The Old Brandeis and the New Madison in Historical Perspective

The modern progressive view of antitrust, which has its origins in Louis Brandeis’s famous attack on “The Curse of Bigness,” regards large firm size as a source of improper political influence, regardless of the concentration of market power within the industry. That Brandeisian view, championed today by Tim Wu and Lina Khan, now in the Biden administration, calls for a vast expansion of antitrust enforcement. Unfortunately, all too often that approach has given a free rein to monopolistic activities in agriculture and labor, while at the same exposing productive firms to political attacks that all too-often work to prop up weaker market competitors. The New Madison approach of Makan Delrahim rightly exposes the overreaching of that approach in dealing with patent pools. And his cautionary attitude also warns against the counterproductive effects of the antitrust laws in other sectors outside the IP sector.

By Richard A. Epstein¹

 

I. INTRODUCTION

One striking feature about modern intellectual debates is how much emotional impact can be packed into a single word — big, as in bad. Today, a debate is raging in the overlapping areas of antitrust and patents. Their combination holds the keys to innovation, which everyone on both sides of the political spectrum seem, at least as at some abstract level, to support. But the means to that end, and its relationship to other social ends, divide the academic and practical worlds into two warring camps. On the one side stands the progressive tradition that attacks bigness. It most closely associated with the work of Louis Brandeis, both before and after he sat on the Supreme Court. Its modern exponents include two individuals, both holding prominent positions in the Biden Administration, who represent a powerful one-two punch. The first is Tim Wu, a Columbia Law professor and the author of The Curse of Bigness: Antitrust in the New Gilded Age,² about which I have expressed³ my deep reservations. The second is Lina Khan, a newly minted Columbia Law professor, recently installed as Chair of the Federal Trade Commission, and the author of a highly cited law review student note,⁴ Amazon’s Antitrust Paradox, which attacks the firm for its big size, even as it provides reliable services at low prices. I have also criticized⁵ her work for its failure to grasp traditional antitrust principles.

On the other side of the line, bigness is not so big. One of the intellectual leaders of the anti-big movement is former Assistant Attorney General for antitrust in the Trump Administration, Makan Delrahim. He has described,⁶ with much immediate impact, what he calls the “New Madison”⁷ approach towards antitrust and patent law. His views align squarely with the classical liberal tradition, with which I am closely associated. It flies under the consumer welfare standard, which insists that the sole function of antitrust law is to maximize the sum of consumer and producer surplus, not just the former. Champions of this approach include Robert Bork in The Antitrust Paradox,⁸ and Richard Posner, who had long written⁹ on the regulation of natural monopolies. It is no mystery on which side of this line my antitrust views are.

The consumer welfare view does not deny that there are other relevant issues — e.g. inequality and wealth distribution — but it does insist that different tools should be used to analyze these quite distinctive questions. Such a separation of topics is common in classical liberal thought for the simple reason no one policy instrument can adequately address two (or more) independent issues, which can converge or diverge in any given case. If the former, focusing on either will get you to the right place. If they oppose, someone has to decide which prevails and why. Hence on this view, in a first best world, taxation and welfare policy should address the inequality question; antitrust policy should seek to counter act behaviors that tend to reduce overall social output, whenever and however they occur. Patent law, meanwhile, encompasses the basic elements of this regulated area: patent-eligible subject matter as against those that fall into the public domain; allowable licensing agreements, which overlaps with antitrust principles; the nature of remedies, be them damages or injunctions or both; and the defenses available in cases of patent infringement. The fundamental empirical assumption under antitrust law is to protect competitive markets but to subject monopoly firms to regulation or break-up, at least if the transaction costs of these remedies are not too high. Such regulation could include structural remedies – breaking up the firm — or, when breakup is either impossible or too costly, conduct remedies — like forbidding tie-in or exclusive dealing provisions. Where patent law overlaps with antitrust is on the contested question of the nature of a patent: whether the exclusive right to sell should be regarded as tantamount to the creation of a permitted monopoly under the antitrust laws.

 

II. BRANDEIS, FIRST

The history of these two battling perspectives is worth expounding. The Brandeis position, which emerged in the two decades after the passage of the Sherman Antitrust Act¹⁰ of 1890, did not see patents as a way to aid innovation and did not see antitrust law as a way to deal with efficient market structures and behaviors. Rather, throughout the early progressives were motivated by a pervasive fear that large combinations of wealth were in themselves the source of major social dislocation. In that sense, they were out of sync with the initial burst of judicial activity on both subjects, which tended to be patent protective on the one side, and worried about monopoly power, not size, on the other. The issues were uppermost throughout government, academic and business circles, as part of a comprehensive response to the massive industrialization of the post-Civil War period.

