Access Official Technical PDF

Share Summary:LinkedInEmail

Patent assertion entities (PAEs)—firms that earn revenue by licensing or asserting patents rather than primarily selling physical products—occupy a contentious place in debates about technology and innovation. For industrial automation companies, whose products often combine hardware, software and communications technologies, patent disputes can affect both intellectual property strategy and the commercial environment for innovation. But do PAEs generally undermine innovation, or can they help technologies reach the market?

A 2018 empirical analysis by Noel Maurer and Stephen Haber examined publicly traded firms identified as PAEs by patent-industry expert RPX Corporation. The authors assessed company filings and share-price data to test claims about research and development (R&D), profitability, investment risk and the scale of any transfer from operating companies. Their results challenge a simple account of PAEs as uniformly low-investment, low-risk businesses built on nuisance litigation.

Patent assertion entities and the competing explanations

The “patent troll” hypothesis describes firms that acquire low-value patents and use threats of costly litigation to secure settlements that may have little connection to the patents’ economic value. In this account, payments and legal expenses divert resources from productive investment, including R&D.

A different explanation emphasizes patent licensing as a form of intermediation. An inventor or technology developer may not have the capital, expertise or commercial position to bring a product to market. Licensing or selling patent rights can provide a way to earn returns from an invention, while allowing an operating company to develop and sell products using the technology.

These explanations are difficult to evaluate if “PAE” is applied inconsistently. The study used RPX’s identification of 26 publicly traded PAEs rather than creating its own classification. The authors then examined data drawn mainly from company filings with the U.S. Securities and Exchange Commission, covering 2000 through 2016. This approach creates a defined group for analysis, but it does not mean that every firm or patent-licensing model fits the same pattern.

What the study found about R&D and patent investment

The patent troll hypothesis would lead one to expect little investment in developing or acquiring valuable technology. Yet, for 2011–2016, the 26 firms spent a weighted average of 19% of revenue on internal R&D and a further 9% on patent acquisitions. The comparison group of 153 large U.S. high-technology companies spent an average of 10% of revenue on R&D over the same period.

Those figures do not prove that every PAE develops useful technology, or that R&D spending translates into commercially successful products. They do, however, complicate the claim that the publicly traded firms studied were generally characterized by negligible research investment. The authors also note that patent-acquisition spending may be understated because it did not capture patents obtained through mergers and acquisitions.

The result has a practical implication for technology-intensive businesses: a company’s role in the market cannot always be inferred from whether it manufactures products. Patent ownership, research activity and licensing can coexist in different combinations. For industrial automation organizations reviewing a potential partner, competitor or rights holder, examining the specific firm and its patent portfolio may be more informative than relying on a broad label.

Profitability, risk and market scale

The study also tested whether the firms behaved like consistently profitable, low-risk enterprises. The evidence did not support that description for the group as a whole. Between 2000 and 2016, the publicly traded PAEs collectively recorded $3.1 billion in losses, and only six of the 26 firms generated positive returns for shareholders over that period. The authors report that the firms with strong share performance also exposed investors to substantially greater risk than an investment in the S&P 500.

These results challenge the idea that patent assertion was an easy route to dependable returns across the sample. They are not a finding that abusive litigation never occurs: the analysis concerns a particular set of publicly traded firms, and accounting performance alone does not establish whether an individual patent claim is valid or whether a lawsuit is justified.

To estimate potential sector-wide impact, the authors made a separate calculation for 2011–2016. They added the firms’ revenue to estimated litigation-defense costs imposed on operating companies. On that basis, the estimated transfer averaged 0.28% of revenues in the U.S. high-technology market. The paper distinguishes this transfer from deadweight loss—the economic activity that does not occur because of the transfer—and estimates the latter at a much smaller level under its assumptions.

Important limits on interpretation

The findings should be read within the study’s boundaries. Its core dataset covers 26 publicly traded PAEs, not the entire population of privately held patent-licensing firms. The authors acknowledge that private firms could behave differently, and that evidence about their number, size and performance would be needed to assess how far the results apply beyond the sample. They also examine three large privately held licensing firms, but do not claim that this resolves the broader data gap.

Nor do the results establish that PAEs are necessarily beneficial intermediaries. Showing that a sample does not fit predictions of the patent troll hypothesis does not, by itself, prove that it promotes innovation. That alternative also requires testable claims and systematic evidence. The study’s central contribution is to separate these questions rather than treat all firms that license patents as alike.

What technology decision-makers can take away

For industrial automation leaders, the broader lesson is to evaluate patent-related risk with evidence and specificity. Patent assertions can create legal costs and uncertainty, but the economic significance of a business model depends on factors such as the firm’s investment, financial performance, scale and conduct—not simply whether it makes products.

The study’s evidence does not settle the debate over patent litigation or establish that every PAE operates responsibly. It does show that broad claims about innovation impacts should be tested against firm-level data, clear definitions and the size of the market affected. For companies managing automation technologies and intellectual property, that is a useful basis for more precise risk assessment and policy discussion.

Access Official Technical PDF

About The Author