10602 words as of January 15, entire file. 8000 permitted for the body.
Nos. 22-277, 22-555
----------------------------------
In the Supreme Court of the United States
Ashley Moody, Attorney General of Florida, et al., Petitioners,
v.
NetChoice, LLC, dba NetChoice, et al.
----------------------------------
NetChoice, LLC, dba NetChoice, et al., Petitioners,
v.
Ken Paxton, Attorney General of Texas
----------------------------------
On Writs of Certiorari to the United States Court of Appeals
for the Eleventh and Fifth Circuits
----------------------------------
Brief of Eric Rasmusen and Friends as Amici Curiae
in favor of Respondents in 22-555 and Petitioners in 22-277
----------------------------------
John H. Doe
Counsel of Record
Dewey, Cheatham, and Howe LLP
222 Main Street, Suite 4100
Houston, TX 77002
(713) 632-8000
QUESTIONS PRESENTED
1. Whether Texas House Bill 20’s content-moderation restrictions comply with the First Amendment.
2. Whether Texas House Bill 20’s individualized explanation requirements comply with the First Amendment.
This brief will only address question 1.
TABLE OF CONTENTS
Table of authorities....................................………………………………………………..ii
Interest of amici curiae.......................................………………………………………….3
Summary of the argument........................………………………………………………..4
The argument ..........................................................……………………………………... 5
I. In normal markets, antitrust law can be relied upon to prevent
monopoly................………………………………………………………… 6
II. Some markets are “natural monopolies” where antitrust law fails and a
different kind of regulation is needed.....................................………. 14
A. Increasing returns to scale is the most common reason for natural monopoly.............................……………………………………………………….. 15
B. Network externalities also create natural monopoly………………….... 2
C. An additional problem for large natural monopolies is who controls the corporation……………...........……………………………………...20
1. Control by managers harms all shareholders................……………. 2
2. Control by dominant shareholders harms minority
shareholders.....................................................………………………….. 2
3. Control by wealthy shareholders harms the public good…… 35
4. Control by government hurts all shareholders and the public good. ......................................………………………………….... 2
III. The social media platforms are natural monopolies, suitably regulated by
HB 20......................................………………………………….... 2
A. The social media platforms are natural monopolies………………........... 2
B. Regulation such as that of common carriers helps solve the problem of natural monopoly.…...............………………………………………………… 23
C. The Texas statute addresses “market failure” without falling into
“government failure”………………………………………………………..……...34
Conclusion....................................................…………………………………………. 2
Certificate of compliance with word count ………………………………………..35
Certificate of service………………………………………………………………………..45
TABLE OF AUTHORITIES
To be inserted here.
INTEREST OF AMICI CURIAE[1]
Forty-four amicus briefs in support of NetChoice have been submitted to this Court. Are they all making independent points? No. One wonders about Rule 37: “An amicus curiae brief that brings to the attention of the Court relevant matter not already brought to its attention by the parties may be of considerable help to the Court. An amicus curiae brief that does not serve this purpose burdens the Court, and its filing is not favored.” But these briefs do not violate Rule 37. Their existence is useful even if their contents perhaps are not: the number of them demonstrates the power arrayed on behalf of what we might without hyperbole call “The Lords of Tech” in this case of “TechLords v. Texas”.
Amici Eric Rasmusen and xxx do trust that the content of this brief is, if not useful, at least unique. We believe that economic theory is important to this case, even if though the questions presented concern the 1st Amendment, not money. We believe that the Court should, as background, consider the concept of “natural monopoly” in relation to regulation of Internet platforms and as the idea behind the law of common carriers. For convenience, this brief will be written addressing only the Texas statute, HB 20, not the Florida statute, but these underlying economics ideas apply to both of them.
The argument will be that two-sided internet platforms like Facebook are natural monopolies, markets that will inevitably be dominated by one firm, that their product is Speech, and that because they are natural monopolies they should be regulated. It is important in applying precedent and making gap-filling law to decide whether Facebook is more like a phone company or more like a magazine. Economics can help decide which analogy is appropriate by looking at general characteristics of both kinds of enterprises.
Amicus Eric Rasmusen is a retired professor, the former Dan R. and Catherine M. Dalton Professor of Business Economics and Public Policy at Indiana University. He has also held positions at UCLA and Chicago’s business schools, Harvard and Yale Law Schools, Oxford’s Nuffield College, and the economics departments at Harvard, Oxford, and the University of Tokyo. He is best known for his book on game theory[2], his work with J. Mark Ramseyer on the law and economics of the Japanese judicial system[3], and his seminal paper with Professor Ramseyer and Judge John Wiley on the economics of exclusive-dealing contracts.[4] He has taught in the George Mason economics-for-judges program a number of times, and has authored amicus briefs for the 5th Circuit, the D.C. Circuit, and the Supreme Court of Indiana.[5]
Although not trained as a lawyer, Professor Rasmusen has written many papers in law-and-economics. In this brief he does not pretend to deal adequately with 1st Amendment law.[6] Rather, he hopes to explain some ideas from economics that may be helpful to the Court.
Economics does not have definite answers. It has arguments and models. Thus, both law and precedent will be relatively absent in this brief. It does not rely on authority, not even economic authorities. Rather, the objective is to explain relevant economic concepts clearly enough that the Court can understand them and decide, based on their plausibility, whether to use them.
SUMMARY OF THE ARGUMENT
If someone wants to post a message on Facebook saying “Vote for Trump!” should Facebook be able to refuse?
It depends. The issue is whether Facebook is more like the phone company, or more like a magazine.
If Facebook is more like a local landline phone company, it can’t refuse. When John Doe wants to telephone a friend and tell him, “Vote for Trump!”, the phone company can’t drop Doe as a customer. It may argue that it owns the phone lines or that allowing such phone calls would make people think the company was pro-Trump, but a court will reject those arguments. If Facebook is more like a magazine, on the other hand, it can refuse. When someone wants to publish “Vote for Trump” as a letter to the editor, a magazine has every right to refuse him.
Which is the correct analogy, phone company or magazine?
The question has high significance. If Facebook can ban “Vote for Trump”, it can also ban posts by Trump supporters on any subject. Indeed, it can ban posts by any Republican.[7] It can even make a declaration of support for the re-election of President Biden a legal requirement for use of the site, part of the “terms of service”. This parade of horribles reminds us that we are deciding a question with the potential to profoundly affect American government.