The theme of this movement is not bigness, but contracts in restraint of trade. The roots of this movement date back to 1711 in Mitchel v. Reynolds,¹¹ when an employer sought to limit the freedom of former employer to ply his own trade after the employment contract had been terminated. The two parties may be able to negotiate an optimal deal for themselves, but the deal’s externality — the loss of another competitor — reduces consumer choice. Yet at the same time, the absence of some measure of protection renders the employer less likely to share trade secrets and other information with the employee, so that the business operates at less than ideal efficiency. The case is tailor-made for the rule of reason, which allows for some but not all potential restrictions. And the usual synthesis of “one-year, same location, same line of business” has endured just long enough to make it look like a sensible accommodation.

The hard question is whether that formula or approach can work when the size and scale of the business enterprises both in the United States and abroad have expanded greatly in the decades after the Civil War. The Brandeisian world view answered that question emphatically in the negative, and in so doing it captured the massive social unease with the new economic order. In his book,¹² Other People’s Money, Brandeis invoked the powerful image of “the curse of bigness,” which cast suspicion on the key “robber barons” of the age, based on their size alone, regardless of their behavior in any particular market or class of transactions. That overall suspicion made it unnecessary in this world view to offer any analysis of how this conduct could influence the quantity of goods and services sold, or the prices, terms and conditions on which they were sold.

Brandeis’ fear of concentrated political power in the hands of small elites swept aside any mundane concerns about price, terms and quantity of exchange. In this struggle, antitrust laws should be invoked to cut these large firms down to size, regardless of their performance in the any real world market. At the same time, smaller aggregations of farmers and workers (later joined by renters and consumers) had to enjoy protection for collective action among potential competitors that would expand their own position in the market — so that, in their case, size was not all that mattered, and its relative expansion was justified based on myriad factors. This schizophrenic attitude was quickly enacted into law with Section 6 and Section 7 of the Clayton Act. Section 7 expanded the reach of the antitrust law against firms by allowing the federal government to challenge any mergers whose effects “may be substantially to lessen competition, or to tend to create a monopoly,” even if they did not meet the more demanding standard under Sections 1 and 2 of the Sherman Act. At the same time, the farmers and the labor unions had a field day under Section 7, which exempted their ordinary activities from the Sherman Act with these stirring words: “The labor of a human being is not a commodity or article of commerce.¹³ Nothing contained in the antitrust laws¹⁴ shall be construed to forbid the existence and operation of labor, agricultural, or horticultural organizations, instituted for the purposes of mutual help, and not having capital stock or conducted for profit, . . . .” Little people could ban together to again in the way that big business could not.

The debates over the focuse of the antitrust laws led to huge partisan uproar, at a time when agrarian and labor interests were far stronger than they were today. Indeed, this development has had profound implications for the political economy of government regulation. One of the most famous articles on the subject, George Stigler’s 1971 A Theory of Economic Regulation,¹⁵ contains two interrelated propositions. The first attacks as myth the romantic view that legislatures somehow act solely in the public interest. The second is that in a world of competitive interest group contexts, the regulated industry often emerges as the victor in the struggle for political advantage. As I have recently argued, the activities in the first and second progressive periods (Woodrow Wilson and Franklin Delano Roosevelt) show that the first proposition has a lot of legs, but that the second is far more doubtful. In both periods, the large industrial firms lost out consistently to a group of farmers or workers whose larger numbers translated into votes, allowing for the triumph of the systematic protection of cartels of individuals. That observation does not translate well to all cases (think the organization of the Civil Aeronautics Board). It only means that the initial positions of, and the bargaining strategies by, the various interest groups resist any strong empirical generalization. The welfare implications in both cases are negative, but even here, all things considered, cartels are less efficient systems of production than single monopolies, which do not have to accommodate the same distributional demands that inefficient firms within a cartel do. Bigness has some social advantages.

