the CORE textbook in my department. Most of us put it on our syllabi. It is free, after all. As a comment on Doctorow's take: every Principles textbook mentions the environment because every Principles textbook talks about externalities. I spend a week on the subject.
Again, I confirm my impression that he doesn't know anything about economics or competition policy, but studying it more seriously would be to his benefit, especially in commenting on technology and regulation issues.
"What the market will bear." is a nonanswer to them even though it is accurate because reality is messy with human psychology more so.
Of course from mainstream economics perspective their concepts of value and exploitation are between fallacious (universal global value because labor is assumed as the basis) to pants-on-head levels of crazy (the idea that a fair trasaction becomes unfair when the buyer is able to reap later profit).
How is the theory of market power combined with everything else we know not that? [1] It seems obviously in the firms interest even if their primary motive were still profit since it ensures increased rates of profit (modulo risk of a regulatory smackdown.)
Indeed, I said I thought he was ignorant of economics in part bc he brought up this weird “capital as power” thing which has nothing to do (AFAICT) with actual market power as we understand it.
Furthermore, in this particular case I think you get a much better, more direct explanation of Google’s behavior vis-a-vis its acquisition of Motorola via intellectual property and patent litigation than via market power, let alone via “capital as power”.
https://en.wikipedia.org/wiki/Regulatory_capture
https://www.goodreads.com/book/show/34397551-the-chickenshit...
If everyone in a village economy is capable of growing their own food on the side for about the equivalent of say $15/bushel averaged out but you have the specialization, mechanization to grow it for $1/bushel margins. Even if you wind up the only one growing food others still have their fields.
If we call increasing efficiency "power seeking" then the term has lost any and all meaning and only serves to confuse and make the most mundane actions sinister by interpreting them in the most conspiratorial light. If you drink commodity coffee in the morning you actively to exploit the rural poor, if you go with more expensive fair trade coffee you are engaged in conspicious consumption of veblem goods while people starve, and if you just drink water you are trying to starve the rural poor.
Sure, but simply calling an acquisition like the one discussed in the article "increasing efficiency" interprets it the best possible light, which seems equally in need justification.
We don't call increasing efficiency "power seeking." Yes, I think many anti-capitalists would respond to that by saying that "there is no ethical consumption in capitalism" and that these problems cannot be solved by consumer choices.
Yes, boys and girls, "power" is sometimes a motivator. There are other ones, though. Maybe read a few more books.
The Bork view on anti-trust definitely has favored these companies, and Apple and Google can absolutely be accused of having and using monopoly power to keep competitors out. Google pays to be the default search engine on iOS (keeping competitors out), and Apple denies fair platform access to software it thinks has a chance at competing with it's ecosystem (Tile, etc..). Amazon and Facebook can be accused of this pretty easily too (restricting 3rd party sellers, restricting API access respectively).
The law[1] is pretty clear on this, but bills exactly like [2] and [3] would clear it up more, which I believe Cory Doctrow supports, at least from reading his Twitter.
[0] https://openinventionnetwork.com/
[1] https://www.justice.gov/atr/competition-and-monopoly-single-...
[2] https://www.congress.gov/bill/117th-congress/house-bill/3825...
[3] https://www.congress.gov/bill/117th-congress/house-bill/3816...
As for the Bork view: all those things you mentioned are "consumer harm" and are not protected. It would be a lot more fruitful to look at the lobbyist spending by the tech giants, and the Astroturf "studies" they fund.
I hope the current lawsuits against these companies over the exactly the things I mentioned should be easily decided then, and we won't need the bills I linked.
They buy a patent which looks good, i.e. applies broadly, hasn't been litigated unsuccessfully, has a few more years to run, ideally has been tested by an unsuccessful IPR. The whole goal is to make back the investment in a few years. Contrary to what someone said, there is not an infinite supply of those.
Then they pick out easy targets to sue, i.e. companies which just settle without much fuss. Those settlements don't have to be for large amounts, and typically are not.
I should correct the word "sue." Often there is no lawsuit. If a defendant is formally notified that they might be sued, they can seek a Declarative Judgment (DJ) in a friendly venue, which affirms that they're not infringing. The troll does not want to be DJ'ed. So they just threaten without really threatening ("hey, you might want to take a look at this patent!").
