Appendix K in Strategic Patenting contains a thorough checklist covering patents, trademark, copyright, and trade secret.  Very briefly, patent due diligence requires consideration of: all company owned patents and pending patent applications (U.S. and foreign); all patents and applications in-licensed or out-licensed, all assignments; all prior art references known to the company, its named inventors, and its IP counsel that may render any company patent claim invalid or unenforceable; all patent opinions; employment agreements for all partners, employees, consultants, and other company-related indiv-iduals who might be involved as inventors; list of all IP counsel in the previous 5 years, including statistics on their cost-effectiveness and filing strategies; patents and applica-tions of competitors; procedures for invention disclosure and ensuring secrecy of any trade secrets.

Patents and other forms of intellectual property are intangible assets, merely a legal right conferred upon the patent holder.  Yes, one can physically hold a registration certificate or other indicia of the asset, but one cannot physically feel, touch, smell, or hold, the underlying legal rights.  Add to a patent's intangi-bility the fact that the legal right is by definition unique, and one has the very essence of a difficult-to-value asset.  Despite these difficulties people do value patents all the time.  The common models include:

• willing buyer/seller model
• marginal profit  model
• cost/replacement model
• simple metrics
• management team

Qualitative valuation is probalby most accurately based upon the marginal profit  model.  In that model one makes assumptions about the market size, market penetration, cost of entry, cost of production, lifespan, and profitability of the patented product, and then runs a spreadsheet to estimate profit.  The value of patent is then taken to be the discounted present value of the estimated marginal profit, i.e., the profit over and above profit that which one would expect to be generated without the patent.  Unfortunately, the validity of the marginal-profit model depends entirely upon the validity of the underlying assumptions, and those assumptions can vary wildly among different analysts.  At the end of the day, it all boils down to experience and judgment.

The qualitative value of a patent is based to some extent upon extreinsic factors (expected market size, profitability, and so forth), but also upon scope and enforceability of the claims.  The real issue with respect to claim scope is how difficult it is to circumvent the claims.  In making that determination one must evaluate every element of a claim, and determine whether there is a commercially feasible way of avoiding that element.  If there is at least one element in every claim of a patent that can be avoided, then the claims of that patent have little value.

The law regarding claim scope is set forth in considerable detail in the sample opinion letter of Appendix J of Strategic Patenting.

Absolutely.  If the patent holder is merely licensing the patent to a manufacturer, the estimated profit will be considerably lower (by three quarters or more) than if the patent holder were to manufacture and sell the product.  But that difference is misleading because it fails to account for the risks and costs involved in production, distribution, and patent enforcement.  Most electronic products, for example, cost much more for a newcomer to manufacture than for an established business.  From the distribution standpoint, an established player will likely sell many times that of a smaller player, just because of name recognition and store placement.  From an infringement standpoint, one needs to consider that a competitor will think very hard about trying to outrun the patent of a multi-billion dollar company.

Inventors often want to know what is a reasonable range of royalties for their invention.  Thankfully, the process is not entirely guesswork.    For most mechanical patents the starting point for royalties is about 3%, with extremely valuable patents fetching values of 7-10%.  On the other end, it is not uncommon for a patent royalty to be down in the 1 - 2% range.  Pharmaceuticals can fetch up to 20% and more, but only because the margins can run 80 - 90%, and the barrier to entry of competing products is extremely high.  The standard rule of thumb for estimating royalty rates is that the licensor should receive 20 - 25% of the marginal value of the patent.  Most likely, a license results in increased profit on a per unit basis, as well as increased the total sales.  Thus, the royalty should be calculated as a function of both increased profit per sale and increased sales.

It will come as no surprise that patent owners tend to value their patents more highly than assignees or licensees.  The question is how such disparities can be resolved.  The most important suggestion is to use the marginal profit model of patent valuation.  At the very least, that model provides a mathematical foundation upon which to start negotiations.  Yes, the assumptions will lead the model, but at least one can argue the assumptions more intelligently that one can argue the final number.  I strongly suggest that the parties allocate risks and benefits using (a) minimum royalties and (b) buy-outs.  A patent assignee or licensee is usually willing to forgo a large upfront payment if he is guaranteed a minimum royalty.  Similarly, a patent assignee or licensee is usually willing to pay a higher royalty rate if he can buy out of the contract at some future date.

Although there are exceptions, valuation consulting firms typically provide a nicely prepared, impressively thick report that will tell you exactly what you hired them to say.  But you won't know how they accomplished their analysis, and you won't be able to assess the validity of the analysis.

Perhaps the most frustrating aspect is that such firms tend to prepare their analysis at a level of abstraction that virtually ignores the specifics of the patent claims.  Valuing a patent while ignoring the scope and validity of the specific patent claims makes about as much sense as appraising a diamond while ignoring the size, cut, color, clarity, and defects of that specific diamond.  It can be done, but it can't be done well.

For several years patent donation has been somewhat of a scam.  The IRS finally took notice, and issued a new rule in which a donor’s initial charitable deduction is limited to the lesser of (a) the donated property’s tax basis and (b) its fair market value.  Additional deductions may be allowed in later years based on specified percentages of income received by the charity from owning the donated property.  These additional amounts are calculated using sliding-scale percentages that decline over the years.  Additional deductions after the year of contribution are not allowed for donations of patents and intellectual property to a private foundation - unless it’s a private operating foundation or a private foundation described in 26 U.S.C. § 170(b)(1)(E).  These changes apply retroactively to donations made after June 3, 2004.

It is possible for a patent holder to obtain a "reasonable royalty" during a period of provisional protection.

It is also possible to license or sell the subject matter of a pending application, even though the scope of the eventually issued claims is unknown at the time.  In fact, this happens all the time.  In effect the licensee assumes the risk that the patent will never issue, or that the claims will have little commercial use, in exchange for a confidence that it can push forward with researching or commercializing the invention without having to later try to negotiate a royalty.  To the extent that the license is exclusive, there is also a benefit in possibly precluding a competitor from acquiring a license down the road, after the patent issues.