There is yet another more damning implication in this set of cases. No matter who wins the struggle between cartels and monopolies, the public is always a loser. Thus, the Brandeis view talways looks for some reason to oust competitive markets, but it typically fails to justify the efficiency losses that result from blocking competitive outcomes. In particular, any supposed distributional account has to be suspect because the economic statuses of the winners and losers under these cartel arrangements are hard to understand. Some workers are excluded, and many consumers of both manufactured and agricultural goods and commodities are indeed less well off than the union and agricultural interests that secured the passage of these actions. It is even more ironic, perhaps, market forces being so powerful at this time, that the exemption from antitrust laws did not protect either of these groups from competition by new entrants. A generation later, when labor unions were under the National Labor Relations Act,¹⁶ Section 7 gave exclusive rights of representation that excluded competitors, and farmers received production quotas under the Agricultural Adjustment Acts of 1933 and 1938 that worked to prevent any cheating on the cartel. The Brandeisian model should never be used to justify big cartels.

The point, moreover, was not lost ironically on Brandeis himself. If his Other People’s Money was a tedious populist diatribe written on the concentration of wealth, Brandeis himself was in fact one of the leading practitioners of the classical liberal view that he challenged with his larger philosophy. The most notable instance was his orchestration¹⁷ of the United Shoe Machinery deal, which involved the coordination of multiple parties in the production chain, each with patent protection for certain processes needed to assemble the completed product. The United Shoe Machinery Company had a title that reflected the vertical integration within a single firm that was able to withstand early challenges under the antitrust laws in United States v. Winslow¹⁸ (1913) and United States v. United Shoe Machinery¹⁹ (1918). The argument in favor of this merger was two-fold. First, it meant that outsiders did not have to decide when a product failed, which of the various companies in the supply chain was responsible.

Second, it avoided the double marginalization problem²⁰ whereby each company in the chain of production, acting unilaterally, would charge prices that in to substantially reduced the producer surplus for the product. These are both neoclassical explanations, which did not apply in United States v. United Shoe Machinery Corp.²¹ (1968), when the Supreme Court broke up the company at the height of aggressive antitrust enforcement against mergers, when the firm faced tough Japanese competition that had already eroded any monopoly power than the company may have previously possessed. That decision was by no means alone — similar decisions in cases such as Brown Shoe v. United States²² (1962) and United States v. Von’s Grocery²³ (1966) also blocked standard mergers even though modern Brandeisians have never come up with a nonefficiency rationale to turn these social losses into social gains under standard metrics. The simple model of Oliver Williamson outlined in his article,²⁴ Economies as an Antitrust Defense: The Welfare Tradeoffs (1968), asked the right question about the relationship between operational efficiencies on the one side and increased monopoly power on the other, which has long provided an accurate roadmap. It applies to other situations that the Brandeisians addressed, such as the effects of the Robinson-Patman Act of 1936 in regulating price discrimination in various markets. The Act was justified as a way to protect small businesses against big businesses in the form chain-store competition. Its effect was anticompetitive by propping up inefficient firms to the detriment of consumers whose dollars now did not go as far.

The lesson of these examples is to never believe the antitrust horror story that some juggernaut firm will take over the entire market. In dynamic and resilient markets, small firms, even without legal protection, can specialize in various forms of consumer surplus to create goodwill and survive. It is common in evolutionary biology to note that the ocean does not consist only of whales, but rather has all sorts of different kinds of species, down to minnows, each occupying its own niche, which could grow or fall with time. The same applies to business. It is not only that the direct competitor of the big-chain store survives. It is also (as is common in shopping centers) that anchor tenants bring in their wake smaller businesses that have a synergistic interaction with the larger firms. These patterns are difficult to predict but in general easier to observe. It is not as though McDonald’s and companies have driven out all one-of-a-kind shops.

Lina Khan’s student note resurrected these old views that posited that an attack on Amazon was justified because it was able in the long run to sustain sales at lower prices than many of its smaller competitors. That logic gets matters backwards. The durability of Amazon’s pricing strategies makes it unlikely that these actions were forms of below-cost predation, which in most cases poses no social threat either. Nor does it take into account that, for all the protests about the employment practices of the firm, Amazon was easily able to beat back²⁵ a unionization effort in April 2021 for the simple reason that the employees thought that they got a better deal without going through a union middleman, who would skim off the top. The real beef against Amazon lies in its high-handed approach to selective bans on speech content — actions that would be unwelcome even if it had a tiny share of the market.