Those easy targets build up a war chest, which the troll then uses to go after the bigger and harder targets. These are companies that will fight back, often with an IPR.
So the moral is: make yourself a hard target. Never roll over for those people.
"This is a great strategy in light of US anittrust [sic] theories over the past 40 years, during which time antitrust regulators promised to leave companies alone as they formed monopolies, provided they didn't hike price after attaining monopoly dominance."
No, Cory, that's not what happened and there was no "promise." The new theory of anti-trust was "consumer harm." Hiking prices is certainly one form of that, but so was forming a monopoly. Microsoft was prosecuted for forming a monopoly and reducing consumer choices.
"Google's true operating costs aren't captured by adding up its salaries and servers – a correct accounting must include the costs of acquiring companies and with them, patents. That's the cost Google must incur, if is to retain its power."
No, Cory, those are not "operating costs," those are called "capital costs." They're treated real differently in Accounting.
I was part of this "acquire patents" effort at Google. I actually interviewed people to be in Legal whose job would be to acquire patents. The theory (since discarded) was that a heavy patent portfolio would give you a better negotiating position if you were sued. There was no thought of actually asserting these patents against Microsoft, or anyone.
I actually used some of our internal tools to assess our patent portfolio. Most of those Motorola patents were worthless.
This is particularly important when your view of the underlying issue is actually correct, because if you post like this about it, you end up discrediting not only yourself but also the truth [1], and that hurts everybody. We're really trying to avoid that sort of internet dynamic here.
I certainly appreciate that you have inside knowledge about this topic and are willing to share it with us! That part is great. I also understand (believe me) the frustration that comes up when people make glib claims about a situation they don't have first-hand knowledge of. But if you wouldn't mind reviewing https://news.ycombinator.com/newsguidelines.html and taking the intended spirit of the site more to heart, we'd be grateful.
[1] https://hn.algolia.com/?dateRange=all&page=0&prefix=true&sor...
Disclaimer: I work at Google, and I'm used to keeping things polite when the workforce here is attacked on the forum. So not sure why the reverse can't also be true.
Cory is a well networked, fairly thorough, informed blogger. He participates in all sorts of panels on topics of regulation, listens to a wide amount of voices. His views typically reflect & meet with a wide variety of other expert opinions.
It's really unfortunate that such easy, cheap character assassination is being done against someone who has spoken so well, with such unvarnished genuine interest & care for what technology can be in this world.
Doesn't it kind of prove part of the argument?
Google apparently felt threatened by patents, enough so to acquire "second strike capabilities" as deterrence.
Seems to me Google though patents aggressive tools with the purpose to hurt other companies or threatening to do so (though defensively in this case).
I am really interested in the "since discarded" part, if someone can recommend some reading.
I'm curious: Have you heard of the state's "monopoly on violence" before? It comes up a lot in political theory discussions, and it applies to this attempted rebuttal by analogy.
Especially in light of stuff like "Google training documents advise avoiding monopoly language": https://www.theverge.com/2020/8/7/21358441/google-alphabet-t...
The literature says that in mature markets, the major players all have cross-licensing agreements sharing their giant patent portfolios, shutting out the minor players. They form an oligarchy, in other words.
I may be a few years behind, having been retired for a few years, but I don't think this has actually happened. It's possible that Google and Apple and Microsoft really do have cross-licenses now, but I kinda doubt it. It would be too complicated to negotiate. What they have is mutually assured destruction.
I'm guessing he's aware that's how things work currently; it sounded to me like he was arguing that it shouldn't work that way, although I don't know if he literally believes that accounting practices should change or if that was more just for rhetorical effect.
Interesting! Can you expand on this topic? Does/has the strategy work?
The main problem with this strategy is patent trolls. Some company that doesn't make anything buys up some broad patents. Then there is nothing you can sue them over because they don't make anything so the strategy doesn't work.
It also makes the whole concept of software patents pretty useless, because the "defensive patents" could equally be used to counter anyone who tries to sue you for violating a "legitimate" patent. So the whole system becomes a bludgeon for larger companies to use against smaller companies, because they have more patents, even if most of them are low quality, which means they can tie up anyone who actually makes anything in court indefinitely.
