German Patent Attorney Malte Köllner has written an interesting piece in February/March's Intellectual Asset Management Magazine entitled "The Journey is the Reward" (page 59). Malte - who also co-founded the Triangle Venture Capital Group - is a member of the Geramn Standard's Institute's committee working on a standard for patent valuation (see previous posting here).
Malte points out that much of the thinking and work done on patent valuation to date has been by those with accounting backgrounds who have merely considered the economic aspects in valuing the patents - the technical and legal aspects have often been ignored. He's quite correct in pointing out that there is no such a thing as a "perfect patent" and that there is always a risk that the patent will be revoked (for reasons like prior art not discovered by the patent office, lack of enablement, lack of entitlement, etc).
One problem that he fails to mention in his article is the source of the data on which much patent valuation is based. The licence rates usually come from the information contained in company accounts filed with the US SEC or from publicised damages awards. Particularly in court cases, the validity of the patents has often been challenged and their strength upheld (even if in an amended manner). In a sense there is little or no "risk" element in the value of the royalty rate. Once the validity of the patent has been challenged once, then the chances that it will be later revoked are much smaller.
On the other hand the validity of most patents has never been challenged. There is a much higher risk that, for example, the patent office has found no relevant prior art which means that the scope of patent protection is much narrower than that granted by the patent office. The royalty rates derived from damage awards or licensing negotiations for such patents need to be discounted - the question is to what extent.