Lately we’ve seen many complaints about the proliferation of patent holding companies, which buy patents, usually from small inventors, and then try to extract royalties, by negotiation or lawsuit, from companies that (allegedly) use the patented inventions. Often this is depicted as some kind of outrage. But from a policy standpoint I don’t see a problem.

Now perhaps you believe that the patent system is irretrievably broken and ought to be scrapped or severely reformed. Perhaps you think it should be harder to bring patent lawsuits. If that’s your position, then your policy effort should be spent on reforms that apply to all patent owners and all lawsuits, and not just on holding companies. Why focus specially on patent activity by holding companies, unless your goal is to disadvantage small inventors?

If, on the other hand, you buy into the goals of the patent system, and you think that the system, though imperfect, generally works, then it’s hard to see the problem with holding companies. It seems sensible that the financial return for an invention ought to be the same, whether the inventor works for a big company or freelances in his garage. If the invention really is novel, non-obvious, and useful, then the inventor is entitled to reasonable royalties from people who use the patented technology. Why should small inventors face barriers that large inventors don’t?

An inventor’s ability to negotiate royalties depends, ultimately, on the threat that he will bring a lawsuit if the company using the invention doesn’t agree to pay. Patent litigation is costly and time-consuming, especially if the defendant is using delay tactics. A freelance inventor can’t credibly threaten to bring a suit without financial backing from somebody else. Litigation is risky, too, and the inventor may be risk-averse. The company using an invention knows these things, so a freelance inventor’s lawsuit threat won’t have much credibility, even if the suit would have merit. And so the freelance inventor won’t be able to extract the royalties that a deeper-pocketed inventor could. It’s often argued that the patent system unfairly favors large companies, for precisely this reason.

Why not allow an outside firm to invest in small inventors’ patents, so as to provide the financial resources to support a potential suit and to absorb the risk? Coming from such a firm, a lawsuit threat would have suitable deterrent value. And so, most importantly, suchs will bid against each other for small inventors’ patents. Holding companies can level the playing field by helping small inventors extract the true value of their inventions.

Beyond this, holding companies may develop expertise in patent valuation or negotiating royalties. Holding companies that specialize in valuation and revenue-extraction allow small inventors to specialize in what they do best, which is inventing. This would mirror the structure in large companies, where one subgroup of people handles invention and another handles revenue-extraction. Why treat the small inventor differently from the large one?

Though there is no good policy argument for disadvantaging small inventors, we may see such changes anyway, due to rent-seeking by large companies. Those who support rational patent policy should focus on setting up the right patent rules (whatever they are), and applying those rules to whoever happens to own each patent.

A two-year-old internal memo from Hewlett-Packard predicted that Microsoft would soon launch patent-infringement suits against companies that distribute open-source products such as Linux and Apache, according to a Joe Barr story at NewsForge. (The article reprints the memo in full.) The memo is clearly based on statements made by Microsoft negotiators during a patent licensing negotiation with HP.

My first reaction to the story was that this was just a negotiating ploy by Microsoft, to scare HP into granting better patent licensing terms – an implicit threat to sue HP if you they rejected an offered deal. But after reading the whole memo, that seems unlikely. Apparently the eventual agreement did not protect HP against the threatened suits; so raising a false alarm about such suits would have made the negotiation harder for Microsoft, not easier.

Another possibility is that Microsoft really did intend at the time to file suits. If so, the question is why, two years later, no suits have been filed. Slashdot posters suggested that the SCO-IBM suit is providing the FUD (fear, uncertainty, and doubt) about open-source that Microsoft was hoping to create with its suits, rendering a Microsoft suit unnecessary. But if some FUD is good for Microsoft, why isn’t more better?

Perhaps Microsoft changed its mind, because it couldn’t find a strong enough patent to assert, or because it feared patent-infringement countersuits from the intended targets of its litigation. This one also seems unlikely. If Microsoft’s goal was to create FUD and run up a defendant’s litigation expenses, then even an iffy patent would suffice, and Microsoft would have no trouble at all finding a patent to use. And if the goal of a suit was just to create FUD, then it wouldn’t much matter who they sued, so they would certainly be able to find somebody without a big patent portfolio to go after.

Most likely, the lawsuit threat was a ploy, designed to create FUD at HP about its use of open-source products. And the HP memo is evidence that it was working; the author clearly wanted to reduce HP’s “exposure” from its use of open-source.