The Case Against Patents

Authors: Michele Boldrin and David K. Levine (Washington University in St. Louis) · Published: Journal of Economic Perspectives 27(1), Winter 2013, pp. 3–22 · URL: aeaweb.org, with the earlier version as Federal Reserve Bank of St. Louis Working Paper 2012-035A (September 2012)

One-line: No empirical evidence shows that patents increase innovation or productivity. The political economy of patent systems guarantees they grow stronger and worse over time, so the answer is to phase them out rather than try to reform them.

Saved to Kyle’s Apple Note “Things To Internet” (c. 2014) as “the case against patents’ mentioned on Planet Money.” The paper was the basis of Planet Money Episode 551: The Case Against Patents (July 9, 2014), which was prompted by Tesla opening up its patents and featured both authors.

Summary

The thesis, from the opening paragraph: “there is no empirical evidence that they serve to increase innovation and productivity, unless the latter is identified with the number of patents awarded – which, as evidence shows, has no correlation with measured productivity.” They call this the “patent puzzle.” Patents have grown enormously in number and legal strength without any matching speed-up in technological progress or R&D spending. The authors add that weak patent systems “may mildly increase innovation with limited side-effects,” while strong ones “retard innovation with many negative side-effects.” Their conclusion is to “abolish patents entirely through strong constitutional measures.”

Theory (section 2).

• The hold-up problem. A monopoly grant raises one innovator’s profits but lowers everyone else’s ability to compete. Modern products are built from thousands of patented components, so future inventions face “a gigantic hold-up problem.” Their example is Microsoft collecting Android licensing fees over a patent for “generating meeting requests and group scheduling from a mobile device.”
• Disclosure is a myth. An innovator patents only what couldn’t be kept secret for longer than the patent lasts, and “it is essentially impossible to build a functioning device or software program from a modern patent application.”
• First-mover advantage is enough. The first iPhone shipped June 29, 2007, and the first serious Android rival shipped October 22, 2008. By then more than 5 million iPhones had sold. Engineers are told not to read patents so they can’t be accused of willful infringement.
• Pharma is the partial exception, and not because of high fixed costs. It’s because the chemical formula and the clinical-trial results really are disclosed. The authors suggest prizes as the better tool even there.
• Industry life cycles. New industries start as competitive, imitative bursts: cars in the early 1900s, software in the 1980s and 1990s. Patents matter only once an industry matures and consolidates, when “the dead hand of dying institutions” uses them to tax entrants.

Political economy (section 3). Consumers are absent from “the patent game.” Inventors, patent trolls, the patent office, patent lawyers (legal fees above $7,000 a filing, about $1.5 billion a year on 2010’s 244,341 grants) and the courts all want more patents. Their test case is Amazon’s “one-click” patent, #5960411, which they argue fails the statutory “non-obvious” standard to anyone who has used a soda machine. Their conclusion is that the “optimal” patent system “is not of this world,” and advocating it “only offers an intellectual fig-leaf to the patent system we actually have.”

Evidence (section 4).

• A 2006 meta-study of 24 studies found “weak or no evidence that strengthening patent regimes increases innovation; they find evidence that strengthening the patent regime increases patenting!”
• Josh Lerner’s survey of 150 years of patent-law changes across countries found that strengthening protection did not spur innovation by the country’s own residents.
• In the authors’ own firm- and sector-level data, productivity and patenting mostly don’t correlate. Competition does correlate with productivity growth: the most competitive sectors grow up to 2% a year faster than the least competitive.

Conclusion (section 5). They quote Fritz Machlup’s 1958 report to Congress: if the US had no patent system, “it would be irresponsible … to recommend instituting one.” They add that six more decades of evidence have made the case weaker, citing software patents, which became possible in the early 1990s through judge-made law, and Bessen and Meurer’s Patent Failure. They compare patents to trade restrictions and propose phasing them out gradually through ever-shorter patent terms, which could be reversed if innovation collapsed. Meanwhile they’d take interim fixes: a proper reading of obviousness, real disclosure of working methods, and an independent-invention defense.

Full text

Archived privately against link rot: the St. Louis Fed working-paper version, ../attachments/the-case-against-patents/boldrin-levine-2012-st-louis-fed-wp-2012-035.pdf (25 pages, draft of June 29, 2012). The published JEP version is behind a Cloudflare check at aeaweb.org and wasn’t fetched.

Connections

• Patents: this paper argues the opposite of patents-as-innovation-metric. On its evidence, patent counts track patenting, not innovation or productivity.
• Intellectual Property: the authors treat patents as a “monopoly” right rather than a “property” right, and argue that difference explains why IP only ever expands.
• Open Source: open-source software is their contemporary example of “how collaboration and exchange of ideas thrives absent intellectual property,” alongside the Cornish steam-engine engineers.
• Regulatory Capture: section 3 is a public-choice capture story, citing Stigler (1971) directly. Voters don’t understand patents, so the industry’s own players write the rules.
• Innovation: the competitive theory of innovation (first-mover advantage plus imitation) is set against “Schumpeterian” monopoly-incentive theories.
• Planet Money: Episode 551 (July 9, 2014) is how the paper reached Kyle.