The strategies of patent introduction into patent pools

Abstract

This article explores patterns of patent introduction into seven patent pools over time, analyzing 1337 essential patents. Pools grow significantly after their launch, in particular through the addition of new patents by incumbent members. The generality, width and significance of patents introduced into pools decreases significantly over time. Incumbent members file and introduce patents that are narrower, more incremental and less cited than new entrants. Pool members, however, also introduce patents relevant to larger parts of the standard.

Keywords:

• patent pool
• patenting strategy
• patent quality
• standard-essential patents

JEL Classification:

• L24
• O32
• O34

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. This finding conflicts with other investigations revealing that pools include better patents than appropriate control patents (Delcamp 2012; Joshi and Nerkar 2011b).

2. ‘Moreover, the [MPEG2] pool members themselves had a strong incentive to exclude non-essential patents, since their share of the royalties was a direct function of the number of essential patents they held’ (quoted from the DoJ business review letter of the DVD3C patent pool, http://www.justice.gov/atr/public/busreview/2121.htm).

3. MPEG Licensing Association is one of the currently most important pool administrators (together with Via Licensing and Sisvel). There are currently eight patent pools administered by MPEG LA, including very important pools such as MPEG2.

4. The business review letter of the DVD6C pool states:

although the formula weights the patent count with other factors, each Licensor will benefit monetarily from the exclusion of other Licensors’ non-“essential” patents and accordingly has a strong incentive to encourage the expert to review other Licensors’ patents critically. http://www.justice.gov/atr/public/busreview/2485.pdf

5. For instance, the MPEG 2 pool stipulates: ‘The licensors are bound by the expert's opinion. However, they need not consult the expert if they agree unanimously in good faith that a submitted patent is an essential patent or that a portfolio patent is not essential.’

6. This claim is raised as patent misuse defence in many patent infringement cases, e.g. by disc replicator ODS in its litigation MPEGLA over the MPEG2 patent pool; Landgericht Düsseldorf Urteil vom 30. November 2006, Az. 4b O 346/05; V. b) cc).

7. For instance, the MPEG 2 pool uses the technical essentiality criteria (no alternative available), whereas the DVD 6C pool uses the economic feasibility criteria (no economically feasible alternative).

8. DVD6C is one of the two patent pools licensing out patents essential for DVD specifications, MPEG2, MPEG4 Systems, MPEG4 Visuals and AVC H/264 are patent pools including essential patents for coding standards issued by the Moving Pictures Expert Group, the IEEE 1394 patent pool covers wireless communication technology, and DVB-T is a patent pool for patents on Digital Video Broadcasting technology.

9. Japanese patent application with the same family ID.

10. A significantly larger number of US patents are included into a pool containing already at least one Japanese prior application which is not an equivalent. This is especially frequent in the case of DVD6C; see Appendix 1.

11. Patents included in several pools are not significantly more or less often founding patents or held by founding companies. These patents are however very significantly more general, broader, and more significant, and they cover significantly less sections of the respective standards.

12. The One Blue patent pool first implements a restriction on the number of patents per family (Peters 2011). The pool however also needs to provide incentives to file and include patents in many different countries, as well as continuation applications etc. Indeed, the pool can only attract licensees if it includes enforceable patents.

13. Baron and Delcamp (2012) confirm that the generality index is an especially meaningful indicator in the context of essential patents.

14. To exclude any bias and in line with most empirical research on patent quality, we exclude citations received by patents owned by the same firm; see also Hall, Jaffe, and Trajtenberg (2005).

15. A list of all the variables used in this paper with some descriptive statistics can be found in Appendix 2.

16. This pattern is consistent with the pool creation model of Dequiedt and Versaevel (2013), and is opposed to coalition building models such as Aoki and Nagaoka (2004).

17. This finding is consistent with Nagaoka, Shimbo, and Tsukada (2009), who find that holders of the earliest essential patents do not rely to a greater extent on continuations than other holders of essential patents.

18. Refer to Appendix 3 for an overview of the differences between patents introduced by pool incumbents and outsiders in the different pools of our sample.

19. We test the effect of pool membership on patent characteristics in two different ways. First, we test the effect on all the individual characteristics independently. This approach is justified if the variables are used to indicate the same characteristic, for instance the technological significance of the invention. But we also want to investigate the effect of pool membership on specific patent characteristics. We therefore also run all our regressions controlling for the number of claims.

20. For both models, we also estimate the regressions in one single set of seemingly unrelated estimations to account for potential interdependencies between the different effects. The results are similar to those presented in the body of this paper and available on demand. We also test the robustness of our findings controlling for the technological classes of the patents. The results are similar to those presented in the body of this paper and available on demand.

21. Columns 1 and 2 report the results on the number of claims. In columns 3–5, we present the same results on the generality index of the patent. Columns 6–8 present the results on the number of forward cites. Columns 9–11 introduce the findings on the number of standard sections for which the patent is essential to.

22. Statistically significant at 10% in column 5.

23. The results of the first stage are reported in Appendix 4.