4.1. Defensive strategy

The first situation mainly corresponds to the traditional objective of defending innovation in a narrow sense, namely to ensure one can ward off the risk posed by counterfeiting or the infringement of rights, both preventively and repressively. In this case, the entitled party wants to ensure that they can produce goods and services that include the protected knowledge and obtain income after they have been put on the market, without risking direct competition in this respect from a rival. There are two things at stake. Not only should we make our competitors respect us but also ensure against inadvertently infringing their rights ourselves, as a company that wishes to become involved in a given activity must first check that it is not unwillingly infringing a third party’s intellectual property rights by studying its freedom to operate – namely by referring to patent and trademark registries, etc.

Despite being classic, these considerations about defensive strategies remain very relevant today. As for companies in the French manufacturing sector, Duguet and Kabla [DUG 98] have thus shown that the main reasons behind the use of patents are the desire to prevent imitation (92% of respondents), to avoid trials initiated by competitors (62% of cases), to use patents in technological negotiations (62%), to receive license fees (28%), to reward researchers (18%), and finally to enter foreign markets. For the German industrial companies surveyed by Blind et al. [BLI 06], obtaining protection against imitation is also the respondents’ first concern.

4.2. Licensing strategy

The second type of organization involves entitled parties who monetize their intellectual property through licensing. Thus, income does not derive from the direct use of the knowledge protected but from granting licenses to third parties in return for the payment of license fees. Several large companies have been relying on this method for a long time by transferring technologies deemed nonstrategic to third parties, but in general they use it at most as an additional financial source. Broadly speaking, companies only license out less than 10% of their patents [WIP 11]. What is new is that some companies – a net minority of them – base on this strategy a significant part of their business model. Most of these companies work in IT, telecommunications or the multimedia sector. This is the case, for example, for the Technicolor group, specialized in technologies designed for the media and entertainment industry and the French company with the highest number of patent applications at the EPO. Its intellectual property portfolio, which includes among other things more than 30,000 patents and patent applications, produced a net income of 490 million euros in 2014 and constitutes the foundations for most of its operational profit (Ebitda): 65% in 2014 and 78% in the first semester of 2015!

IBM, an American group, claims to have derived from its intellectual property portfolio a profit of 1.5 billion dollars in 1999, but most of this amount corresponds to the value of intellectual property assets sold. The actual sums resulting from its patent licensing program were markedly lower and hovered around 200 million dollars in 1999 (in 1992 dollars). Furthermore, this is a gross sum that does not take into consideration the fact that IBM pays several hundred patent lawyers for its licensing activity [BES 08].

This licensing strategy generally involves the threat of legal action and, in quite a few cases, the implementation of such actions to win against third parties accused of infringing rights and compel them to sign license contracts to solve their situation. The semiconductor industry illustrates this connection between licensing programs and litigation (Box 4.1).

Always considering the licenser’s point of view, there are several cases in the drug or medical equipment industry where patents allow startups to reach licensing-out agreements, namely to grant to third parties under license inventions that they do not use themselves.

Drawing up such licensing contracts is made much easier by intellectual property rights, both as exclusive rights and to the extent that they indicate the precise outlines of the intellectual assets in question. Without these rights, companies that can negotiate license agreements would be wary of disclosing their secret technological knowledge when it is easy to copy. By playing this role, intellectual property rights – and patents in particular – facilitate specialization in the innovation process [WIP 16].

On the technological knowledge market, these rights contribute to a sort of cognitive division of labor that entails as a benefit improved effectiveness on a global level, as they represent the main institutional foundations that allow knowledge markets to work properly while also favoring the specialization of the agents considered in relation to their comparative advantages [CIM 08]. Taking into consideration that companies in the best position to conceive inventions are not necessarily the most skilled in marketing them, it is a priori advantageous to society to organize transfers of knowledge among them. In line with this division of labor between inventors and innovators, Arora et al. [ARO 16] emphasize a specific category of actors they call technology specialists, who correspond to independent inventors, academics, and R&D service providers. They show that, when companies base their innovations on inventions acquired from third parties, the most highly valued inventions are acquired from these providers of specialized technologies.

