Patent thickets are dense webs of overlapping intellectual property rights. They are common in industries ranging from semiconductors to smartphones to pharmaceuticals.  When many firms own the underlying patents, thickets complicate licensing negotiations and increase the risk of holdup and litigation. These costs can be enormous: In 2021, for example, Intel paid $2.1 billion to VLSI Technology after a jury ruled that it had infringed two of VLSI Technology’s semiconductor patents. Conversely, firms that build their own patent thickets can use them to defend against litigation or even to crowd out competitors. This strategy allowed pharmaceutical companies such as AbbVie, Amgen, and Merck & Co. to add an estimated total of $500 billion in sales in recent years.[1] These costs and benefits have important implications for firms that acquire innovation through mergers and acquisitions. In a new paper, we find that patent thickets significantly affect the selection of M&A targets and thus create frictions in the acquisition market that may reduce social welfare by interfering with the efficient allocation of capital.

A series of maneuvers by Google in 2011 illustrate how patent thickets can precipitate acquisitions. That year, a thicket of about 250,000 patents covering the smartphone[2] led the likes of Apple, Google, HTC, Microsoft, Motorola, and Samsung to sue one another for patent infringement in what came to be known as the Smartphone Patent Wars. To protect itself, Google, which owned fewer than 1,000 patents at the time, attempted to build its own patent thicket by buying over 6,000 of Nortel Networks’ telecommunications patents.[3]^,[4] That  would have put Google in a stronger position to settle infringement claims by cross-licensing patents with accusing firms. After the attempted purchase failed, Google acquired Motorola Mobility and its 17,000+ patents instead.[5] Google specified that a main benefit of the acquisition was to protect its space in the patent thicket; its press release stated that “Motorola Mobility’s patent portfolio will help protect the Android ecosystem.”

We examine the relation between patent thickets and M&A by focusing on patent-owning acquirers and patent-owning targets in an M&A sample that includes nearly 6,000 deals over almost four decades. Our patent sample spans nearly 50 years, includes over 6 million patents granted by the United States Patent and Trademark Office, and allows us to construct a patent citation network, which we use to calculate a novel measure of patent thicket density. We conduct our analysis by matching each target firm to control firms and investigating how patent thicket density is associated with the probability of being acquired within each set of matched firms while controlling for characteristics such as patent count, patent citations, and technological overlap.

Our measures of patent thicket density use patent citations, which reference prior patents that a patent builds on. Since patent citations represent a limitation on the legal boundaries of a patent’s claim, they are well suited to measure patent thickets. The costs or benefits of a patent thicket depend on its ownership structure, so we construct proxies for two types of patent thickets.

We refer to the first type as an external thicket, which is comprised of patents owned by many firms. A firm hoping to commercialize technology dependent on patents in an external thicket faces costs (i.e., licensing negotiations, holdup risk, litigation risk) that increase with the density of that thicket. We use network analysis to develop a novel measure of external thicket density based on the density of citations around the firm’s patents in the patent citation network. For example, a firm is in a denser external thicket if it cites the patent portfolios of firms that often cite each other. Consistent with acquirers wanting to avoid the costs associated with external thickets, we find that the probability of being acquired decreases with the density of the potential target’s external thicket. External thickets therefore create frictions in the acquisition market that interfere with the efficient allocation of capital and thus may reduce social welfare.

We refer to the second type of patent thicket as an internal thicket, which is comprised of patents owned by the firm in question. Firms with a denser internal thicket are better able to defend themselves against infringement claims and deter competition. Our measure of the density of a firm’s internal thicket is the extent to which the firm cites its own patents. Consistent with acquirers wanting to procure the benefits associated with internal thickets, we find that the probability of being acquired increases with the density of the potential target’s internal thicket.

We also investigate how the probability of being acquired is associated with thicket co-occupation, or the extent to which the acquirer and another firm occupy the same external thicket. In such cases where a firm can directly impose thicket-related costs on the acquirer, we find that firm is more likely to be acquired.  This result is consistent with the acquirer attempting to reduce thicket-related costs.  Since the costs created by external thickets reduce social welfare, this result suggests that acquisitions may help mitigate the negative welfare effects of external thickets.

Interestingly, when an acquirer and another firm are in the same thicket, and that firm can have thicket-related costs imposed on it by the acquirer but not vice versa, we find that firm is less likely to be acquired.  This result is consistent with the acquirer preferring to extract rents from that firm rather than acquire it outright.

Finally, we find that internal thicket density is an increasingly important factor in selecting a target for acquisition. This implies that capital is increasingly allocated to firms that can protect their intellectual property and ability to innovate. On one hand, if this protection involves crowding out competitors by limiting the spread of knowledge, our results imply that capital is increasingly allocated to activities that harm society. On the other hand, if this protection involves defending against attempts to impose thicket-related costs, our results imply that capital is being increasingly allocated to activities that benefit society.

Overall, we find that the large economic effects of patent thickets create significant distortions in the acquisition market. In the case of external thickets, this means that thicket-related costs may further reduce social welfare by impeding the efficient flow of capital. In the case of internal thickets, however, the welfare effects remain uncertain.

ENDNOTES

[1] https://www.pcmanet.org/patent-thickets-are-anti-competitive-and-lead-to-higher-drug-costs/

[2] https://www.nytimes.com/2012/07/30/technology/apple-samsung-trial-highlights-patent-wars.html

[3] https://techcrunch.com/2011/07/01/apple-microsoft-rim-google-nortel-patents/

[4] https://www.washingtonpost.com/blogs/faster-forward/post/google-looks-to-cultivate-its-own-patent-thicket/2011/04/04/AFNvrXcC_blog.html

[5] https://www.google.com/press/motorola/

This post comes to us from professors Logan P. Emery at Erasmus University’s Rotterdam School of Management and Michael Woeppel at Indiana University’s Kelley School of Business. It is based on their recent article, “Patent Thickets and Mergers and Acquisitions,” available here.