But, if the neo-Brandeisians cannot bring back the anti-chain store attitude of the 1930s, they can both in academic work and in proposed legislation try to do away with the framework of Section 7 of the Clayton Act by claiming that certain nascent mergers, as argued by Scott Hemphill and Tim Wu,²⁶ could eliminate future competitors that could go head-to-head with today’s giants. At least these efforts purport to work with traditional efficiency explanations, but in general the overall venture is ill-advised.²⁷ The current regime, while surely not perfect, is not way of whack. Yet, this new proposal gives no assurance that, left to its own devices, an acquired firm could have reached the same level of success. There are many small firms that are not acquired, but which fail because they do not have either the expertise of capital to mount a successful campaign. It may well be that the securities law makes things more difficult by imposing onerous requirements on taking private firms public, but if so, the remedy is not to distort antitrust law. It is to remove the barriers that stifle the operation of the securities market. Nonetheless, new legislation, as I have argued elsewhere,²⁸ is moving exactly in the wrong direction, including the proposal by Amy Klobuchar in her Consolidation Prevention and Competition Promotion Act.²⁹ Two wrongs never cancel out; instead, they tend to cumulate. We should then look upon with dread the prospects of these misguided Brandeisian reforms.

 

III. THE NEW MADISON APPROACH

The new Madison approach does not attempt to grapple with the Brandeisians across the full range of issues, but they do join issues on a number of point where the patent and the antitrust law might arguably collide. The source of this interaction is the claim that a patent confers a monopoly position on its holder, and thus automatically attracts antitrust scrutiny. But in fact, that false equivalence is mistaken both for real and intellectual property. The exclusive right to a given piece of land puts it in competition with other owners of other parcels. As was well recognized in the 1855 case of Allen v. Hunter.³⁰

Patentees are not monopolists. . . . No exclusive right can be granted for anything which the patentee has not invented or discovered. . . . [T]he law repudiates a monopoly. The right of the patentee entirely rests on his invention or discovery of that which is useful. And which was not known before. And the law gives him the exclusive use of the thing invented or discovered, for a few years, as a compensation for ‘his ingenuity, labor, and expense in producing it.’

In the modern context, patents might give someone the exclusive right to a given drug, but never give a right to exclude everyone else who wants to use a similar drug within the same class to treat the same or similar conditions. The device or the formula will differ which makes for competition between different sellers with exclusives over different products. Indeed, the secret for understanding how the patent universe operates is to draw,³¹ as I have argued elsewhere, analogies from the physical space to the intellectual property space.

Here is one recent example,³² offered by Robert Greenspoon, of how that process of analogy works. Patrick Leahy and Thom Tillis have proposed in the misnamed Endless Frontier Act³³ that the failure to register a patent should exclude the patentee from seeking extra damages for the willful infringement of a patent right, even for valid holders of the patent. But, the analogy in physical space is that the owner of newly purchased property cannot seek extra damages for willful trespasses unless that title is recorded. In both cases the nonsequitur is clear. To win the case against trespass or infringement the claimant has to show a valid title that is superior to the claim of the outsider. The purpose of recordation is to prevent double dealing in titles, but there is no reason why a party who fails to protect itself against one risk should be unnecessarily exposed to a second. It is a common feature of the Brandeisians to constantly seek to reduce the scope of patent damages, such that infringement carries with it little or no consequences. But weak remedies encourage infringement, which then puts a greater strain on the universe of voluntary transactions. A firm stand against deliberate infringements is therefore critical in dealing with tort remedies, whether for real of intellectual property, while nonetheless allowing for more accommodations with innocent intrusions that do not carry the serious risk of repetition, both by the initial wrongdoer and by eager imitators. The world in physical space cannot follow a regime where anyone can walk to an automobile dealership, take out a car, and then tell the original owner to bring a damage action. The whole set of antecedent criminal and civil remedies are needed to preserve markets.

