Hall (2012) - A Study Of Patent Thickets

From edegan.com
Jump to navigation Jump to search
Article
Has bibtex key
Has article title A Study Of Patent Thickets
Has author Hall
Has year 2012
In journal
In volume
In number
Has pages
Has publisher
© edegan.com, 2016

Reference

  • Hall, B.H., Helmers, C., von Graevenitz, G. and Rosazza-Bondibene, C. (2012), "A Study of Patent Thickets", Draft Report to the UK IPO, pp.1--66
@article{hall2012study,
  title={A Study of Patent Thickets},
  author={Hall, B.H. and Helmers, C. and von Graevenitz, G. and Rosazza-Bondibene, C.},
  journal={Draft Report to the UK IPO},
  pages={1--66},
  year={2012},
  abstract={},
  discipline={Econ, Policy Report},
  research_type={Empirical},
  industry={},
  thicket_stance={},
  thicket_stance_extract={},
  thicket_def={},
  thicket_def_extract={},  
  tags={},
  filename={Hall (2012) - A Study Of Patent Thickets.pdf}
}

File(s)

Abstract

This report analyses whether entry of UK enterprises into patenting in a technology area is affected by patent thickets in a technology area. The aim is to contribute to our understanding of the role of patent thickets as a barrier to entry into patenting for UK enterprises, in particular small and medium sized enterprises (SMEs). To do this we review the literature on patent thickets, discuss factors contributing to thicket formation and growth, and evaluate to what extent patent thickets might be considered a barrier to entry in some technology areas. We also summarize the limited existing empirical evidence regarding effects of patent thickets on R&D investments and competition.

We find overwhelming evidence in the literature that patent thickets arise in specific technology areas. This conclusion is based on a comprehensive review of the empirical literature on patent thickets of the last 15 years. This literature consists of surveys of firm representatives as well as of econometric analyses of firm level data. The literature on thickets contains more than 100 peer reviewed papers and a number of extensive studies undertaken by competition regulators.

Our main contribution in this study consists of an empirical analysis of the effects of patent thickets at the European Patent Office on entry into patenting by UK firms. Using a new measure of patent thickets developed by Graevenitz et al. (2012), the report provides a descriptive analysis of the growth of patent thickets in the European patent system and an analysis of the growth of the exposure to these thickets of UK entrants into patenting. Econometric analysis of the probability of entry into patenting by technology area shows that the density of a patent thicket is associated with reduced entry into patentin in the technology area in the data set used for this study. We discuss the limitations of the data used and suggest how further work might test the reliability of our findings.

Review

Measures of thicket

Following Graevenitz, patent thickets are measured by counts of "triples" within (33) technology sectors:

  • "Triples" are a group of three firms, each of which has a recent patent application that cites to "critical" prior art of the other two firms;
  • A "critical" prior art (also called an "X and Y citation") is a reference in a patent application with the European Patent Office (EPO) that indicates that the patent being applied for references a prior art that limits one of more of the claims in that patents;
  • Counts of "triples" in each industrial sector normalized to total patenting in each sector, (triples per 1000 patents).

Sample

  • Nearly 1 million observations on 29,435 Small and medium sized UK firms from 2002-2009 drawn from the Bureau van Dijk's Financial Analysis Made Easy (FAME) database:
    • Excludes industrial sectors with little patenting;
    • Selects a 1% sample of non-patenting firms for comparison.
  • Patent citations used for the triples measure of patent thickets are drawn from EPO PATSTAT database.

Results

Descriptive statistics:

"the count of triples in Electrical Engineering far outstrips the counts of triples in any of the other main technology areas. This is commensurate with...the increase in patent applications at USPTO after 1984...primarily due to firms operating in Information and Communications Technologies."
  • The most triples are found in the sub-segments of the Electrical Engineering group known as "Telecommunications, Audiovisual Technology, and Computer Technology";
  • Figure 6, p.40 gives a telling graphic of the network of critical references in the Technology Area Telecommuniations in 2005.

Entry into patenting in a technology area:

"A doubling of patenting is associated with a 70 per cent higher probability of entry (standard error 2.0%). However, when we include the triples density in the class, we find that it depresses entry. Doubling the intensity of triples in a class is associated with a highly significant 12 per cent lower hazard of entry into that class (standard error 0.7%)."
  • The coefficient on the log of triples density in a class is -.12 and significant at 5% level.
  • The coefficient on the log of patents in class is -.05 and significant at 5% level.
"...we interact the log of assets with the log of patents and the log of triples density to see whether these effects vary by [firm] size. They do, and in the expected way. The impact on larger firms from aggregate patenting weakens and the impact on triples strengthens slightly. That is the impact of aggregat patenting (positive) and triples density (negative) on [small and medium sized enterprises] is stronger than it is for larger firms. ...Firms in the lower range of the size distribution (assets less than 10,000 pounds) are much more likely to enter a sector with high aggregate patenting if they enter at all, but their hazard of entry falls 15 per cent if the triples density doubles in that sector. On the other hand, for the few firms in the upper range of the size distribution (assets greater than 100 billion pounds), the hazard of entry falls only 7 per cent if the triples density doubles."
  • The coefficient on an interaction betweeen the log of triples density in a class and log of firm assets is .004 and significant at 5% level.
  • The coefficient on an interaction betweeen the log of patents in a class and log of firm assets is -.032 and significant at 5% level.

Social Welfare Consequences

"As we find thickets to affect entry negatively, there is a strong indication that thickets represent some kind of barrier to entry in those technology areas in which they are present. However, we must emphasize that the simple finding of a barrier to entry created by patent thickets is not proof positive that reducing that barrier and increasing entry would lead to welfare improvements in the innovation/competition space. Rather it is the existence of evidence that the presence of patent thickets reduces entry combined with the large literature we have reviewed that shows that currently patent systems do not work as well as they should."

Dependent Variable and Model

  • The Dependent variables is the time until a firm's entry into patenting in a given technology area.
  • Log-logistic regressions are estimated with stratification by industrial sector, and control for:
    • Patent thicket as measured by the triples density in technology class (triple count per 1,000 patents in class);
    • Total EPO patents in each industrial sector in a year;
    • log of reported firm assets (in British pounds);
    • Indicators of 2-digit SIC industry code;
    • Interactions bettween 1) triples density and log of assets, 2) number of patents and log of assets.