One way to decide between phone company and magazine is to look at the law of common carriers, e.g. the five-part Thomas test of infra xxx. We need to go deeper, however. The internet is a new technology and neither common law nor constitutional should be applied without attention to their first principles. Precedent must be considered, of course, but which precedents are still on point? We must ask of each precedent why it was first established. Even followers of originalism and textualism must ask this question; it is not just a matter of what is good policy. Thus: What would the Founders have intended to do with the Internet? That is no more absurd than to ask what they would have thought of an Air Force. What does “freedom of speech” mean when applied to political arguments on the internet? —Not necessarily the same thing as when applied to a speech in the village square. As Ganesh Sitaraman and Morgan Ricks say,
Both Google’s response to Ohio’s complaint and some of the scholarly literature discussing common carriers and public utilities typify our formalist era. Rather than reasoning analogically, Google’s lawyers read common law opinions as if they are analyzing statutory text, and scholars seem perplexed when they cannot identify some singular criterion that will define what is or is not a common carrier for all time, in all places, and in all contexts. This formalistic impulse too often leads to the conclusion that the enterprise is futile. Tech Platforms and the Common Law of Carriers, Duke L. J. (forthcoming as of Dec. 20, 2023).
At the center of the cases is the idea of “natural monopoly”. This is the economic idea behind the legal idea of “common carrier”. In economics a monopoly is marked not so much by the firm’s market share as by its market power: its ability to raise price or reduce quality without attracting competition. A firm may have 95% of sales but zero market power because if it raised its price it would instantly lose all its sales to its competitors. Market power can arise in various ways. Antitrust law deals with what we might call “artificial monopoly”: market power created by firms merging with their competitors, by conspiring with them to keep prices high, or by driving them out with unfair practices such as threatening suppliers. But some monopolies are “natural”. They arise even if nobody commits a crime, and even if nobody has an advantage in cost technology or product quality.
The classic example of a natural monopoly is the electric company. The first company in a city to lay cables to each house will have an unbreakable monopoly without violating antitrust law. Any new competitor would have to lay new cables at high cost, after which competition between the two rivals would reduce the price too low for either of them to recover their cable-laying investment.
The electric company is a natural monopoly on the supply side, the cost side. There also exist natural monopolies on the demand side, the product side. In the days of landlines, only one local phone company could survive. Partly, that was because of the cost of the lines, as with the electric company. More important, though, is that customers want to join whichever company has the most other customers. Nobody wants to be a phone company’s only customer; the phone company with the biggest network would attract all of the customers.
Economists use the term “network externalities” for this idea that when customer joins the network he generates a positive spillover onto other customers, who are all the happier to be part of the network.[8] Conditions change. In 1924 the phone industry was a natural monopoly. In 2024 it is not, because technology (and law) makes customer interconnection between companies easy, and technology has evolved to where there is room for several companies to survive with reasonable profits.
Ordinary markets have higher prices and lower quality if they are monopolies. Natural monopoly, however, has a curious policy implication. An industry with a natural monopoly ought to be monopolized, because splitting it into two companies would be bad for consumers. Consumers do not want ten phone companies in a city; they want one, so they can call 100% of other people with phones. Even if the city government were in charge, it would set up one company, not ten competing companies. This is not because the one company would be particularly talented or innovative; it’s just that bigger is better. One badly-run big company is better than ten well-run small companies. Most industries are not like this, but most industries don’t have network externalities.
Despite its network advantage, however, a natural monopoly will set its prices too high and its quality too low because it’s immune to competitive pressure. It will keep to itself the advantages of bigness. It will be lazy, and the Invisible Hand of Providence will fail: the seller’s greed will not ultimately benefit the customers, but its shareholders (at best), but possibly just its managers, its majority shareholders, or even the government— if the government uses its power to extract advantages from the single business in control of the market.[9]
And so countries regulate their natural monopolies. Most commonly, the state utility commission requires a firm to serve all customers, maintain quality, and keep prices low. The law has come to call power companies “common carriers” to justify utility regulation, but the underlying idea is natural monopoly: a power company is carrying electrons, not customers, and its common carrier status is a legal fiction. A judge in 1750 would think it absurd to say that a tangle of wires was like a stagecoach. But it is. Both are natural monopolies within their large or small markets, because only one company can survive at each moment in time and space.[10] The law is right to look at the underlying idea instead of the exact product sold. As Isaac Asimov said,
I discovered, to my amazement, that all through history there had been resistance...and bitter, exaggerated, last-stitch resistance...to every significant technological change that had taken place on earth. Usually the resistance came from those groups who stood to lose influence, status, money...as a result of the change. Although they never advanced this as their reason for resisting it. It was always the good of humanity that rested upon their hearts.
For instance, when the stagecoaches came into England, the canal owners objected. Not that they would lose money, although they would, but they feared for humanity. Because as the stagecoaches tore along at fifteen miles an hour, the air whipping past the nostrils of the people on board, would by Bernoulli's Principle, suck all the air out of the lungs. . . . Well naturally the stagecoach people laughed heartily, and all they had to do was run a stagecoach at fifteen miles an hour with people inside and show them there's no harm. But they memorized the argument... for when the railroads came in. (The Future of Humanity: a Lecture by Isaac Asimov, speech at Newark College of Engineering, November 8, 1974.)
Asimov is right. Today the United States has come to accept the desirability of product innovation, even though blogs and webzines are driving newspapers and magazines out of business. But we still see a vigorous struggle to stifle the legal innovation needed to deal with technological innovation.
We will discuss the concept of natural monopoly in more detail below and argue that it fits large internet platforms. In this context, the worrisome result is not high prices or low quality of service, but the suppression of user speech.
THE ARGUMENT
The first part of the argument consists of a lengthy explanation of natural monopoly. The impatient reader who already knows the theory but wonders whether it fits social media platforms may skip to Section III.
I. In normal markets, antitrust law can be relied upon to prevent monopoly
Suppose a company makes 10 airplanes per day at a cost of $10 million in overhead and $20 million dollars for labor, materials, and machinery. The “average cost” is then $30 million/10 airplanes, which is $3 million/plane. We say that the $10 million for overhead is a “fixed cost”, because it must be paid regardless of whether 1 or 10 airplanes are made. If only one airplane is made, the average cost is $12 million/plane, ($10 million + $2 million)/(1 airplane). If two airplanes are made, the average cost is $7 million/plane; if three, $5.33 million; if four, $4.25 million; and so forth on to $3 million/plane for ten planes. The average cost falls as more planes are produced.
