For the half-year to 30 June 2014, the IPKat's regular team is supplemented by contributions from guest bloggers Alberto Bellan, Darren Meale and Nadia Zegze.

Two of our regular Kats are currently on blogging sabbaticals. They are David Brophy and Catherine Lee.

Monday, 23 January 2012

Katonomics 9: IP valuation

In this, the third post in the second series of Katonomics, the IPKat weblog's resident and much-loved Katonomite, Dr Nicola Searle continues her introduction to economics for intellectual property enthusiasts, scholars, practitioners and policy-makers. This week Nicola tackles a subject that some IP experts view as a precise science and others see as a hybrid discipline that lies somewhere between guesswork and witchcraft. To find out what Nicola says, just read on:
"Valuation of IP 
I have two favourite answers in economics: “it depends” and “diminishing marginal returns”.  Both are useful phrases and can be applied to many situations. However, for this post, only one is relevant. So, how do you value IP? Answer: It depends. 
The most common valuation techniques for IP are cost, income and market methods. Cost methods values IP as the costs incurred in the creation of the IP (typically R&D costs). Income methods value the IP at what it earns (or could earn), often in the context of discounted cash flow analysis. Market methods value the IP at what the market is willing to bear. This WIPO guide explains the methods well. 
These methods are not unique to IP or to intangible assets. For example, the commercial real estate market uses similar methods. Under the cost method, a building is worth its replacement cost. The income approach entails a cash flow analysis of rental income and costs. A market analysis would look at sales of similar buildings. This analogy also highlights how the three different methods interact; for example, cheaper construction costs and lower rents will depress market sales. Ultimately, an asset is only worth what someone is willing to pay for it. 
But, as I said, it depends. For some IP, the cost model may be inappropriate. What if the IP in question was developed through a flash of genius? Or if an upcoming PR disaster is going to damage a brand severely? The IP may be worth much more or less than its R&D cost. 
Market valuation is tricky as there may not be comparable sales or available data on those sales.     
The income methods are particularly appealing as they avoid some of the above problems associated with cost and market methods, and are unpinned by the well-established cash flow valuation analysis. However, they are not infallible, as they still require projections and assumptions about markets. 
One innovative solution is to treat IP rights as probabilistic rights and apply real options analysis. Lemley and Shapiro put forth the idea that, instead of ironclad rights, IP should be treated as probabilistic rights. IP protection can be lost through a variety of means including invalidation, market obsolescence, damaged goodwill and loss of secrecy.  A patent is not a right to exclude, but a right “to try to exclude.” Thus, an IP right represents a probability of having a right, rather than an absolute right. 
Real options analysis captures the probabilistic nature of IP rights in its valuation. As noted by German economist Baecker, real options valuations incorporate the bundle of options associated with a patent including the option to commercialise and the option to litigate. For example, British economist Van Reenan and his American counterpart Bloom use real options analysis of patents to investigate how uncertainty affects productivity. They find that policy attempts to reduce uncertainty will increase productivity. 
Real options have long been a strategic tool for the modelling of investment decisions. The theory behind real options comes from the finance world and incorporates the principles of cash flow analysis. However, as Greek economist Andrikopoulos describes, the statistical assumptions and restrictions of the model may not be in tune with the business reality. Given the recent performance of financial markets, readers will be forgiven for their scepticism. Nonetheless, real options analysis could address some of the shortcomings of other methods. 
As a policy measure, the valuation of IP can have an impact on the use of IP. Consider the case of litigation of patent infringement. Scotchmer argues that there is circularity between valuations used in disputes and those in negotiations. A potential licensee chooses between licensing and infringing. If the valuation used in court is lower than a licence, the potential licensee may choose to infringe. At the same time, a licensor’s proposed royalty payments are limited by the values used in court. Thus, valuations in disputes affect bargaining positions in seemingly unrelated transactions. 

Here we touch on the world of economics of crime, a discipline that has exciting tag lines such as “the socially optimal rate of murder”.  Economist Becker put forth the idea that lawbreakers weigh the costs and benefits of breaking or obeying the law and find in favour of breaking the law. The same logic applies to the would-be patent infringer, industrial spy or counterfeiter. Assuming equal benefits, if the costs of not infringing outweigh the costs of infringing (adjusted for the likelihood of getting caught), then infringing is a logical decision. The value of the IP in question plays into this cost-benefit analysis. 
At this point in the post, I’m heading into diminishing marginal rates of return. What is your preferred valuation method? Does the would-be patent infringer take potential damages payment into account?"


Anonymous said...

Is there any study on the time dependence of the value of IP. My working model (no great insignt claimed) is that in the early stages the value is a bimodal probablity distribution, the modes being an awful lot and nothing at all. Later on the value crystalises, again to something or nothing at all, but more clearly one or the other. Some benchmarks by industry on when you should know that it is worth something or nto might be interesting to know - for example when reviewing patent portfolios that will be costing national phase filing, examination and renewal fees. They might also inform policy on how long examination should take. Fast examination, as is the stated aim of patent offices, will at the extreme kill patent protection being obtained through cost but may be unncessary if in short while the inventor will know whether the thing has legs and if not no examination is needed.

When it comes to the valuation method won't the valuation method to be used depend on where one is in that progression? If you have a product licensed to 20 companies then sure you can do an NPV calculation. At the other end, how many inventor's invention disclosure forms have you seen that confidently says the thing will be worth $10billion per annum?

A final question: I read a while back about a project to develop valuations for lending on patents. Obviously if one can lend to enough people the uncertainty in the value of early stage ip is not so much of a problem since you only have to be right on average. Has anyone heard if there has been progress on that project?

Anonymous said...

Since this is a bean count, why not crowdsource?


Nicola said...

A reader pointed out this paper using the Black-Scholes equation to value patents by Heald and Denton.

I enjoy the abstract's opening line, 'Current patent valuation methods have been described charitably as "inappropriate," "short-sighted," "inherently unreliable," and a "guestimate."'

Subscribe to the IPKat's posts by email here

Just pop your email address into the box and click 'Subscribe':