To his great credit, Delrahim takes the same attitude on the more limited set of questions that he approaches. Most of his proposals overlap closely with work that I have done with Kayvan Noroozi — Why Incentives for “Patent Holdout” Threaten to Dismantle FRAND, and Why It Matters³⁴ — to develop the similar point that a robust set of patent rights is needed throughout the entire cycle of patent development, from the creation of the invention, formula ore device, to licensing, to production, to remedies for breach. Delrahim takes the right attitude on all the points that he mentions. He does not begin his analysis with issues of patent eligibility or nonobviousness. Instead he starts with the proper modes for the commercialization of patentable technologies, often by coordinating multiple distinct patents into more usable packages, chiefly by a range of licensing techniques. This effort is absolutely critical³⁵ because (as was the situation in United Shoe Machinery) the boundary lines around a single invention are often rather small. So assume that we have 100 patents, arrayed in a 10 by 10 grid, each of which is needed to development some special patentable device, like a handset. If each of the patents is treated as an island onto itself, the number of boundary lines in this grid is 220, with pitfalls along each of these margins. Yet if all of the patents are complements to each other, then the number of boundaries drops to four, each of which is longer and clearer than the multitude of other boundaries. The delineation of the new space makes it easier for any third party to license the bundle of patents, and easier for the group to fend off outsiders. The full protections of these devices do not carry any serious antitrust risk, which would arise if owners of substitute patents agreed to restrict output and raise prices, in what would be in any market a per se violation of the antitrust laws.

In light of these considerations it is clear that Delrahim is correct when he recommends that “an SEP [standard essential patent] holder reneging on their agreed-upon obligation with a Standard Development Organization (“SDO”) to license their patent to implementers on fair, reasonable and non-discriminatory terms (“FRAND”)³⁶ is fundamentally not an antitrust violation, but rather, fraud or a breach of contract.” The point here is that we want to encourage individuals to enter into these types of arrangements. That will not take place if the formation of a SDO becomes a trap door, so that once the agreement is made, it slams shut so that outsiders are entitled to treat actions that are said to renege on the basic agreement as the opportunity to bring a treble damage action against the party who is said to have reneged. The point is even more important in practice because the determination of who has reneged is often far from clear in litigation. There are constant readjustments in these pools, which could turn on the need to renegotiate the terms of the initial deal in light of changes in the relevant value of the various components. The antitrust hammer could easily lie over these transactions as well. The fraud and the contract remedies use only common law tools to deal with the usual pitfalls on contractual negotiation, which is all to the good.

It also follows from the basic understanding of these arrangements that innovation will be better achieved through Delrahim’s proposition that “an SDO should not be an apparatus for a coordinated effort by market participants to favor implementers over patent holders.” That outcome will not emerge if it is the innovators who are entitled to determine the scope and condition of their patent pool, subject to the antitrust limitation that substitute products cannot be put into the same patent pool. It is, of course, the case that the innovators have to have direct contact with the implementers in order to ensure some degree of complementarity. And, it is also the case that these pools will not stay together if injunctive relief is denied against third parties that seek to use one or more of the patents in the pool without obtaining the proper consent. Thus, Delrahim is also correct on his third point, already discussed above, that “an essential component of patent rights is the right to exclude, which should encourage courts to avoid limiting that right by disfavoring injunctive remedies.”

eBay v. MercExchange³⁷ remains as wrong now as when first promulgated in 2006 by deviating from the correct rule that injunctive relief is presumptively justified in all cases of breach until shown otherwise. It is always a mistake to use damages as a first cut remedy in these cases. The correct approach starts with injunctions, and then asks whether they should be conditioned or limited in some cases, including those of innocent infringement where massive dislocations can be offset by less intrusive remedies. The same rules that apply in the end cases should apply here as well.

The correct overlap between antitrust and patent law therefore shows the soundness of Delrahim’s last point, which is that “antitrust laws should consider a unilateral decision by a patent holder not to license a patent as ‘per se’ legal.” The point here follows from the general proposition that antitrust law should deal with cases of collective actions, not individual decisions, which deserve the same level of respect for intellectual property that they do for land or any other asset. The patent is thus subject to competition from other devices that are covered by other patents, so that they operate largely in a competitive environment. And, even if there is no immediate rival, the knowledge of a gap of competition in any given niche is likely to spur another firm to enter that space, so that the potential competition from entry exerts a sensible restraint on prices even if no new competitor has yet to appear.