Suppose, however, that going beyond 10 planes per day requires workers to be paid overtime and puts extra strain on the machinery and materials budget, so that each additional plane incurs an additional cost-- called the marginal cost in economics-- of $4 million. Then 11 planes would cost a total of ($10 million + 10 * $2 million + 1 * $4 million)/11 planes, which comes to an average cost of $3.09 million/plane. Since the average cost was only $3 million/plane when producing 10 planes, now making more planes increases the average cost instead of reducing it. It turns out that making 12 planes would cost $3.17 million/plane; making 13 planes would cost $3.23; and jumping a bit, making 20 airplanes would cost $3.5 million.
This pattern is the common one for businesses. If a firm is too small, it can’t spread its fixed costs over enough units of output. If a firm is too big, its costs tend to rise faster and faster as it strains its capacity and its ability to manage the enterprise. The minimum average cost will be somewhere in between. If we graph the average cost on a diagram of, for example, dollar per ton of production on the vertical axis and tons of production on the horizontal axis, as in Figure 1, we get what is called a “U-shaped cost curve”.[11] Figure 1 shows a U-shaped cost curve, the curve labelled “average cost”. The diagonal line represents a smoothly rising marginal cost, the additional cost at each point of increasing output further.[12]
Consider the implications of the U-shaped cost curve on how much the industry is concentrated in a few firms. If the minimum average cost is at an output of 10 airplanes per month and about 50 airplanes per month are demanded by airlines, we would expect them to be produced by 5 companies, competing the price down to the minimum average cost. No company could dominate the market, because if it tried to produce all 50 planes, its average cost would rise and it couldn’t compete with a smaller company. If demand were 150 planes per month, we would have 15 companies; if it were 20 planes per month, we’d have only 2 firms.
For markets to be competitive, it must be the case that the cost curves are U-shaped and there is a big enough market to support a number of firms selling at minimum average cost. We would then expect competition to drive down the price to that level, since any firm trying to raise its price would lose business to the others.
What we might call “artificial monopoly” ruins this picture of an industry producing at the minimum efficient scale and selling at a low price. For example, if one company was well-connected politically, it might induce the legislature to pass a statute giving it the sole right to sell in the market, perhaps under the guise of making that quality was higher or that pollution was lower if just that company was operating. The lucky company would be able to charge a high price, even though its costs would be higher.
Another form of artificial monopoly is when an industry starts out competitive, but mergers and acquisitions are used to consolidate an original ten businesses into just one. This is how U. S. Steel was formed in 1901. The one company might have higher costs because it would be harder to manage and coordinate, but it would take time for new companies with lower costs to enter the market, and in the meantime the merged company would make extraordinary profits. In the case of steel, U. S. Steel’s market share fell continuously over the next 50 years due to the creation of Bethlehem Steel and other new entrants into the industry. With the passage of the Clayton Act in 1914, mergers like those that created U.S. Steel became illegal, and the Justice Department evaluates each proposed merger for its anti-competitive effect.[13]
A third form of artificial monopoly is the price conspiracy. Even if there are 10 firms in the industry, they could make an agreement--- also illegal now--- to all charge a high price. This is noteworthy, because courts have ruled it illegal for businesses just to talk to each other about prices, even if they do not make a formal agreement. This is a ban on speech, but courts have decided that this kind of government regulation does not violate the First Amendment.[14]
II. Some markets are “natural monopolies” where antitrust law fails and a different kind of regulation is needed
Our reasoning so far rest on the cost curves for the industry being U-shaped, so average cost first falls and then rises with output. That, in turn, depends on the marginal cost rising with output, so the extra cost of producing extra units rises if output rises beyond the output of the minimum average cost.
Suppose, instead, that the marginal cost is small and never rises. In our earlier example, suppose the overhead cost is $10 million and the marginal cost is $2 million/plane, but the marginal cost never rises to $4 million--- the firm can produce as much as it likes without straining its capacity or its management. Then the average cost will never rise. It will continue to fall, slowly approaching $2 million as the overhead is spread over more and more units of output. Bigger is better, forever.
As we will see below, “bigger is better” sometimes also applies to a product’s desirability to customers, the demand side.
III-A. Increasing returns to scale is the most common reason for natural monopoly
Not all industries have U-shaped cost curves. Consider Figure 2. It shows what happens when firms have a fixed overhead cost and then a constant extra cost for each unit produced. As the overhead is spread over more and more units, the average cost falls. It never rises again, because the extra cost per unit never rises.
Only one firm can survive in this industry unless they form a price conspiracy. Indeed, only one firm should survive. There is no sense in making two firms incur the overhead. If the customers owned the industry, they would want to consolidate it into one firm. The problem, however, is that though having one firm minimized the cost of product, it also allows that firm to operate as a monopoly, raising the price and reducing quality.
In Figure 2 the average-cost curve is falling over its entire length, and the marginal cost curve is flat. This means the company has a large fixed cost but constant marginal cost; it has to pay for overhead, but its extra production cost to produce an extra unit is always the same, no matter how much it produces. Thus, the bigger a company’s sales, the lower its average cost.
Suppose there are two firms in the industry and we think about having them compete price down to marginal cost and split the market, our usual ideal. What would happen? The problem is that our usual equilibrium conditions of price equalling marginal cost and zero profits are contradictory. Suppose the price equals marginal cost, and the firms split the market, at outputs of 50 each. For one thing, the price will be below average cost, so each firm will make a loss. For another thing, one of the firms would deviate by cutting its price, so it could increase its market share, since with the increased output, its marginal cost would be lower and it would profit from the extra sales. Whichever firm is bigger will have a cost advantage. So we would expect a costly war of attrition as the two firms each struggled along making losses in the hopes that the other firm would drop out, leaving it free to raise prices drastically as a monopoly.
III-B. Network externalities also create natural monopoly and that is a problem for Internet platforms
Internet platforms do have increasing returns. The company incurs a big fixed cost to set them up, and a smaller marginal cost to maintain them and to sell advertising. But that is not the main concern. Rather, the concern is with the natural monopoly created by network externalities. Network externalities are the demand-side equivalent of the supply-side’s increasing returns to scale. Increasing returns to scale make bigger better because as scale increases, the average cost falls. Network externalities make bigger better because as scale increases, each consumer is willing to pay a higher price.
Network externalities are harder to talk about than increasing returns to scale. Diagrams don’t help. Diagrams are good for economies of scale, as they illustrate the equations that form the most rigorous explanation. For network externalities, however, equations can also model the situation precisely, but there no good way to illustrate the equations. We will have to resort to words, since we cannot expect non-experts to spend the time to understand not just one equation, but a series of them that interact with each other.
