Jonathan Barnett’s Spring 2025 Regulation article “The Perils of ‘Free’ Information,” arguing for stricter intellectual property (IP) rights, confounds many different arguments about ideas and IP. He starts by ascribing to IP liberalizers a strawman argument, the “simple narrative in which IP owners exploit a monopoly.” As it happens, pretty much everyone—except Barnett (apparently)—agrees this is true, albeit with the proviso that the monopoly is government-granted. However, the narrative of students of IP—pro and con—goes well beyond that. It recognizes that awarding IP rights creates additional incentives to produce ideas, and that existing IP serves as a disincentive to create ideas. Hence, it is an empirical question as to which effect predominates. As Barnett’s arguments are entirely theoretical, he sheds no light on this central issue.

Having set up a straw man, the balance of his argument repeatedly toggles between two notions of “free”: free as in freedom and free as in free beer. The former refers to my freedom to repair my house after reading a book on home repairs, while the latter refers to my obligation to pay for the book. This makes the article difficult to follow because Barnett often seems concerned about IP-protected material being given away for free (as in beer) by the owners. Insofar as this is a problem, the solution is surely not IP.

Tech / A good portion of the article is a diatribe against YouTube and Android, and it is littered with misconceptions. YouTube, in Barnett’s view, steals content by paying creators to post their creations, which of course often includes clips from copyrighted materials. He dismisses the fact that, if they object to this use, copyright holders can get their content removed by filing take-down notices with YouTube; indeed, the company is so aggressive in responding to these notices that scammers have used them to get YouTube to remove content for which they do not hold the copyright. The bottom line, however, is that YouTube has created an entire ecosystem for creating content. If, in fact, the content was stolen and creators were not paid, then nobody would create content and there would be no YouTube. Instead, YouTube is flourishing.

Barnett’s discussion of Android confounds branding with IP. Android software runs on top of Linux, an open-source operating system that is by design IP-free. That refutes the idea that IP is needed for the creation of ideas.

Concerning web browsers, he writes:

This was the fate of Netscape, which pioneered the internet browser in 1994, but by 1999 had lost the market to Microsoft, which replicated the technology and gave it away as part of the Windows Office suite. For a single-product innovator like Netscape, the power of free was impossible to overcome.

Here he means free as in beer. As it happens, Netscape was protected by IP, and its founder, Marc Andreessen, is a billionaire. Second, Netscape didn’t vanish; it became free as in freedom, and it ultimately evolved into Firefox, which currently holds a share of the browser market comparable to Microsoft’s. Finally, Barnett overlooks that in the early introduction of software, products and firms come and go. Before Excel there was VisiCalc (labeled a “killer app” at the time), then Lotus 1–2‑3 (which killed the killer app). Now there is Google Sheets. Despite eventually losing the market, Visicorp and Lotus Development all made money for their owners, and their founders are all rich today.

Pharmaceuticals / Barnett considers pharmaceuticals to be a simple slam-dunk case for stronger IP rights. But there is nothing simple about pharmaceutical products, wrapped as they are in layers of government regulation (clinical trials) and protection (market exclusivity as well as patents). More to the point, there is plentiful evidence that IP and patents are not essential to the development of new pharmaceutical products.

Barnett recognizes that, in the absence of IP, profit is still possible by providing complementary goods and services, but he seems to think these cases are quite exceptional. They are not, in any industry, let alone the pharmaceutical industry. To understand this, realize that pharmaceuticals ideas are cheap, but the provision of products is expensive. In “Should Patents on Covid-19 Vaccines Be Waived” (Economic Observatory, May 14, 2021), Michele Boldrin, Flavio Toxvaerd, and I examine the costs and benefits to pharmaceutical firms in providing Covid-19 vaccines. The invention of the vaccines was extremely cheap, in one case taking a handful of people two weeks. As it happens, there is an important complementary good to the idea of a vaccine, and that is the vaccine itself. That is, you don’t need to sell the idea, just the vaccine. Now, if vaccines were easy to copy and produce, then everyone would be making and selling vaccines in their garages, and it would be hard for the inventors to earn a return on their two weeks of effort. As it happens, it is extremely difficult to produce vaccines and, despite huge investments, vaccine production took time to ramp up. Meanwhile, supply was small, demand large, and profit opportunities great.

Barnett’s view of supply chains suffers a similar problem. Take NVIDIA, source of the magical chips that power artificial intelligence. What does NVIDIA sell exactly? It sells blueprints for building chips. Would it go bust if those blueprints were not protected by IP? Hardly. It is no easier to produce a chip from a blueprint than it is to produce a vaccine from a chemical formula. Let’s be clear: China almost certainly has access to NVIDIA blueprints, so why doesn’t it produce NVIDIA-like chips? Because it is hard, and the expertise that designed the blueprints is needed to implement them. When you are building a chip foundry and you have a question, who would you prefer to ask: the person who copied the blueprints or the person who invented the blueprints?

Barnett winds up with a non-sequitur: an inaccurate historical analysis of IP rights over time that confounds the demand for patents with the effects of patents. I can only wish that IP rights were as weak as he believes.