Changes

Jump to navigation Jump to search
no edit summary
 
*This page is referenced in the [[Patent Thicket Literature Review]]
*This page is listed on the [[PTLR Core Papers]] page
On the basis of a novel data set of 612 European patents and related inventions from five different industries, it is shown that multiple patents per invention are filed in both discrete and complex technologies. Multivariate analysis of the data suggests that in selected discrete technologies, patent ‘fences’ may serve to exclude competitors whereas in complex technologies, ‘thickets’ represent exchange forums for complementary technology. The results expand on traditional views of profitable patent exploitation across industries and elaborate on the most recent findings by Cohen et al. (Cohen, W.M., Nelson, R.R. and Walsh, J.P. (2000) Protecting Their Intellectual Assets: Appropriability Conditions and Why U.S. Manufacturing Firms Patent (or not). Cambridge, MA: NBER.) The analysis suggests that different legislative issues arise from multiple patenting per innovation in complex and discrete technologies depending on the degree of technological complementarity. The results have unexpected policy implications in that they illustrate how patentees could eliminate competition in the form of substitute technologies through fencing. They have wide managerial implications regarding the valuation of patent portfolios and the design of corporate IP strategies.
 
==Review==
 
===Measures of thicket===
 
The article does not measure impact from patent thickets using measures external to the firm. Instead, the article considers the impact of the firm's own patent bulk (number of patents covering the firm's invention) on patent value, and finds limited results supporting links between patent value and use for "thicket" versus "fence" industries.
 
===Sample===
 
Size, data source, industries, geography:
*612 inventions protected by European patents from five industries.
*Data is drawn from a questionaire asking about the value of patents covering various inventions sent to patent-holders and conducted by the EPO in 1994.
 
===Results===
 
*No significant results about patent value and use for "thicket" industries:
:''"Neither in the field of electronics, machinery, nor manufacturing of transport equipment there are significant interaction effects to be found between the bulk size and the dummy variables classifying the patents as bargaining patents. Consequently, the notion of a significant ‘thicket’ effect on individual inventions’ values is not supported."''
 
*Chemical patents are used to exclude competitors as expected in "fence" industry:
**Chemical patents for exploitation by royalties, interacted with bulk is positive.
**Chemical patents used to access foreign market have significantly higher value, but the interaction with bulk is negative.
:''"Another interesting finding relates to the interaction effect for bulk size and those chemical inventions that are primarily intended to exclude competitors. Here, the coefficient is positive, supporting hypothesisH1.3 that a ‘fencing’ effect might increase the value of the whole patentprotected invention by protecting a set of related substitute technologies."''
 
===Social Welfare Consequences===
 
*Impact on value can depend on industry and use of patent. In particular patent value is lower for:
**Drug patents used for access to foreign markets;
**Drug patents exploited for royalties/licencing interacted with bulk;
**Electronic patents used to prevent copying;
**Transport patents for exploited for royalties/licensing;
:''"...there is some private value in bulk patenting for both patentees who exchange technology in complex industries (‘thicket patenting’) and patentees seeking exclusiveness in discrete technologies (‘fencing’). The private value of the bulk seems to depend crucially on the use that is made of the patents."''
 
===Dependent Variable and Model===
*The dependent variable is the value patent(s) of patents covering an invention:
**Value obtained from a survey question that asked the respondent to estimate the added annual value from the patent(s) on a categorical scale from 1 to 5 representing logarithmically increasing intervals of value;
**The monetary values in Euros associated with ordinal value were 1) 0-10,000, 2) 10,001-100,000, 3) 100,001-1 million, 4) 1,000,001 to 10 million, 5) over 10 million, with a median value of 2;
**Unit of observation is the invention covered by the patent(s) as the questionaire asked about the single patent or set of related-patents needed for a given invention.
*A reduced form analytical model of patent value is estimated by OLS or ordered Probit accounts by adding controls in stepwise fashion:
**Age, or the duration of the patent from time of granting until it expires (averaging 9.5 years old);
**Firm size of the patent-holding company measured by number of employees (average of 1,800 workers);
**Primary use of patent: prevent copying invention, exploitation of royalty licensing, or access to foriegn markets;
**Industrial sector of the firm or business unit, likely reflecting the technology of the related invention (chemicals, drug, machinery, electronics, transport equipment);
**Patent "bulk", the number of patents belonging to same invention;
**Interactions between patent "bulk" and industry and use.
Anonymous user

Navigation menu