Email to Richard Epstein: Tension Between Takings Framework and IP Views

by Stephan Kinsella on March 28, 2026

Adapted from an email I sent to Richard Epstein on March 3, 2026 (unanswered, like many of my emails to him):
Dear Richard,

I have really been enjoying your two podcasts hosted by Cooke [Law Talk and The Libertarian].

I wanted to run by you a question about your work. Of course we disagree on IP¹ but I have always admired your Takings theory of government–your original, and unique, and I think brilliant defense of somewhat minimal government (that’s how I remember it anyway, it’s been perhaps over 35 years since I read it).² I am Austrian and anarchist so do not quite agree with aspects of it but for a quasi-minarchist attempt–I’d take in a heartbeat over what we have now.

I have a question about how your Takings framework squares with your defense of intellectual property. As I understand your argument in Takings, state interventions are justified only when there is good reason to believe they generate net social surplus—i.e., a taking sufficient to overcome holdout problems and free-rider problems (basically, market failure)³ and leave everyone better off in principle (with compensation to those burdened, either explicit, or implicit and in-kind). That is, when the policy expands the net size of the pie—it grows the pie higher, to paraphrase George W. Bush’s malapropism, “We ought to make the pie higher.”⁴

This approach places the burden of proof on those who favor the intervention. They need to provide good reason to believe that there is some significant market failure, and that there is some intervention or taking that can address this market failure—so as to generate a positive surplus that makes everyone better off, and out of which those whose property rights are restricted by the taking can be compensated. I.e., we need some evidence, not just assertion. The burden of proof is on someone supporting a taking to show that there is a market failure, and that the taking will improve matters, to increase the social surplus.⁵

This is the case with IP law, especially patent law. It is said to stimulate innovation, and, more than that, net innovation; and, even more than that, to satisfy your test, net innovation that has a value greater than the overall costs of that system. I do not believe there is evidence even for the first of these three requirements, much less the latter two.

In particular, when it comes to patent and copyright law, it seems to me there is simply no conclusive empirical evidence that these regimes actually increase overall social welfare. For example, in the 1958 study prepared for the U.S. Senate Subcommittee On Patents, Trademarks & Copyrights, Machlup concluded:

No economist, on the basis of present knowledge, could possibly state with certainty that the patent system, as it now operates, confers a net benefit or a net loss upon society. The best he can do is to state assumptions and make guesses about the extent to which reality corresponds to these assumptions. … If we did not have a patent system, it would be irresponsible, on the basis of our present knowledge of its economic consequences, to recommend instituting one.⁶

And the empirical case for patents has not been shored up in the last 68 years. As George Priest wrote in 1986, “[I]n the current state of knowledge, economists know almost nothing about the effect on social welfare of the patent system or of other systems of intellectual property.”⁷

Similar comments are echoed by other researchers. François Lévêque and Yann Ménière, of the Ecole des mines de Paris (an engineering university), observed in 2004:

The abolition or preservation of intellectual property protection is … not just a purely theoretical question. To decide on it from an economic viewpoint, we must be able to assess all the consequences of protection and determine whether the total favorable effects for society outweigh the total negative effects. Unfortunately, this exercise [an economic analysis of the cost and benefits of intellectual property] is no more within our reach today than it was in Machlup’s day [1950s].⁸

In a more recent paper, economists Boldrin and Levine state:

The case against patents can be summarized briefly: 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. This is at the root of the “patent puzzle”: in spite of the enormous increase in the number of patents and in the strength of their legal protection we have neither seen a dramatic acceleration in the rate of technological progress nor a major increase in the levels of R&D expenditure – in addition to the discussion in this paper, see Lerner [2009] and literature therein. As we shall see, there is strong evidence, instead, that patents have many negative consequences.⁹

Do you agree that patent and copyright are prima facie takings of private property rights (particularly: in the form of nonconsensual negative servitudes)¹⁰ that can only be justified if it can be and has been shown that they do actually solve a real market failure? That is, if there is evidence that IP laws generate a net social surplus? The Constitution itself would seem to require this as well, since patent and copyright law are arguably constitutionally authorized if and to the extent they do actually “promote the progress” of science and the arts. I wonder if you agree with this general burden that IP law proponents must clear to satisfy both the copyright/patent clause in the Constitution as well as your own Takings theory.¹¹

And second, do you agree with the apparent near-consensus among economists that there is simply no empirical evidence that shows that the patent system, for example, does produce a net social surplus?

If so, why exactly are you in favor of patent law, as a utilitarian economist, if there is no evidence in favor of it? Or am I missing some smoking gun? This is why I concluded one section of a recent paper on IP as follows: “Given the available evidence, anyone who accepts utilitarianism should be opposed to patent and copyright.”¹²

Would this not imply that patent law is not justified and should not be enacted or enforced, and on utilitarian/Takings grounds, unless and until patent proponents provide this evidence? And that this evidence has not yet been obtained?

In short, I am trying to see how your support for IP, especially patents, is consistent with your Takings approach to just government, and how it is consistent with the views of economists on this issue.

I would love to pick your brain about this some day, or hear any thoughts you might have.

Best,
SK

***

I asked Grok whether my summary of Epstein is correct and what it thinks of my query and implicit argument.