In words: network externalities exist when each consumer values the product more if more other consumers are buying it. The classic example is the telephone. A telephone is useless if nobody else has one. If two people have them, they can call each other, but the usefulness is still limited. If a hundred people have them, each is willing to pay quite a bit more. If a hundred million people have them, each is willing to pay even more, since the group of consumers will include many people they know.
As a result of network externalities, the biggest phone company will be able to charge the highest price, because its product is more useful. If no government regulation is imposed, only one company will survive. Government regulation, however, in the form of a requirement that each phone company to place calls to its customers, can restore competition to the market by eliminating network externalities; all companies are in one big network.
Social media platforms have obvious network externalities, just like phone companies. People who want to post links or short message open to everyone to see will prefer the large company Twitter/X to any small company that tries to compete. People who want to post short or long messages just to friends will prefer the large company Facebook to a small company. People who want to post pictures to show other people will prefer Instagram to a small company. And so it goes. Note the difference from Google, another large company. Nobody cares if anyone else uses Google instead of Bing or DuckDuckGo. People prefer whichever search engine is best. Google is the biggest search engine, but it is not a natural monopoly. It succeeds because most but not all people like it better than its competitors. Google is not a platform; it is an app, and so has weak or nonexistent network externalities.
NetChoice’s response brief in the 5th Circuit says:
“Unlike the cable companies in Turner (and phone companies and railroads), websites have no natural monopoly over physical infrastructure. Reno, 521 U.S. at 870. And websites do not possess any bottleneck that would “destroy[]” an entire speech medium used by half of the country. Hurley, 515 U.S. at 577. Platforms lack “the physical power to silence anyone’s voices.” Zhang, 10 F. Supp. 3d at 437, 441.”
This is correct but irrelevant. A natural monopoly does not have to be based on economies of scale, the “physical infrastructure”, or on “possessing a bottleneck”. Invisible, non-material advantages are just as powerful, or perhaps even more powerful since at a big enough scale almost any company will start to have managerial problems that cause marginal cost to rise. [15]
In the current proceeding, the brief of amicus Digital Economists says at 24,
While network effects can have negative effects on competition, they also can be pro-competitive. Network effects mean that larger platforms can be more efficient than smaller ones, all else being equal.
The Digital Economists are correct that that network externalities means that larger platforms are more efficient. They are wrong in saying that this is pro-competitive. Consumers are better off with a large monopoly platform, usually, than with no platform at all. But this is true of any monopoly, even an artificial one. When U. S. Steel was formed by merging together 90% of America’s steel capacity, that made steel prices increase, but America was nonetheless better off than if all those steel mills evaporated. In the case of a natural monopoly, we can go even further. If the town’s monopoly telephone company is split into ten competing firms that don’t interconnect, prices will rise, but the product will be so much more useful that consumers will be worse off. As we’ve explained earlier, though, the efficiency of network externalities does not increase competition; it results in monopoly. “Efficient” is not the same as “Pro-Competitive”. Best of all, though, would be to have one phone company to get the benefit of the network externalities, but with regulated prices to avoid the excessive profits of a monopoly.
Also at 24, the Digital Economist’s brief says:
Even in Ohio v. Am. Express Co., antitrust violations were not found in the “indirect network effects” of two-sided platforms in merchant credit card networks or in anti-steering provisions. Ohio v. Am. Express Co., 138 S. Ct. at 2281.
This too is true but irrelevant. A natural monopoly, unlike an artificial one, does not violate antitrust laws. Neither the Sherman Act nor the Clayton Act says that being big, or being a monopoly, is unlawful. They just say that engaging in unfair business practices, or merging, or conspiring with other firms to create a monopoly is unlawful. If a company just happens to grow so much that its competitors go bankrupt, the company has not violated any laws and it is free to enjoy its high profits without government interference. Such is the case with natural monopoly. When Microsoft ran into antitrust trouble, it was not because Microsoft was overwhelmingly dominant in the market for computer operating systems. Rather, it was because Microsoft was engaging in unfair business practices related to bundling their products and making it difficult for consumers to buy the Windows operating system without also having to buy the Internet Explorer search engine.[16]
II-C. An additional problem is who controls the corporation
We must also consider who actually controls the internet platform This is important not just for economic efficiency or fairness; who controls the corporation goes to the heart of the 1st Amendment problem. The speech of the corporation may not be the same as the speech of the owners of the corporation. Shareholders do not just vote their shares to decide which users to ban and which to boost. Rather, by majority vote they elect corporate directors, who vote for the corporation’s president, who hires other employees--- all of them subject to government pressure. What emerges may not be the speech shareholders want. As a result, regulating a corporation’s speech may, rather than compelling it, actually lift the burden of compelled speech from its shareholders. It can save the shareholders from being associated with—and paying for--- corporate speech with which they disagree.
This problem of who controls the corporation has four aspects: managerial control, large-shareholder control, rich-shareholder control, and government control.
II.C.1. Control by managers harms shareholders
A corporation’s speech might not actually be the speech of its owners, but of the owners’ employees. Twitter had a transgender person in charge of censorship, and he banned The Babylon Bee.[17] Was Twitter’s President aware of this? What about all the other users banned by the censor? What about the shareholders? The Bee is a popular satire site, and would draw a lot of view and thus a lot of advertising revenue. HB 20 would have prevented the Bee from being censored. Would that be compelled speech for the shareholders? Instead, we must consider the possibility that HB 20 would have freed the shareholders from being compelled to speak by disloyal employees, including, perhaps, the President.
To be sure, if corporate managers are using corporate assets to push their personal political beliefs instead of maximizing profits, a takeover that replaced the directors and fired the managers would be a profitable venture. Hostile takeover is difficult if the corporation is large, however--- and it is precisely large corporations that HB 20 regulates. To be sure, that did work in the case of Twitter. Elon Musk saw an opportunity to profit by taking over the corporation. Whether he can make a profit is unclear, however, and if his acquisition proves disastrous, it will be the exception that proves the rule.[18]
II.C.2. Control by dominant shareholders harms minority
shareholders
The problem of managerial control arises because if there are many small shareholders, none can muster enough votes to replace faithless directors. If the corporation has a majority shareholder, or a few large shareholders who can successfully control the board of directors, the problem of managerial control disappears, but it is replaced by the problem of violation of the rights of minority shareholders. Corporate law prevents a dominant shareholder from such things as draining assets from the corporation telling his directors to sell them to himself at a cheap price, but the business judgement rule makes it difficult for minority shareholders to prevent the dominant shareholder from using the assets to advance his personal political positions. Minority shareholders of large platforms would like protection against that abuse of their rights.
