When Licensing is a Winning Choice
Licensing is most likely in industries where firms can produce technologies that are useful for several product markets, and when markets do not compete highly with each other. In this case, firms can license their technologies in other markets without fearing increased competition in their home market.
Alfonso Gambardella and Marco Giarratana (Department of Management and Technology and CRIOS) conclude this after investigating the conditions under which general purpose technologies (technologies that are valuable in multiple product markets) can facilitate licensing without increasing competition in General Technological Capabilities, Product Market Fragmentation, and Markets for Technology (Research Policy, Vol. 43, Issue 2, March 2013, Pages 315-325, doi: 10.1016/j.respol.2012.08.002), where these arguments are formalized in a robust theoretical model and tested empirically. Their findings imply that property rights protection and the assets required for production are not the only strategic factors that should matter to a firm's technology strategy.
The authors use a formal model to prove their hypotheses, which takes into account the effects of two factors on demand and supply. The first is whether the licensed technology is specific to one product market (dedicated) or is useful in multiple product markets (general purpose). On the one hand, it is easier to implement a dedicated technology, so this will increase demand, but on the other hand, a general purpose technology can be used for multiple product markets, which attracts more potential buyers. Furthermore, the product markets can be homogeneous (all competing strongly with each other) or fragmented (little competition between product markets). The main insight is in the interaction between these dimensions: The benefit of general purpose technologies regarding licensing is only present in fragmented markets because there is too much competition between markets in the homogeneous case.
After mathematically formalizing these arguments, the authors derive two formal hypotheses that they test empirically in the security software industry by looking at how the patent structure (broad or narrow) of the firm relates to the amount of licenses granted.Indeed, they find that in fragmented markets firms that are able to produce general technologies (which have a broad patent structure) have a higher chance of issuing a license than those which are able to produce dedicated technology (have a narrow patent structure). In homogeneous markets, this difference is negligible since there is not so much supply to begin with.
As indicated earlier, these findings imply that firms that want to produce general purpose technologies are in a better position to profit from licensing. Furthermore, this implies that firms should align their product market strategy with their technology strategy. For instance, when positioning themselves in fragmented markets, firms can develop capabilities to adapt general technologies to their product market, and then buy general technologies from firms in other markets. Alternatively, firms can specialize in one product market, but generate general purpose technologies which it can also sell in other product markets.