What You Should Be Reading: July 2025

Welcome back to “What You Should Be Reading,” a monthly blog series in which the Taxpayers Protection Alliance distracts itself from the hottest and steamiest summer months by diving into the cool and refreshing waters of the best recent works of public-policy research.

July’s edition includes the censorship of the European Union’s Digital Services Act, an obscure — though not unimportant — threat to the First Amendment, and the facts about the American semiconductor trade.

So, without further ado…

House Judiciary Committee: The Foreign Censorship Threat: How the European Union’s Digital Services Act Compels Global Censorship and Infringes on American Free Speech”

Technology has wrought miracles. As one author observed more than a half century ago, “Any sufficiently advanced technology is indistinguishable from magic.” In the 19th century, the telegraph and the railroad overcame difficulties of time and space, contracting the extended American republic. Today, digital technologies’ victory over the distances once imposed by time and space are nearly total.

For all its power, though, technology cannot rid the world of disagreement and, in many circumstances, falsehood. Bad information is endemic to the human condition and to human interaction in two senses. First, perception and reason are fallible, even if unclouded by interest or passion. Never yet was the person born whose opinions were not in some way faulty. Mistakes and inaccuracies will, therefore, be committed — online and offline, both.

Second, those who set themselves up as censors are — like everybody else — imperfect. the definitions of disinformation and misinformation — let alone malinformation — often extend to information that cannot, in any objective sense, be considered false. The theory of COVID-19’s likely origin in a lab in Wuhan, China migrated from a conspiracy to likelier-than-not in the course of a few months; yet in the interim, analysts who propounded and offered evidence for this theory were maligned as quacks. Put differently, charges of disinformation are often leveled by the censorious in order to skirt the critical truth-seeking process of debating contentious and uncertain issues of national import.

In 1644’s Areopagitica, John Milton questioned the notion that censors will light upon the truth. “It cannot be denied but that he who is made judge to sit upon the birth or death of books, whether they may be wafted into this world or not, had need to be a man above the common measure, both studious, learned, and judicious,” Milton wrote of England’s book-licensing code; “there may be else no mean mistakes in the censure of what is passable or not; which is also no mean injury.” Milton’s warning holds good today. When authorities undertake to declare some opinion true and others false, and so to dispense with further argument, they will invariably err.

Despite all this, many seek to render the internet free of inaccuracy and (subjectively) objectionable speech. The European Union (EU)’s Digital Services Act (DSA) leads this effort. In a recent interim staff report, the House of Representatives Judiciary Committee documents the DSA’s efforts to censor online information of which the European officials disapprove.

The mechanisms of the DSA’s censorship include:

• “Article 21 mandates that platforms allow certified third-party arbitrators to resolve content moderation disputes. These arbitrators must be independent from the platforms, but do not need to be independent from the European regulators who certify them, incentivizing arbitrators to heed regulators’ censorship demands.”
• “Similarly, DSA Article 22 requires that platforms give priority to censorship requests from government-approved third parties known as ‘trusted flaggers.’ In practice, these trusted flaggers are uniformly pro-censorship, and in many cases, they are government-funded, meaning that these so-called ‘trusted’ flaggers are incentivized to censor speech critical of politicians or the current regime.”
• “The core of the DSA is the risk assessment and mitigation framework set out in Articles 34 and 35. These provisions encourage platforms to censor a wide variety of speech. Tech companies are directed to identify ‘systemic risks’ present on their platforms, which are defined to include ‘misleading or deceptive content,’ ‘disinformation,’ ‘any actual or foreseeable negative effects on civil discourse and electoral processes,’ and ‘hate speech.’ Platforms are specifically warned that this systemic risk may include ‘information which is not illegal.’ Then, under the DSA, platforms must mitigate these risks, meaning they ultimately must remove content that European regulators deem ‘misleading,’ ‘deceptive,’ or ‘hate[ful].’”

The DSA does not merely interpose between users and the kinds of speech that the First Amendment does not protect. Quite the contrary: “Documents produced to the Committee under subpoena show that European censors at the Commission and member state levels target core political speech that is neither harmful nor illegal, attempting to stifle debate on topics such as immigration and the environment,” Judiciary relates. One post flagged by Polish authorities stated that “electric cars are neither ecological nor an economical solution.”

Efforts to rid the internet of officially disfavored speech will end not only in violations of users’ right to free speech but in a shrunken, impoverished internet, bereft of truth. Classical sources liken truth to light; under the regime of the DSA — and other laws like it — the online world has become a far darker place.

Read the full report here.

TechFreedom: Comments in the Matter of Omnicom/IPG

The Federal Trade Commission (FTC) objects to the work of advertising agencies that have declined to place their clients’ advertisements alongside content they deem objectionable. The First Amendment objects to the FTC’s attempt to control what sort of speech private businesses choose to associate with.

The agency characterizes the consent order in its scrutiny of the acquisition The Interpublic Group of Companies by Omnicom Group Inc. as a matter of competition enforcement. However, according to comments submitted by TechFreedom’s Berin Szóka, the case is better understood as a matter of free speech.

The FTC’s settlement includes a “behavioral restriction [that] prohibits the companies from ‘steer[ing] advertising dollars away from publishers based on their political or ideological viewpoints’ — a rule that turns entirely on the content of the publisher’s speech,” Szóka argues. “The Commission may believe it is targeting anticompetitive collusion, but it risks chilling legitimate, and protected, editorial judgments by private actors.”

Szóka further warns against the regulation of speech by “soft law” — through settlements, whose constitutionality and lawfulness cannot be reviewed by judges. “This ‘soft law’ approach was dangerous and rightly criticized by those concerned about FTC,” he writes. “No matter how many cases the Commission brings, if the Commission settles them all, such a common law ‘is really just a series of unadjudicated assertions.’”

Threats to liberty often arise slowly, creepingly, and quietly. Obscure cases advanced by administrative agencies rarely appear on the front pages of newspapers. That such efforts are often obscure, however, does not mean that they are unimportant.

Read the full comments here.

American Enterprise Institute: “How Does Semiconductor Trade Work?”

“America doesn’t make anything anymore.” This claim is ubiquitous among protectionists. It is, however, patently false. Nonetheless, the notion that free trade has “hollowed out” the American industrial base persists among those who wish for towering trade barriers and Washington, D.C.–managed international markets.

One such misunderstood industry is the semiconductor-manufacturing industry. The simple protectionist proposal — to tariff semiconductor inputs — overlooks the realities of the sector and threaten to stifle the production of American firms. The American Enterprise Institute’s Chris Miller correct these errors in a recent report.

The report has many points of interest, but consider this: Miller notes that “many of the chips that the US is importing are actually chips that are manufactured in the US and then packaged abroad before being imported back into the US.” The critical public-policy implication: “Imposing tariffs on these chips would penalize manufacturers who are already conducting their high-value manufacturing steps in the United States.”

Moreover, Miller writes, “because the US has hardly any chip packaging capacity, we know that the chips imported into the US are almost all converted into final goods in the US. In other words, they are assembled into cars, medical devices, aerospace equipment, and other products. Increasing tariffs on imported chips would make manufacturing in these segments more expensive and less competitive.”

Simplistic conceptions of the American economy rush blindly into simplistic policies, which in turn generate tremendous unforeseen carnage. The errors, if left uncorrected, damage the prospects of the very industries trade barriers purport to protect.

Read the full piece here.

Note: TPA highlights research projects that contribute meaningfully to important public-policy discussions. TPA does not necessarily endorse the policy recommendations the featured authors make.