II.C.3. Control by wealthy shareholders harms the public good
Even if there is just one shareholder, one who owned 100% of the shares, having such a wealthy person controlling the corporation can harm the public good. There is then is no problem of managerial control, or of abuse of minority shareholders. We still must worry, however, about the influence of wealth on government. A billionaire is more willing than someone owning stock in his 401(k) to be willing to sacrifice dividends in order to influence a presidential election. The ordinary shareholder may have strong political beliefs, but he generally wants the corporation to maximize profits rather than sacrifice them to political causes. Thus, he will favor political involvement by the corporation to be limited to activities that increase its profits, such as lobbying for government policies that affect the industry. A billionaire shareholder, on the other hand, will be willing to spend more on politics, just as he is willing to spend more on homes or yachts. We recognize this with statutory limits on political contributions to candidates, but if the contribution is in the form of corporate speech, we do not. If a corporation is run to make money, greed is good, as Adam Smith said in the quote earlier in this brief (supra at xxx). If a corporation is run to obtain political power, even nobility is suspect, since in a democracy we are apprehensive about someone controlling debate even when his motives are unselfish.
II.C.4. Control by government hurts all shareholders and the public good
Statute HB 20 also saves Facebook from government pressure to censor. This is unlawful, but it happens anyway:[19]
As the Supreme Court held in Norwood v. Harrison (1973), it is an “axiomatic” principle of constitutional law that the government “may not induce, encourage or promote private persons to accomplish what it is constitutionally forbidden to accomplish.” That’s exactly what the Twitter Files show officials from the Federal Bureau of Investigation, the Centers for Disease Control and Prevention, the Central Intelligence Agency, the Department of Homeland Security and other federal agencies doing—inducing and encouraging Twitter to censor constitutionally protected speech. Jed Rubinfeld, Facebook Bowed to White House Pressure, Removed Covid Posts, The Wall Street Journal (2023).
HB 20 would restrict the platform corporation from sacrificing profits under government pressure, and this would both prevent illegal pressure’s corruption of democracy and help all the shareholders of the corporation.
III. The social media platforms are natural monopolies, suitably regulated by HB 20
But are the social media platforms natural monopolies? And is HB 20 an appropriate way to regulate a natural monopoly? Even if it is appropriate in theory, will it be politicized and abused by the government? We will now address these questions
III.A. The social media platforms are natural monopolies
Are the social media platforms natural monopolies? Consider Twitter/X. Geoff Manne tells us in his amicus brief that Facebook has market share of 50%, Instagram 16%, Twitter 15%, and YouTube 2% (https://www.statista.com/statistics/265773/market-shareof-the-most-popular-social-media-websites-in-the-us/). With a market share of 15%, how can Twitter be called a monopoly?
Recall that to the economist, the problem of monopoly is not that only one firm is in a market, but “market power”: that a firm can raise its price or reduce its quality without losing customers. This avoids the problem of how define “market” and “product”. If we define the market as “social media platforms”, Twitter has a market share of 15%. If we define the market as “computer software”, adding in every computer program ever written, Twitter’s market share becomes miniscule. If we define the market as “social media platforms that limit posts to 280 characters or less”, Twitter has 100% market share. Rather, we need to ask if Twitter has market power. If this were an antitrust case, expert witnesses would spend months doing statistical analyses to find out whether advertising rates would rise if Twitter were to merge with You-Tube, comparing the market power of the companies now with the increase in market power if they merged. The Court would follow the Clayton Act in asking whether the merger “the effect of such acquisition may be substantially to lessen competition” (15 U.S. Code §18). The companies and the Antitrust Division would each present their witnesses in a lengthy trial.
One way to deal with HB 20 is to remand the case to District Court for a trial to conduct a trial like this as part of considering the question of whether HB 20 violates the First Amendment rights of these companies. But the question is whether Twitter could drastically change its moderation policies without losing users to other social media platforms. The Musk takeover makes the answer obvious. Under Musk’s policies, liberals are unhappy because Twitter tolerates fellow users such as The Babylon Bee. They say they’ll leave for other platforms, or even create new platforms that will be less tolerant of conservative posts. But they stay. Twitter is still the social media platform for journalists and pundits. On the other hand, conservatives were very unhappy under the previous management that banned the Bee. Yet they remained on Twitter (if allowed to) unhappy though they were. Twitter can change its policies drastically and keep its customers even though large numbers of customers hate the changes.
Does anyone really believe Facebook or Instagram or You-Tube are different—that if they became more tolerant of dissenting political voices liberal customers would leave, or more conservatives would join if they didn’t have their current strict policies? These are all social media platforms, but no one has succeeded in entering and competing with them head to head, despite the technological ease of doing so and the large advertising profits that could be earned. Nor do they compete with each other. Journalists unhappy with Twitter do not switch to Facebook or Instagram. Podcasters unhappy with You-Tube do not switch to Twitter. Users who are banned do switch, of course, but the switch is not voluntary and the substitutes are vastly inferior, which is why banned users complain. If the big platforms had no market power, users could laugh off being banned—they would simply switch to another platform and have just as many viewers and just as much ad revenue. But users hate to be banned, even if they complain about their current platform’s moderation policies. This is sufficient evidence that the large platforms have market power.
Casual observation also tells us why amicus Francis Fukuyama’s middleware argument fails. “Middleware” is software that picks and chooses among pages on different Internet platforms to choose the mix that a given user tells the software to select. For example, a consumer might use a service that selects the pages with the 5 lowest prices for AA batteries among all the seller websites on the Internet. Similarly, a consumer could select both posts from Twitter and posts by users Twitter had banned and sent to tiny pseudo-Twitters. Fukuyama says at 12,[20]
Another way of empowering users and decentralizing control over speech is to require interoperability between social media platforms. In technical terms, systems are interoperable when they speak a common language, or when systems provide the capacity to share, interpret and present data in a way that the other systems can understand. Mike Masnick has shown how the existence of common protocols for information can disrupt the control over speech that is currently “centralized among a small group of very powerful companies.” Mike Masnick, Protocols, Not Platforms: A Technological Approach to Free Speech, Knight First Amendment Inst., Colum. Univ. (Aug. 21, 2019), https://perma.cc/UPY2-CRL6 (last visited Dec. 5, 2023). . . .
Given that social media platforms naturally create network effects, these effects can be limited by allowing users to choose the moderation regime most appropriate for them. Producers of moderation regimes can compete with one another, including the original platform. See Fiona Scott Morton & Michael Kades, Interoperability as a Competition Remedy for Digital Networks (March 19, 2021), https://perma.cc/DYQ4- PRFW (last visited Dec. 5, 2023)
Regulation could take the form of interoperability requirements enforced by a government agency, as with phone companies and the Federal Communications Commission, rather than must-carry requirements, as with HB 20.