Formal intellectual property rights, and patents in particular, facilitate the transfer of technological knowledge among businesses, on what Gans et al. [GAN 08] call the “market for ideas”. These experts, however, add that this market must be defined by effective and timely deadlines to work properly, so that an innovation can be quickly launched on the market. Yet, this market presents several imperfections, which are more or less significant in relation to the technological fields and generally derive from different phenomena: information asymmetry, search costs, difficulties in transferring tacit information to potential licensee, etc. To evaluate them properly, we should recall, for example, that a startup involved in a patent commercialization strategy with a partner must consider the right time to do this, since although an early agreement may increase productivity and enable a more rapid launch on the market, a later agreement may confer more negotiation power and enable a more effective technological transfer. Three main factors of uncertainty may thus influence a similar license agreement: uncertainty about the ability to enforce the patent and, prior to this, the very fact of whether this patent is effectively granted or not, but also the time it takes for a patent to be granted knowing that this period tends to be long but can vary quite a lot. In the United States, this period is estimated to be on average 28 months, but it can vary quite widely (standard deviation of 20 months). An agreement like this differs in relation to whether the patent in question has been already issued or not. If the patent application is still pending, there is also incertitude about its scope; moreover, this uncertainty is still present even after the granting of the patent, as a court may always limit the scope of the patent or even invalidate it at a later stage. We finally need to consider the very strong uncertainty about the economic value of the patent, which depends on several factors especially on a technological and business level. In the biopharmaceutical industry, for example, very few patented inventions end up being marketed [GAN 08].

The value of patents is therefore very uncertain and mostly unrelated to that of the underlying technologies. Taking also into consideration how high entry costs and potential litigation fees are, these “derivative markets for science and technology” remain very concentrated and dominated by large-size companies [CIM 08]. These various market frictions or weaknesses thus limit the development of the licensing strategy.

4.3. Cooperative strategy

Unlike the licenser’s point of view, which relies on licensing-out, a cooperative strategy relies mostly on the opposite perspective, namely licensing-in, which allows licensed individuals to legally exploit the intellectual assets of third parties. Relevant surveys about the United Kingdom (for the period between 2009 and 2012) and the United States (for 2003) show that purchasing technologies from third parties, whether through R&D services or similar license agreements, represented for these two countries 40% and 44% respectively of the companies’ total R&D spending [ARO 13]. The underlying logic mostly involves collaborative innovation, which is often called “open innovation”. It means that a company, in order to innovate, does not rely only on its internal resources but also draws on the knowledge coming from several external partners. From this perspective, which aims more to include than to exclude [COH 11], intellectual property rights can reduce certain costs and expedite some procedures by obtaining the cognitive resources of third parties, notably in return for license fees, and using these rights as assets that can be negotiated in collaborative innovation projects or following a logic of reciprocity. Generally speaking, companies that innovate in partnership may benefit from using tools like patents, as patents help them both define the respective rights of the different parties involved and find common ground especially during negotiations carried out by partners. This suggests that intellectual property rights, far from being a simple locking tool based on the right to exclude, may also cement collaborative innovation.

4.4. Movement strategy

Finally, a fourth type of strategy banks, as in the first case, on the exclusive power of rights – to ward off competition – but this time more offensively or preemptively than defensively, with the goal of gaining a strong economic position and excluding competitors by obtaining high-margin competitive positions. The notion of a “blocking patent” is often used to summarize this logic, which is in most cases similar to a sprint. This type of practice aims, in relation to the situation, to hinder rivals, send false signals to them, or even prevent them from patenting. Cohen et al. [COH 00] also refer to “preemptive patents”, insofar as filing certain patents may involve creating obstacles to prevent competitors from patenting related inventions. The notion of a blocking patent may be developed in different ways, which have been analyzed by [GRA 99] and more recently described in detail in Corbel and Le Bas’s work [COR 12]:

