In the last blog I recited extensive findings from a 2003 IPO survey to the effect that 88% of the corporate respondents considered trade secrets as their really important intellectual assets. Interestingly, Dennis Crouch in his March 12 blog, offers a summarization written by James Bessen and Michael Meurer of their recent book Patent Failure. In it they note that

“Surveys also find that in most industries (pharmaceuticals are the exception!) R&D managers report that lead time, goodwill, trade secrecy and other means of appropriatia are more effective than patents in obtaining returns on their R&D investments (Levin et al. 1987, Cohen et al. 2000). For this reason, it is not surprising that survey research also finds that most inventions are not patented (Arundel and Kabla 1998, Cohen et al. 2000). On average, large European firms applied for patents on only 36% of product innovations and 25% of process innovations.”

Their answer to the obvious question “Why don’t patents reliably encourage R&D and growth?” is that “it is hard to sustain patent laws and institutions that make patents work like property.” That is an abstruse answer, to say the least, because patents do work like property because they are property. A better answer is that patents are not the be all and end all, as was recognized in the 2003 IPO survey, because they have limitations and weaknesses, i.e. strict patentability requirements, early publication and invent-around opportunity. And that is not all because the existence of three dozens of reasons for invalidity or unenforceability of patents cannot be overlooked. Likewise, other attrition factors for patents exist, such as:

• getting a patent and getting an enforceable patent are two different things,
• the average patent life is only five years due to non-payment of maintenance fees,
• enforcing a patent is a daunting and expensive task,
• only limited coverage is obtained in foreign countries, etc.

Still it is very important to go the patent route but it should be complemented by trade secret protection for collateral know-how to assure a fall-back position a la Pizza Hut decision, where patents were invalidated on summary judgment on on-sale bar grounds but the complementary trade secrets were validated after trial.

In prior blogs I railed aplenty about the many misconceptions about the patent/trade secret interface and the deep-seated belief that patents and trade secrets are incompatible.

A typical, illustrative and specific sample of such misperceptions comes from “The Trade Secret Handbook — Protecting Your Franchise System’s Competitive Advantage” by Michael Lockerby, American Bar Association, 2000.  In it the following relevant passages appear:

    How Do Trade Secrets Differ from Other Forms of Intellectual Property? (p.11)
    The issuance of a patent obviously destroys trade-secret protection.  However, the franchisor should maintain the confidentiality of any patentable trade secrets while the patent prosecution process is pending.  One reason that this is important is that a patent may never be issued.
    Can Trade-Secret Protection Coexist with Other Forms of Intellectual Property Protection? (p.12)

    There is considerable overlap between the scope of copyright protection and the scope of trade-secret protection.
    In contrast, patent protection is fundamentally inconsistent with trade-secret protection.  The quid pro quo for what is essentially a government-granted monopoly (sic) with respect to the patented invention is full disclosure so that it will be in the public domain upon expiration of the patent.  However, the patent prosecution process does allow for protection of trade secrets until such time as the patent actually issues.

These passages reveal misunderstandings and half-truths at best.  Let me parse these passages to see what is wrong with them.

“Issuance of a patent obviously destroys trade-secret protection.”

This is clearly too broad a statement and a half-truth.  Trade secret protection ceases to exist only for precisely that which is disclosed in the patent specification.  But patent applications are normally filed very early in the research stage, after a first reduction to practice, in order to obtain the earliest possible filing or priority date.  The specification of such an early application typically describes in relatively few pages only rudimentary lab or shop experiments done and samples or prototypes obtained.  Hence, patents rarely disclose the ultimate scaled-up commercial embodiments.  Tons of associated or collateral know-how not found in the specification are needed for commercial production and use of patented technology.