Fukuyama is correct in saying that this kind of middleware would reduce network externalities, perhaps even eliminating them. Indeed, this is how weblogs work: the user chooses which authors to follow. Each user follows a different set of blogs, and each blog has no more market power than an author has in the market for novels. But this is not a solution. Social media platforms replaced blogs for a reason. They provide network externalities that blogs do not, by making it easier to communicate with other users. Middleware is supposed to solve that problem, by emulating the networks and convenience social media platforms provide. But no such middleware exists at present. It does for shopping—though Amazon still has market power--- but it does not exist for social media, at least in a way that has appreciable impact. Dismissing HB 20 as unnecessary because middleware solves the problem is like saying that the Antitrust Division should allow the merger of all natural gas companies because with the impending advent of cheap solar panels for roofs, natural gas would have no market power in the home heating market. Maybe eventually—but not now.
III.B. Regulation such as that of common carriers helps solve the problem of natural monopoly
There are several policies to deal with natural monopoly. Sometimes the government owns the natural monopoly, as with the water supply in many cities. Sometimes the government sells the right to provide the natural monopoly’s product, as with contracts for garbage disposal. Sometimes the government establishes a regulatory commission, as with the Interstate Commerce Commission or state public utility commissions. Any of these might help in the case of Internet platforms; all have their own problems. The problems are particularly difficult because the problem is not so much high prices for advertising as discrimination in access to the platform.
The Texas statute uses numbers of customers to decide which firms have market power. It would be costly, though possible, to measure the market power of each company individually, as antitrust law does in merger cases. We don’t do that for public utility regulation, however: electrical distribution is generally a natural monopoly, so we don’t examine the electricity industry in each city separately to decide whether the sole producers could get away with raising prices. Here, number of users is a practical way to implement regulation, like regulations that apply to companies with more than 50 employees, or tax rules that apply only to large taxpayers. Indeed, we use that in antitrust: only mergers with values greater than xxx must report the planned merger to the Justice Department for examination.
“The challenged laws would apply to such entities based on monthly users at the national level or gross revenue. See Fla. Stat. §501.2041(1)(g)(4) (covered providers must have at least 100 million monthly users or $100 million in gross annual revenue); Tex. Bus. & Com. Code §§ 120.001(1), .002(b) (covered social media platforms have 50 million monthly active users). But raw revenue or user numbers do not show market power. It is, at the very least, market share (i.e., concentration) that could plausibly be instructive— and even then, market power entails a much more complex determination.
Geoff Manne amicus brief at 27 (cleaned up).
“Narrowly tailored” means a remedy for a problem that will not sacrifice too much free speech to obtain enough of another good thing. The reason for it is that we want to promote free speech. For example, wartime censorship sacrifices the free speech of reporters in order to obtain success in battle, but that does not require complete censorship of newspapers.
Here, though, we are not sacrificing a little free speech to get a lot of national security; we are sacrificing a lot of free speech to get a little free speech. We are trading off the free speech of a few giant internet-platform corporations against the free speech of millions of U. S. citizens. Or, to our mind, we are trading off the property rights of the corporation against the free speech of the citizen, since it is whimsical to call the hosting of a few opinions while hosting a thousand different other opinions to be compelled speech. To be sure, the argument is, more precisely, that if Facebook allows a pro-Nazi post, then people will think that Facebook wants to tolerate pro-Nazi speech; that is to say, Facebook is alarmed that people might think they support the First Amendment. The argument works exactly the same for common carriers such as bus companies. If a bus company must allow passengers to argue in favor of Trump’s election, the company might say that is compelled speech, because tolerating such people means the public will think the company supports Trump. If Facebook can kick Trump supporters of its servers to avoid compelled speech, why shouldn’t Greyhound be able to kick them off of its buses?
Common carrier law is a form of natural monopoly regulation that recognizes the impracticality of precisely measuring market power but the ease of determining situations where it is likely to be high.
Writing in the early 1900s, Bruce Wyman suggested that in the infancy of England’s trade economy in the 14th and 15th centuries, the special common calling duties applied to all trades and businesses, because in any area, few persons were engaged in each trade and the problem of monopoly or market power abuse was thus endemic.
(See Adam Candeub, Appendix I, Joint Appendix.)
The Thomas Test for whether a company is a common carrier asks:
(1) whether a firm exercises market power,
(2) whether an industry is affected with “the public interest,”
(3) whether the entity regulated is part of the transportation or communications industry,
(4) whether the industry receives countervailing benefits from the government, or
(5) whether the firm holds itself out as providing service to all.
Biden v. Knight First Amendment Inst., _ U.S. _, 141 S. Ct. 1220, 1222-23 (Thomas, J., concurrence on denial of certiorari).
Monopoly should be seen in terms of market power-- the ability to raise the price or change the quality (demand elasticity; sensitivity of quantity demanded to price). Common carrier doctrine does for that, even if sellers are small in size and there are many of them.[21] A medieval ferry was a small business, but it had a natural monopoly over crossing the river.[22] There were many stagecoaches in England, but maybe only one on the route you want to travel, and certainly only one for the time and route you want to trave. They were small natural monopolies. There was a real danger that if you showed up one morning to ride from Cambridge to Ely, the driver might size up the value of your clothing and decide to charge triple the usual price.[23] Economists have noted that in small towns, the prices of plumbers, tire stores, and dentists are highest in the smallest, where there is room for only one business, and fall sharply once the market is big enough to be served by two. Timothy Bresnahan and Peter Reiss, Entry and Competition in Concentrated Markets, 99 J. Pol. Econ. 977 (1991).
Google is a large business, both absolutely and in comparison to a rural dentist, but Google is not a natural monopoly, in contrast to the social media platforms in Paxton v. Netchoice. When someone uses Google to look up the meaning of “estoppel” yet again, who cares whether other people are using the same search engine? All that matters is how good a job Google does. There are no network externalities. Google has a large market share, but that is because it is very good at searching, something it had to be to beat the previously dominant search engine, Yahoo, and in part because Yahoo, too, faces the threat of competition from new entrants such as Bing. Don Reisinger, What would it take to beat Google? We take a look at the key success factors and the opportunities for other companies in the search engine space. CNET (Jan. 2, 2009). Thus, the suit launched by the State of Ohio against Google in 2021 to have it declared a common carrier was misguided. Ohio ex. rel. Yost v. Google LLC (complaint filed June 8, 2021) Common Pleas Court, Delaware County, Ohio.