1. – patent minefield: it somehow involves blanketing or flooding with patents that cover more or less significant inventions. Some minor patents are then used for their ability to do damage in order to slow down competitors;
2. – fencing: in order to consolidate a key invention that has already been patented, this other type of strategy involves filing a series of patents that aim to cover different technological solutions that can potentially produce a result similar to that of the major invention, so as to block any alternative path available to the competitors. In this sense, a typical example is provided by DuPont. This company, in the wake of the first patent filed for nylon in February 1937, patented in the 1940s more than 200 substitutes for nylon, namely molecular variants of polymers with similar properties. Thus, this use of patents as obstacles to the detriment of competitors can be observed more in those sectors that rely on discontinuous technologies, like the chemical industry;
3. – surrounding: a company may also surround an important patent held by a rival with a series of minor patents in order to block its commercial use;
4. – decoy patenting: a company aware of being observed by its competitors, thinking that filing patents related to a certain technological sector may be interpreted as a sign that it desires to follow a given path, may carry out decoy patenting to try to mislead its competitors by dropping a red herring.

In nearly 19% of cases, patent holders claim that their patents are not used internally (commercially or industrially) or licensed out to another user, but that they are used exclusively to block competitors, according to the PatVal survey that considers 9,000 patents granted by the EPO between 1993 and 1997 and whose inventors are located in six Western European countries: Germany, Spain, France, Italy, the Netherlands and the United Kingdom [GIU 05].

The relative significance of blocking patents has also been highlighted by the aforementioned Carnegie Mellon survey focused on industrial companies in the United States and the reasons why they resort to patents. As for product innovation, the desire to prevent rival companies from patenting related inventions – a motivation that corresponds to the notion of blocking patent – comes after the desire to prevent third parties from copying and before the protection against the risk of being taken to court for infringing the rights of others, the promotion of the company’s reputation, the reinforcement of its position in negotiations with other companies (especially in cross-licensing situations), the collection of license income and, finally, the assessment of the internal performance of engineers (Figure 4.2). The results obtained for process innovation are similar, even if the fourth and fifth place are inverted in comparison with the previous classification.

In practice, this movement strategy is particularly common in some sectors, especially in ICT where the increase in the number of patent filings is not only similar to an arms race but also often leads to actual “patent wars” (Box 4.3).

These four categories (Figure 4.1) clearly represent ideal types. In reality, the companies’ practices often correspond to intermediate and hybrid situations. Thus, the movement and cooperative strategies may converge towards the goal of technological conquest based on speed. Let us consider an example. Some large companies – especially in ICT, the multimedia sector, and the software industry – deliberately release some of their inventions into the public domain or license them out to third parties either free of charge or on very advantageous terms. This way, they hope that the technologies in question will be adopted by users more quickly and in mass, and that in the long run they will become ipso facto essential standards on the markets considered. In other words, a strategy of technological conquest does not necessarily involve the maximization of revenue from licenses in the short term. This shows that a relatively liberal policy in terms of licensing out may help the dissemination of the technology involved both among partners and competitors. This scenario illustrates more generally the notion of “coopetition”, which combines a logic of cooperation and competition.

From a microeconomic perspective, similarly, these four ideal strategies (Figure 4.1) are often complementary and may be combined. For example, the same company may simultaneously carry out licensing-out and licensing-in activities. Similarly, patents are often double-edged as they involve a strategy that may include both a defensive component and an offensive dimension, provided that the company in question is able to win a dispute. In the sectors involving complex technologies analyzed by [COH 00], companies often use patents to protect their own inventions, block their competitors and take them hostage by controlling the technologies they need, and as bargaining chips in negotiations.

How these companies combine these different uses depends especially on the technological fields and sectors considered. Companies eager to make the most of their innovation effort learn from the developments in management either by strengthening their own internal ability to think strategically or by resorting to specialized consulting firms.

Overall, intellectual property rights are thus an essential aid for companies and help them to situate themselves in their environment as well as possible in relation to their competitors or partners. Their contribution is multifaceted and cannot be reduced to the rights to exclude, since these rights may be used in a context of cooperation or for inclusive goals [COH 11]. Furthermore, in addition to the most traditional uses that correspond in the present analysis to the defensive and licensing strategy, the last few years have seen the emergence of new uses that involve patents used as a negotiation tool for cross-licensing agreements, as a signaling tool, or as a way of preventing competitors from patenting for preemptive purposes.