 

IV. CONCLUSION

One way to assess the strength of these principles is to see what happens when they are violated. A key case is of course the 2019 decision of the District Court in Federal Trade Commission v. Qualcomm,³⁸ where District Court Judge Lucy Koh entered a massive judgment against Qualcomm by violating all these principles, as I have argued elsewhere.³⁹ Unilateral market decisions by Qualcomm on how to market its various devices and its general technology were subject to extensive review by the Court who then posited that it was appropriate to force the company to license its direct competitors world-wide, subject to supervision of the Court, which wanted to turn itself into a rate-making entity. In so doing, she asserted that a demand that the purchasers of Qualcomm chips had to obtain a Qualcomm license would have undercut the power of injunctive relief, even when Qualcomm had a constant practice of licensing its technology to any persons who wished to purchase their chips from other suppliers, including the highly successful Taiwanese company, MediaTek.⁴⁰

In so doing, the Court did not once consider whether the various schemes that Qualcomm had devised had resulted in higher prices, which, as shown⁴¹ by Alexander Galetovic and Stephen Haber, had not been the case, in light of the massive improvements in quality that paralleled sharp declines in prices resulting from new entry from other competitors. The market was working as it should, and it was a welcome relief that a strong Ninth Circuit⁴² decision overturned the District Court judgment before it could work serious harm. There is an evident lesson here, which is that the inefficiencies engendered by false accusations of monopoly power have no redeeming features. It is a lesson that the ascendant New Brandeisians should learn before it is too late.

---

1 Richard A. Epstein, the Peter and Kirsten Bedford Senior Fellow at the Hoover Institution, is the Laurence A. Tisch Professor of Law, New York University Law School, and a senior lecturer at the University of Chicago.

2 Tim Wu, The Curse of Bigness : Antitrust in the New Gilded Age.

3 Epstein, ‘The Curse of Bigness’ Review: Revisiting the Gilded Age, Wall Street Journal, December 2, 2018, available at https://www.wsj.com/articles/the-curse-of-bigness-review-revisiting-the-gilded-age-1543786330.

4 Lina M. Khan, Amazon’s Antitrust Paradox, 126 Yale L.J. (2016), available at https://www.yalelawjournal.org/note/amazons-antitrust-paradox.

5 Epstein, The Bezos Busters, Monday, September 17, 2018, available at https://www.hoover.org/research/bezos-busters.

6 Delrahim, Broke. . . but Not No More: Opening Remarks–Innovation Policy and the Role of Standards, IP, and Antitrust, Washington, DC,, September 10, 2020, available at https://www.justice.gov/opa/speech/assistant-attorney-general-makan-delrahim-delivers-remarks-leadership-virtual-series.

7 Department of Justice, New Heights for the New Madison Approach, available at https://www.justice.gov/atr/division-operations/antitrust-division-update-2020/new-heights-new-madison-approach.

8 Bork, Robert H. The Antitrust Paradox: A Policy at War with Itself. New York: Basic Books, 1978.

9 Posner, Richard A., “Natural Monopoly and Its Regulation,” 21 Stanford Law Review 548 (1968)., available at https://chicagounbound.uchicago.edu/journal_articles/1862/.

10 Sherman Antitrust Act, 1890, available at https://www.ourdocuments.gov/doc.php?flash=false&doc=51.

11 Mitchel v. Reynolds, Court of King’s Bench 24 Eng. Rep. 347 (1711), available at https://www.quimbee.com/cases/mitchel-v-reynolds.

12 Brandeis, Louis D., Other People’s Money, available at https://louisville.edu/law/library/special-collections/the-louis-d.-brandeis-collection/other-peoples-money-by-louis-d.-brandeis.

13 See https://www.law.cornell.edu/definitions/uscode.php?width=840&height=800&iframe=true&def_id=15-USC-509055121-1913737444&term_occur=999&term_src=title:15:chapter:1:section:17.

14 See https://www.law.cornell.edu/definitions/uscode.php?width=840&height=800&iframe=true&def_id=15-USC-66100559-1913737444&term_occur=999&term_src=.

15 See Stigler, George J., The Theory of Economic Regulation, The Bell Journal of Economics and Management Science, Vol. 2, No. 1 (Spring, 1971), pp. 3-21, available at https://www.jstor.org/stable/3003160?seq=1.

16 See https://www.nlrb.gov/about-nlrb/rights-we-protect/the-law/interfering-with-employee-rights-section-7-8a1.

17 See Epstein, Richard A., Antitrust Consent Decrees in Theory and Practice: Why Less Is More Aei Press (March 15, 2007).

18 United States v. Winslow, 227 US 202 – Supreme Court 1913.

19 https://supreme.justia.com/cases/federal/us/247/32/.

20 For illustrative images, see https://www.google.com/search?q=double+marginalization+problem&client=safari&tbm=isch&source=iu&ictx=1&fir=GLBkOQ307GucFM%252Cny06XdVEZUoUfM%252C_&vet=1&usg=AI4_-kR7uQov6d45D5OXZ44hQmTrQX0gmA&sa=X&ved=2ahUKEwiXoJ6P_7zxAhUJbs0KHRhfD1wQ_h16BAgMEAE.