III.D. The Texas statute addresses “market failure” without falling into “government failure”
In 1982, George Stigler won the Nobel Prize in Economics for his work on “government failure”. Adam Smith showed that free markets maximize national wealth. Later, economic theory made precise the limitations of free markets, the most important work being perhaps A.C. Pigou’s 1920 The Economics of Welfare. Economists label the exceptions to the efficiency of free markets as “market failure”. This is not a critique of free markets generally, but a list of exceptions and caveats that includes spillover effects (e.g. pollution), information imperfections (e.g., fraud), unclear property rights (e.g., ill-developed property law in developing countries), and market power (e.g., price conspiracies). This last is what concerns us here. Natural monopoly results in market power, and justifies government regulation, as an exception to the general efficiency of the free market. In most markets, regulation would reduce national wealth (socialism being the extreme example), but if we can point to specific market failures, regulation increases it.
Stigler and others, however, pointed out that government regulation is not necessarily the best response even in markets where there is obvious market failure. The problem is government failure: that governments don’t always choose the optimal regulations. The cure can be worse than the disease. He brought to the notice of economists something policymakers have always known: policy is made by pressure from interests groups, who often, perhaps usually, care only for their only welfare, not for the public welfare. One might say there is market failure in the market for policy. If government failure is likely if the government is allowed to regulate a market, the best policy is laissez faire even if the market is not working perfectly. The cure is worse than the disease.
The classic article is Stigler and Friedland’s 1962 “What Can Regulators Regulate? The Case of Electricity.” (George Stigler and Claire Friedland, 5 J. L. & Econ. 1) They took a classic example of natural monopoly-- electric utilities-- and argued that regulation had been “captured” by the regulated companies and operated for their benefit, not for the customers. They said that although state utility commissions were set up with the intent of helping customers, consumer ignorance and inattention led to utility commissions actually helping the utility companies by preventing competition between them.
Whether Stigler and Friedland were correct about public utility commissions has been debated, as has whether the cure is truly worse than the disease, or merely, like most powerful medicines, has severe side effects but is still better than not treating the patient at all. Nonetheless, all economists now accept that in policymaking both market failure and government failure need to be considered. Professors may know what cures market failure, but politicians know what wins elections.
In the case of internet platforms, the potential for government failure is obvious. Suppose Facebook does discriminate against Republicans, since it is a large company and monopoly. One solution would be to regulate it by nationalizing the company. But do we really think Republicans, and even most Democrats, would benefit if President Biden controlled the company instead of Mark Zuckerberg? Of course not, even if Congress passed a statute saying that the executive branch should avoid political considerations in its operation of Facebook. Another solution would be to establish a federal agency to regulate Facebook. This would fare no better, whether it was directly under political control like the Justice Department or less directly such as the Federal Trade Commission. Nor would an even more indirect system of agency control such as the Federal Reserve work. A “Freedom Reserve” would be much more independent of the President, but what about the “Freedom Chairman”? Now the power would be with him, and we are back to the same problem as when Zuckerberg controlled the natural monopoly.
Government failure is a powerful argument. Is it really better to have the Governor of Texas controlling Facebook instead of Zuckerberg, with all the problems we have been discussing? No.
The policy in question in this case, however—HB 20--- has taken government failure into account. It neither nationalizes Facebook nor puts it under the control of a state regulatory commission. Instead, Facebook is put under a law. This is what “narrow tailoring” is all about: finding a policy remedy to a problem that is narrow enough not to be abused by the government.
HB 20 does not give the Governor of Texas either the power to censor people’s social media posts or to impose his own. Nor does it give any power to an agency or commission—not even the power to enforce the statute. HB 20 is merely a law, a government directive to be enforced by the Attorney-General of Texas in the same way he enforces law against burglary, speeding, or violation of minimum wage laws. As in every case of government enforcement, the remedies and procedures are crucial but neglected in our discourse, not only in public discussion but even in judicial challenges. The relevant section of HB 20 is this:
SUBCHAPTER D. ENFORCEMENT
Sec. 120.151. ACTION BY ATTORNEY GENERAL.
(a) The attorney general may bring an action against a social media platform to enjoin a violation of this chapter.
(b) If an injunction is granted in an action brought under Subsection (a), the attorney general may recover costs incurred in bringing the action, including reasonable attorney’s fees and reasonable investigative costs.(https://capitol.texas.gov/tlodocs/872/billtext/pdf/HB00020F.pdf#navpanes=0)
The Attorney-General does not have day-to-day control of the internet platform, nor is that his only mission, as it would be if an agency were created to enforce HB 20, nor does he have a large staff of experts focussed on HB 20. He only has this law added to the many other laws under his jurisdiction. Any effort by him to enforce the law will be in a public filing before a court of law, and if he persuades the court that a violation has occurred, the only result in an injunction and his cost of bringing the action, not jail time, and not even a civil fine. Government failure is minimized, and in fact will more likely be by the Attorney-General’s neglect of enforcement than by overenforcement.
CONCLUSION
The decision of the 11th Circuit in No. 22-277 should be reversed, and the decision of the 5th Circuit in No. 22-555 should be affirmed.
Respectfully submitted,
/s/ John Doe
John H. Doe
Counsel of Record
Dewey, Cheatham, and Howe LLP
222 Main Street, Suite 4100
Houston, TX 77002
(713) 632-8000
Counsel to amici
Certification of compliance with the word limit
This Amicus filing supporting Appellant’s motion contains sdfsf words, excluding the parts of the motion exempted by rule. This filing complies with the typeface requirements of Rule 32(a)(5) and the type-style requirements of Rule 32(a)(6) because it has been prepared in 12-point Century Schoolbook using Microsoft Word.
Respectfully submitted,
/s/ John Doe
John H. Doe
Counsel of Record
Dewey, Cheatham, and Howe LLP
222 Main Street, Suite 4100
Houston, TX 77002
(713) 632-8000
Counsel to amici
Proof of service to the parties
Rule 29 says:
3. Any document required by these Rules to be served may be served personally, by mail, or by third-party commercial carrier for delivery within 3 calendar days on each party to the proceeding at or before the time of fling. If the document has been prepared as required by Rule 33.1, three copies shall be served on each other party separately represented in the proceeding. If the document has been prepared as required by Rule 33.2, service of a single copy on each other separately represented party suffces. If service is by mail or third-party commercial carrier, it shall consist of depositing the document with the United States Postal Service, with no less than frst-class postage prepaid, or delivery to the carrier for delivery within 3 calendar days, addressed to counsel of record at the proper address. When a party is not represented by counsel, service shall be made on the party, personally, by mail, or by commercial carrier. Ordinarily, service on a party must be by a manner at least as expeditious as the manner used to fle the document with the Court. An electronic version of the document shall also be transmitted to all other parties at the time of fling or reasonably contemporaneous therewith,
Respectfully submitted,
/s/ John Doe
John H. Doe
Counsel of Record
Dewey, Cheatham, and Howe LLP
222 Main Street, Suite 4100
Houston, TX 77002
(713) 632-8000
Counsel to amici
[1] No counsel for a party authored this brief in whole or in part, and no counsel or party made a monetary contribution intended to fund the preparation or submission of this brief. No person other than amici curiae or their counsel made a monetary contribution to its preparation or submission.