As for big companies suing smaller ones: citations (in the mobile space), please?
It is within a space of possibilities but a sustained lack of evidence decreases the probability of it being so as numbers increase. It is possible that a given person drove drunk without being caught. But if they drove drunk everyday throughout many different towns and villages? Less likely.
Also, how many times does this have to happen before it's a problem? Maybe it's not thousands of times. If it happens each of the small number of times that someone would otherwise have the opportunity to unseat a monopoly, that's still really bad.
> No, Cory, that's not what happened and there was no "promise."
Borkism's "consumer harm" theory has, in almost all cases, been centered around hiking prices. Given the lack of interference we've seen for the past 40 years, whether or not there was this "promise" seems moot: there has been a pathetically low level of anti-trust enforcement. Whether promised or not, there has certainly been a delivery.
That you can cite a single 20 year old Microsoft case does little to convince me that there is efficient & active anti-trust regulation. It was a pretty heinous act, by Microsoft, at a time when there was only one mainstream consumer OS. But now that there's two OSes, there's no need for anti-trust, seemingly, and Apple is allowed to engage in far far far worse behaviors on iOS with regards to only permitting Safari technology to run, hampering the web. Social media companies have faced no scrutiny in their bonanza of competitor purchasing. The semiconductor market has undergone a radical vast wave of consolidation, again, seemingly entirely unchecked. Whatever the new Robert Bork "consumer harm" standard might grant it license to regulate, it certainly doesn't seem to be doing much. Proving those harms is much harder & more specific a task than what came before, and has, imo, been enormously clearly for the worse, as Cory continues to rightly argue.
> No, Cory, those are not "operating costs,"
The entire basis of this post is that the accounting doesn't reflect the truth of the matter. The accounting of acquisitions as a one time capital costs isn't really genuine or fair, when a constant ongoing stream of acquisitions is required to maintain the power & position of dominance. The dominant player has to keep feeding the meter to continue operating from a position of power. Each of those may be a one time cost, but the macro behavior is that this is not so different than the grid hookup keeping the data-centers running. It sustains the dominance of the entity. It's an operating cost, the cost of staying where they are.
Your point about patents is entirely besides the point. This isn't about companies using courts to try to destroy one another- something we haven't ever really seen at scale in the modern world. This is about Saturn Eating His Son. This is about the gigantic titaneous overlords of the planet insuring nothing new ever happens, preventing disruption. By acquiring all freshness & devouring it, bodily.
OK, do you want some more? How about [1]?
If your point is that the tech giants have not faced enough anti-trust litigation: you'd have to define "enough" as "sufficient to break them up." Because they're still here.
The EU has been all over them, and almost every state AG is joining in an anti-trust suit against them. Congress investigates them every week, and we might eventually see some legislation. Democracy is slow. But yeah, more is better.
Your last two paragraphs rely on a false premise: "a constant ongoing stream of acquisitions." If there are acquisitions going on, it's not for patents -- it's to eliminate competitors. That's not a defect of IP laws; it's a matter for legislation and anti-trust action. Which I support.
[1] https://www.ftc.gov/enforcement/cases-proceedings/151-0065/s...
Anyways, the correct link is https://pluralistic.net/2021/10/20/vizio-vs-the-world/#capit....
In THEORY, when the fragment is (semantically?) significant, the site could dynamically alter the link rel="canonical" value to include the fragment such that you could check whether to keep it or not, but in practice I think that canonical links are intended for web crawler deduplication and so nearly always strip fragments off.
About the only thing I could suggest would be to add a checkbox to the submission form that controlled fragment-preserving vs stripping, but you would probably need to add a review queue to double check it wasn't being abused (not that abuse is a particularly attractive target, in this case), and I am unsure whether the minor additional code and UI complexity is worth it vs. the ad-hoc process you have now.
A second possible approach would be to invert that suggestion and simply add a review queue to check whether a removed fragment should have been preserved instead (ordering or filtering the queue by the rank of the submission rather than chronological order, for sanity purposes).