Patently Misguided

NOTE: I did a piece on patent reform for my post this month on the Wharton blog network.  They cut it significantly to fit their length requirements, so I’m posting the full version here.


Last month Google announced plans to acquire Motorola Mobility for $12.5 billion.  One would think this union of two legendary innovators would occasion some reflection on America’s history of technological achievement.  Instead, reporters and investors honed in on a single asset that made Motorola attractive to Google: its 17,000 patents. Motorola’s products, revenues, employees, brand, and supply chains were deemed insignificant. And Google isn’t interested in Motorola’s patents to create new products, but to protect those it already developed against lawsuits from other companies.  Something is wrong here.

Patents are good things.  They are important enablers of investment in new technologies and processes.  We need legal protections so that true inventors and not just opportunistic free riders benefit from the inspiration and perspiration involved in bringing valuable new products to market.

But let’s not kid ourselves about how new ideas develop.  From the calculus to the telephone, a shocking percentage of major scientific and technological innovations were created independently by two or more people at approximately the same time. In information technology, any significant product is a bundle of dozens or more ideas.  Apple’s singular Jobsian vision created the iPhone, for example, but the company stood on the shoulders of numerous giants and quite a few midgets in assembling its hardware, software, user interface, and manufacturing elements.  Moreover, successful technology innovations are usually as much a function of timing and marketing as fundamental technical acumen.  Patents, by their nature, protect small, discrete ideas, divorced from market context.

A patent is a legally-defined property right. Once you have a patent, you can sue for damages when a company infringes.  You can even get a court or an obscure international trade body to issue an injunction, forbidding that company from selling its product.  That’s an incredibly powerful competitive weapon.  If the legislature that defines the shape of the patent rights, or the Patent & Trademark Office that reviews patent applications, or the adjudicators that hear those claims, don’t strike the proper balance, the tax on actual innovations can grossly exceed the added incentives for new inventions.  And there’s abundant evidence that all of those institutions are wildly imperfect when it comes to patents.

The patent wars have only recently broken out in earnest.  For years, major technology companies such as IBM and Microsoft created an environment of mutually assured destruction.  Their deep patent portfolios allowed them to threaten a massive counter-attack against any assault.  That truce has broken down in part because of business shifts in the Internet market, and in part with the rise of “non-practicing entities” such as Intellectual Ventures that hold patents but don’t themselves produce products. Such intermediaries can help create more rational, liquid markets around intellectual property assets, but they can also suck licensing revenues out of the system with limited innovation going back in.

Making the problem worse is the flood of patents that should never have been granted.  The scope of patentability has been increased dramatically during the past three decades, in particular with regard to software and business methods.  As companies in information technology recognized the importance of patents for defensive purposes or as a lottery ticket, filing rates increased.  The Patent and Trademark Office has been overwhelmed by the volume and technical complexity of applications, leading to long delays and imperfect decisions about which patents to grant.

In certain information technology markets, patent litigation is becoming a preferred tactic of competitive business warfare.  Multimillion dollar damage awards are becoming commonplace.  Virtually every smartphone vendor is suing each other.  Venture capitalists fret that patent litigation will torpedo innovative small companies that have neither the time nor the resources to defend themselves adequately.

Or take Java, an important software technology developed by Sun Microsystems.  Sun wanted to promote widespread adoption of Java, so it didn’t aggressively go after companies that pushed the envelope on incorporating it into their products.  When Oracle acquired Sun, everything changed.  Oracle is now suing Google over Java in its Android smartphone platform. Oracle is within its rights to seek its day in court.  Let’s be realistic, though, and recognize this as a story of aggressive competitive tactics rather than anything remotely connected to rewarding or incentivizing inventors.

So, what can be done to tip the balance back toward innovation?  I don’t have room for a thorough discussion of such a complex subject, but here are some ideas:

Reduce the scope of patentability.  Software isn’t patentable in Europe, and business methods aren’t patentable in many countries.  Neither category lends itself to the solo inventor model of the patent system.  The personal computer industry developed just fine even though Apple, IBM, and Microsoft couldn’t assert patents on the graphical user interface (they could today).  And the reality is that software-based startups succeed because of effective execution rather than uniqueness of their ideas.

Split patents into industry categories.  A major impediment to patent reform is that not all industries face the same problems, because patents serve different roles in IT, manufacturing, and life sciences. The innovations and business ecosystems of Lipitor and the Blackberry are of a fundamentally different character.  Why not handle them with different kinds of patents, just as trademarks are divided by industry category, and there are special copyright rules for music.

Limit injunctive relief.  The threat that a court will shut down a business gives patent holders too much leverage.  It’s not necessary to redress the harm if a patent is infringed, and tilts the balance of power too far in one direction.

Crack down on abusive practices.  Some patent holders form shell companies, hide patents from business partners or standards bodies, shop around the country for advantages forum states to sue in, and otherwise game the legal system. Let’s have much greater disclosure of patent portfolios and ownership.

Improve patent quality.  One proposal with broad support is a “gold plated” patent, which would have stronger protection in return for a more searching review process at the front end.  Couple that with a better process to re-evaluate poor patents, and to leverage crowdsourcing techniques to find prior art (as in the Peer to Patent project).

Let the patent office do its job.  In the US, the Patent & Trademark Office easily pays for itself through filing fees. However, the budget the agency can actually spend is appropriated by Congress.  This political process starves it of the resources to hire, train, support, and compensate enough good examiners. Have the patent office run as a non-profit business that must get to break-even, but otherwise does whatever it takes to serve its mission most effectively.

The biggest impediment to patent reform is conceptual.  We need to recognize that patents are not the same as innovations, any more than a copyright is the same thing as a great story or song. If the current design of patents in information technology is failing to achieve the accepted goal of promoting innovation, let’s muster the courage to change it.