[2] Eric Rasmusen, Games and Information (1st ed. 1989,4th ed. 2006, also translated into Japanese, Spanish, Italian, French, Chinese (Taiwan) and Chinese (Mainland).
Here and infra, citations are underlined to indicate hyperlinks in the pdf file of the brief, provided for the convenience of readers to take them to the full source. This is in lieu of printing the web addresses. In the working paper version, these are also in the blue font that has become standard for documents with hyperlinks; in the brief as filed, all fonts are black, as required by Court rules.
[3] Much of it summarized in J. Mark Ramseyer and Eric Rasmusen, Measuring Judicial Independence: The Political Economy of Judging in Japan (2003).
[4] J. Mark Ramseyer, Eric Rasmusen, and John Wiley, Naked Exclusion, 81 Am. Econ. Rev. 1137-1145 (1991).
[5] Barnes v. Indiana, 946 NE 2d 572 (Ind. S.C. 2011); U.S. v. Marshall, 771 F.3d 854 (5th Cir. 2014) In Re Flynn, 973 F.3d 74 (D.C. Cir. 2020).
[6] Notably, NetChoice’s opening brief does not mention natural monopoly or common carriers at all. Might NetChoice be saving that for the Reply brief because they wish to diminish the salience of the argument? Or perhaps they are right and it doesn’t matter. The Court will decide.
[7] Can Facebook ban posts by Roman Catholics, or women, or Blacks? If Twitter is like a magazine rather than a common carrier, why not? From the 5th Circ. majority, at 2:
“Twitter unapologetically argues that it could turn around and ban all pro-LGBT speech for no other reason than its employees want to pick on members of that community, Oral Arg. at 22:39–22:52.”
[8] A better term than “network externalities” would be “network spillovers”. The confusion began because Alfred Marshall, the inventor of the supply and demand curve, discussed how a company setting up shop in a region could create benefits to nearby companies “external” to itself, “external economics”. Principles of Economics, 1st ed. (1890). Paul Bator invented both the term “externality” (Capital, Growth and Welfare: Essays in the Theory of Allocation, Ph.D. Thesis. Cambridge, MA: Massachusetts Institute of Technology (1956) and The Simple Analytics of Welfare Maximization, 47 American Economic Review 22-59 (1957)); and “market failure” (The Anatomy of Market Failure, 71 Q. J. of Econ. 351-379 (1958)), as explained in Steven Medema, Exceptional and Unimportant”? Externalities, Competitive Equilibrium, and the Myth of a Pigovian Tradition, 52 Hist. Pol. Econ. 135–170 (2020).
[9] The “Invisible Hand”:
He intends only his own gain, and he is in this, as in many other cases, led by an invisible hand to promote an end which was no part of his intention. Nor is it always the worse for the society that it was no part of it. By pursuing his own interest he frequently promotes that of the society more effectually than when he really intends to promote it.” Adam Smith, The Wealth of Nations, ch. 2 (1776).
On Providence, see Helen Joyce Adam, Smith and the Invisible Hand, g plus.maths.org (2007). For a more Straussian view, see Gavin Kennedy, The Hidden Adam Smith in His Alleged Theology, 33 J. History of Econ. Thought, 385-402 (2011).
[10] It might be instructive to look in detail at the arguments used historically to extend the concept of “common carrier” so far beyond its original application, and, in particular to look at the arguments against its extension in each particular case. Professor Candeub’s report is a good place to start.
[11] Cite to Viner here.
[12] In our example earlier, the marginal cost started at $2 million/plane and then rose to $4 million/plane at the output of 11 planes. On a graph that wouldn’t be Figure 1’s continuously rising diagonal line, but rather a step curve that would be flat at $2 million from 1 to 10 planes and then would jump to $4 million at 11 planes and stay flat thereafter.
[13] Add footnote here on Clayton Act, US Steel, etc.
[14] Cite to a case that says talking about price is illegal conspiracy.
[15] Eric Rasmusen and Todd Zenger, Diseconomies of Scale in Employment Contracts, 6 J. L., Econ., and Org. 65-92 (1992).
[16] Cite to the Microsoft antitrust case around 2000.
[17] Cite to a news story about the Twitter censor and the Bee.
[18] Whether Musk will lose money is unclear. Cite Charles Haywood or somebody on how Musk cut costs, but revenue fell, and how stocks generally fell after he made his offer.
[19] See this useful story. Internal Meta emails say pressure from Washington was behind a decision to take down posts attributing pandemic to man-made virus. See the writing of Greenwald. Zuckerberg complains that he was pressured to censor. See also, for shadowbanning, this story.
[20] See also, re the Masnick cite below: Mike Masnick, Ohio Files Bizarre and Nonsensical Lawsuit Against Google, Claiming It’s a Common Carrier; But What Does That Even Mean?, TECHDIRT (June 8, 2021), [https://perma.cc/Z3VX-ZMYE]
[21] 3 Blackstone at 164 says that a public innkeeper offers “an implied engagement to entertain all persons who travel that way; and upon this universal assumpsit an action on the case will lie against him for damages, if he without good reason refuses to admit a traveler.” Matthew Hale says that when someone builds the only wharf in a port, “the wharf and crane and other conveniences are affected with a public interest, and they cease to be juris privati only.” Matthew Hale, De Portibus Maris, in A Collection of Tracts Relative to the Law of England 77–78
[22] Justice Newton of the Court of Common Pleas ruled that a ferryman is “required to maintain the ferry and to operate it and repair it for the convenience of the common people.” Trespass on the Case in Regard to Certain Mills, YB 22 Hen. VI, fol. 14 (C.P. 1444).
[23] Note that if the stagecoach were not restricted by common carrier law, that might kill its business. To avoid sudden increases in the fare at the minute they chose to travel, customers might give up on stagecoaches and rent riding horses instead. As with being required to comply with contracts, businesses would want to be subject to legal penalties for tricking customers.