Patent breadth as effective barrier to market entry

ABSTRACT

Firms protecting their invention by a patent have the legal right to exclude competitors from using the patented matter. However, despite patent protection, competitors could enter the market by ‘inventing around’. Provided that inventing around becomes more difficult the broader a patent is, the strength of protection against market entry increases in patent breadth. Building on a theoretical benchmark formalizing the relationship between varying patent breadth and the threat of market entry, our empirical analysis supports the prediction that inventors perceive broad patents as effective market entry barriers.

KEYWORDS:

• Patent breadth
• inventing around
• threat of market entry
• market entry barriers
• IPC codes

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. To isolate the effect of varying patent breadth on the threat of market entry, we disregard the possibility of licensing agreements as well as deliberate patent infringement. See Choi (1998) for an investigation of patent infringement and subsequent litigation in the context of a ‘timing of entry’ game.

2. See Zaby (2010) for a detailed derivation of equilibria with and without patent protection.

3. To isolate the strategic effects of patent breadth, we implicitly assume that the length of a patent exceeds the time that the competitor needs to invent around the patent. Therefore patent breadth is the only dimension of patent protection relevant for our analysis.

4. In the words of Green and Scotchmer (1995) and Denicolo and Zanchettin (2002), patent breadth is thus ‘leading breadth’.

5. The theoretical implementation of an exogenously given patent breadth disregards that inventors can influence the breadth of their patents by writing patent claims accordingly (see Yiannaka and Fulton 2006, for a theoretical approach). In the course of patent examination the scope of patent protection is further specified and possibly narrowed down. One could even argue that the ‘true’ scope of patent protection is only revealed after a patent has been validated in the course of litigation, as then (at least some of) its ‘borders’ are clearly specified (see Waterson 1990). For our investigation of the effectiveness of patents as barriers to market entry it is insubstantial who determines patent breadth, as we are interested in the ex-post effect of patent breadth on inventors' competitive situation.

6. We cannot exclude that a reverse effect might exist if patent breadth is chosen endogenously by patentees. Then the competitive situation could affect subsequent patent breadth decisions. However, to offset this reverse causality we use patent applications up to the year 2003 and include a time lag between the latest application dates and the time of the assessment of the competitive situation in the year 2005.

7. The German CIS contribution originates from the Mannheim Innovation Panel (MIP) which is conducted yearly by the Centre for European Economic Research (ZEW) Mannheim, the Fraunhofer Institute for Systems and Innovation Research (ISI), and the Institute for Applied Social Sciences (infas).

8. The merge was conducted by Thorsten Doherr, ZEW, Mannheim.

9. If firms stated to hold a patent but had no equivalent entry in the EPO data, we dropped these observations.

10. Section 33(1) of the German Patent Law states,

As from the publication of a reference pursuant to section 32(5) the applicant can claim compensation which is reasonable in the circumstances from any person who used the subject-matter of the application although he knew or should have known that the invention he used formed the subject-matter of the application […].

Where section 32(5) states that publication of a patent specification is mandatory 18 months after filing the patent.

11. Note that large firms possibly have more than one product market. In such a case, our procedure relates the breadth of a firm's entire patent portfolio to its main product market only. By this we implicitly propose that the perceived threat of entry on the main product market is influenced by patents which are possibly not technically related to this market. Several arguments support this procedure. First, we argue that a firm will carefully protect its most valuable technology, i.e. a substantial part of the firm's patent portfolio should consist of patents relevant for the firm's main product. Second, as the mean number of employees in our sample is 97 we argue that firms in the sample do not have a highly diversified product portfolio (see Gollop and Monahan 1991). Third, as we do not investigate actual market entry in reaction to patent protection as Cockburn and MacGarvie (2011) do, including all patent applications allows for the possibility that patents which are not technically related to a firm's main product market nevertheless affect the market entry threat firms perceive. However, despite these arguments, we test whether the use of our measure of ‘general’ patent breadth confounds our results and investigate the subsample of small- and medium-sized firms, which are assumed to have one product market only, as a robustness check.

12. To examine whether this distinction of variations in the IPC codes is an appropriate measure of patent breadth, we use the regression-based robustness check implemented by Lerner (1994). We use patent data from 1995 and corresponding citations up to 2005 to avoid the problem of recentness. Our Poisson regressions reveal two significant effects: a strongly positive effect of our broad patent indicator and a negative effect of our narrow patent indicator, i.e. patents we define as broad are associated with more citations, whereas patents we define as narrow are associated with less citations. For further robustness checks we move the frontiers of each category.

Additional information

Funding

Both authors acknowledge funding from the research project ‘Strengthening Efficiency and Competitiveness in the European Knowledge Economies’ (SEEK 131213 to DH, AKZ); State of Baden-Wuerttemberg, Germany.