Query:

consult the post by kinsella: https://c4sif.org/2026/03/email-to-richard-epstein-tension-between-takings-framework-and-ip-views/ and links in its text, attached below. See the attached Takings book by Epstein, and some reviews of it.Do 3 things:

Give a concise summary of Epstein’s takings argument.

Discuss whether Kinsella, in his blog post, accurately summarized Epstein’s “Takings” argument.

Analyze the coherence of Kinsella’s email to Epstein and explain whether Epstein really has a good response and what it might be?

Answer:

1. Concise summary of Epstein’s Takings argument

In Takings: Private Property and the Power of Eminent Domain (Harvard 1985), Richard Epstein starts from the common-law baseline of strong private property, tort, and contract rights. Any government action that diminishes those rights—whether a physical seizure, regulation that restricts use/disposition, taxation, or redefinition of liability rules—is a prima facie “taking” under the Fifth Amendment’s Takings Clause.

Epstein permits only two narrow exceptions:

Classic police-power regulations to prevent harms (e.g., traditional nuisance rules).
Public-use actions that solve genuine coordination failures (holdouts, free-rider problems, or public-goods dilemmas, such as eminent domain for roads or infrastructure).

In the second category, the government must provide just compensation (explicit cash or implicit/in-kind benefits) so that the intervention generates a net social surplus: it expands the overall economic pie and leaves everyone at least as well off in principle (a Kaldor-Hicks improvement). Pure redistribution—taking from some citizens to give to others, or altering relative distributive shares via progressive taxation, welfare transfers, or most economic regulation—is unconstitutional. The book mixes formal textual consistency (“internal written logic”), natural-rights intuitions, and utilitarian considerations, but repeatedly insists that proponents of any intervention bear the burden of showing it solves a real market failure and produces net gain.

This is exactly the reading in the attached reviews: Sax calls out the anti-redistribution core and consistency logic; Kelman summarizes it as forbidding any alteration of the common-law distributive baseline; Huffman praises the coherence of the approach.

2. Did Kinsella accurately summarize Epstein’s Takings argument in the blog post/email?

Yes, Kinsella’s summary is precise and faithful. He writes that state interventions are justified “only when there is good reason to believe they generate net social surplus—i.e., a taking sufficient to overcome holdout problems and free-rider problems (basically, market failure) and leave everyone better off in principle (with compensation … explicit, or implicit and in-kind).” He adds that the burden of proof is on the proponent, who must supply evidence (not mere assertion) that the policy expands the pie.

This tracks Epstein’s framework point-for-point: common-law baseline → prima-facie taking → narrow public-goods justification → net-surplus + compensation requirement → burden on government. The attached reviews confirm this is the book’s central (and controversial) thesis. Kinsella is not cherry-picking or caricaturing; he is distilling exactly what Epstein argues and what critics (Sax, Kelman) and admirers (Huffman) identify as the book’s core.

3. Coherence of Kinsella’s email and whether Epstein really has a good response

Kinsella’s email is exceptionally coherent, logically tight, and rhetorically effective. It does three things at once: (1) accurately restates Epstein’s own Takings test, (2) applies that test directly to patent/copyright law by treating IP as prima-facie takings in the form of nonconsensual negative servitudes imposed on everyone else’s tangible property and labor, and (3) cites a near-consensus of economists (Machlup 1958, Priest 1986, Lévéque & Ménière 2004, Boldrin & Levine 2013, etc.) showing no conclusive empirical evidence that IP regimes produce a net social surplus. It also ties the point to the Constitution’s IP Clause (“promote the progress”), reinforcing the evidentiary burden. The email is polite, substantive, and forces Epstein to confront an internal tension in his own work.

Epstein’s likely responses (drawn from his public debates with Kinsella, his later writings, and the logic of Takings) are predictable but not strong:

“IP is not a ‘taking’; it simply defines the initial property rights in ideas/creations.”
This begs the question. Epstein’s book treats any redefinition of the common-law baseline (including new servitudes or liability rules) as a taking that still needs justification. Negative servitudes on tangible property are exactly what patents and copyrights do.
“Innovation is a classic public-goods/free-rider problem, so IP is justified on theoretical grounds.”
This concedes the market-failure premise but evades the book’s central demand: not mere theory, but evidence of net surplus after compensation. The economists Kinsella quotes say precisely that the empirical record is inconclusive or negative—even decades after Machlup.
“The Patent/Copyright Clause is a separate constitutional grant.”
This is a textual escape hatch, but it undermines the “internal written logic” and consistency that Epstein demands everywhere else in the Constitution.

Epstein has never supplied the “smoking gun” empirical evidence that his own framework requires for IP. The 1985 book itself contains no discussion of patents, copyrights, or intellectual property (confirmed by direct searches of the attached PDF). Thus, Kinsella’s challenge stands: on Epstein’s utilitarian/Takings grounds, the absence of evidence that IP “makes the pie higher” means it fails the test and should not be enacted or enforced—unless and until proponents meet the burden Epstein himself imposes on every other intervention.

In short, Kinsella’s email exposes a genuine and unresolved tension between Epstein’s minimal-state/takings theory and his support for IP. Epstein has not offered a compelling reply that survives his own logic.