This morning I have read this article on Tech Crunch mentioning that – great news – one of the senior executives of Google has finally said it loud what we have thought for years: [most] software patents are detrimental to innovation. Only, he falls short of the point, because he asks for a "real patent reform," whereas the only suitable reform is a software patent abolition (if not an entire patent system abolition, but that is another discussion). All software patents are detrimental to innovation. Period.
But software patents are also detrimental to competition. Especially if they are used in an anticompetitive way. The Microsoft case, once again, gives us good food for thought and leads me to think that there is more than one antitrust concern over the sale of Nortel patents. May I attempt a few short answers to reasonable questions that are lingering around when I mention the antitrust issue. I plead guilty of omitting much of the necessary background.
Q: Nortel patents were there already, why should they create concerns?
A: because they have not been paid for their "market" value, but because of their strategic value. Buyers (or some of them) are arguably not interested in trading them at fair market value (estimated USD 800M), but as a way to stifle the competition from a disruptor with a different business model. You can't compete with a free product and you cannot trade your product for free? Make it a paid-for product and it's not that much of a disruptor anymore.
Q: But why should Google not pay for patents if anybody else is doing it?
A: Again, we must think out of the box. Google has bought a company, On2, just to have all of their rights on a codec, and donated it as WebM. It bid 3.14 (or pi) billion USD to have the same Nortel patents, four times their estimated market value (for the same and opposite reasons as the successful bidders). And it still planned to keep Android [F|f]ree. If the consortium valued them even more, it's because it thinks it can extract much more value out of them. That is, as I said, unrelated to the remuneration on the investment to produce an innovative idea (if any), it is about the strategic value.
Controlling a platform is much more lucrative than just the revenues deriving from selling some software. There is the network effect, everything is about network effects. Network effects are the air that inflates the IT bubble.
Q: yet isn't the "fair market price" the one that is set... by the market?
A: This is true, but only if the market is free to set the price and if you are trading quid pro quo. If you want to trade your patent rights in a normal fashion, you have a straightforward and correct game: you threaten litigation and ask for a price. If the price is too high, you face the risk to be forced to litigate, then you could face counterclaims an lose your patents, or simply lose the case and have no royalties. So you set a price as high as possible, but eventually settle for something reasonable, because if you set it too high you increase the odds of losing much more than your marginal gain. There is some balance in it, marketwise.
In the game where platform dominance and fighting against a disruptor are concerned, the game is totally different. You have more than an incentive to set the price arbitrarily high, eventually hoping the price is not met, or not bothering at all to set a price. This is the fundamental premise of all the antitrust-based forced licensing decisions, form Magill to Microsoft. You can end up with 4 different scenarios, at least, or some combination and permutations:
Scenario 1: the disruptor pays the arbitrarily high price, you gain an awful lot of money, yet this is your worst-case scenario, because still the disruptor might remain in good business.
Scenario 2: you don't trade with the disruptor, but with its clients. Maybe you start with setting some example by convincing the least reluctant to give in first. You therefore extract revenues from the clients of the disruptor and increase the cost its product. So you get paid and you diminish the attractiveness of the product for the disruptor. It's a win-win strategy: if the clients pay the price, you get paid and the product is less appealing; if you further increase the price, you don't get paid because you scare the client the disruptor out of its competing product, and you win even bigger.
Scenario 3: the disruptor refuses to pay, you sue it and you win: best case scenario, you put the disruptor out of business. The disruptor hence must find a way around your patents (and facing possible new litigation for a patent which held through judicial scrutiny) or it shall pay through the nose, if you are gracious enough to set a price, whatever price.
Scenario 4: the disruptor refuses to pay, you sue it and you lose. Yet again you win overall, because you have engaged your opponent through an ordeal of immense magnitude, scaring the hell out of its clients as in Scenario 2, you have obtained some revenues nonetheless through Scenario 2 and you have implanted the idea that your opponent's product is not that convenient as the price tag might suggest. This is a classic FUD strategy, and if your opponent is lucky enough, it will only last one-two years, reducing much of the snowball effect of its disruptive competition.
In all cases, the value extracted from the patents is totally unrelated to the merit of the patent, and it is not a fair market price. The situation presents a huge incentive to increase the price to a multiple of its fair market value.
As always, comments are welcome through my identi.ca federated instance: http://status.piana.eu/carlopiana