21 United States v. United Shoe Machinery Corp., 391 U.S. 244 (1968), available at https://supreme.justia.com/cases/federal/us/391/244/.

22 Brown Shoe Co., Inc. v. United States, 370 U.S. 294 (1962), available at https://supreme.justia.com/cases/federal/us/370/294/.

23 United States v. Von’s Grocery Co., 384 U.S. 270 (1966), available at https://supreme.justia.com/cases/federal/us/384/270/.

24 Williamson, Oliver E. Economies as an Antitrust Defense: The Welfare Tradeoffs, The American Economic Review, vol. 58, no. 1, 1968, pp. 18–36. JSTOR, www.jstor.org/stable/1831653.

25 Herrera, Why Amazon Workers in Alabama Voted Against Union, WSJ, Aptil 10, 2021, available at https://www.wsj.com/articles/why-amazon-workers-in-alabama-voted-against-union-11618066800.

26 Hemphill, C. Scott and Wu, Tim, Nascent Competitors (June 11, 2020). University of Pennsylvania Law Review, Forthcoming, NYU Law and Economics Research Paper No. 20-50 , Columbia Law and Economics Working Paper No. 645, Available at SSRN: https://ssrn.com/abstract=3624058 or http://dx.doi.org/10.2139/ssrn.3624058.

27 Epstein, Richard A., Is There An Antitrust Crisis In Big Tech?, October 12, 2020, available at https://www.hoover.org/research/there-antitrust-crisis-big-tech.

28 Epstein, Richard A., Klobuchar’s Antitrust Blunder, available at https://www.hoover.org/research/klobuchars-antitrust-blunder.

29 See https://www.congress.gov/bill/116th-congress/senate-bill/307/text.

30 Allen v. Hunter, 1 F. Cas. 476, 6 McLean 303 (1855).

31 Epstein, Richard A., The Property Rights Movement and Intellectual Property, available at https://www.oxfordhandbooks.com/view/10.1093/oxfordhb/9780198758457.001.0001/oxfordhb-9780198758457-e-7.

32 Greenspoon, Robert P., Reflections on Unintended Consequences of Proposed Patent Law Amendments, available at https://www.ipwatchdog.com/2021/05/28/reflections-on-unintended-consequences-of-proposed-recordation-amendments/id=134046/.

33 See https://www.congress.gov/bill/117th-congress/senate-bill/1260.

34 Epstein, Richard A & Kayvan Noroozi, Why Incentives for “Patent Holdout” Threaten to Dismantle Frand, and Why It Matters, 32 Berkeley Technology Law Journal 1381.(2017). Available at https://chicagounbound.uchicago.edu/cgi/viewcontent.cgi?article=13768&context=journal_articles.

35 See https://www.innovationpolicyplatform.org/www.innovationpolicyplatform.org/printpdf/innovation_policy_platform_-_patent_pools_and_antitrust_-_2015-10-01/index.pdf.

36 See https://www.justice.gov/atr/page/file/1228016/download.

37 eBay Inc. v. Mercexchange, Ll, 547 US 388 – Supreme Court 2006, available at https://scholar.google.com/scholar_case?case=4819344338954570996&q=eBay+v.+MercExchange+2006&hl=en&as_sdt=8006.

38 FTC v. Qualcomm Inc., 411 F. Supp. 3d 658 – Dist. Court, ND California 2019, available at https://scholar.google.com/scholar_case?case=6588913420164874313&q=ftc+v+qualcomm+inc&hl=en&as_sdt=8006.

39 https://core.ac.uk/download/pdf/286364339.pdf.

40 https://www.mediatek.com/.

41 Galetovic, Alexander and Haber, Stephen H., SEP Royalties: What Theory of Value and Distribution Should Courts Apply (September 4, 2019). Available at SSRN: https://ssrn.com/abstract=3447641 or http://dx.doi.org/10.2139/ssrn.3447641.

42 FTC v. Qualcomm Inc., 969 F. 3d 974 – Court of Appeals, 9th Circuit 2020, available at https://scholar.google.com/scholar_case?case=5420885724600138871&q=Qualcomm+v.+FTC+ninth+circuit+2020&hl=en&as_sdt=8006.