What You Should Be Reading: June 2025
David B McGarry
July 3, 2025
Welcome to “What You Should Be Reading,” a monthly blog series from the Taxpayers Protection Alliance (TPA) for which our editors told us that we may not (under any circumstance) write about trade policy for the third straight month. (We protested, but HR sided against us.)
June’s edition includes the political economy of the Inflation Reduction Act, the dubious assumptions of competition policy technocrats, and misinformation about misinformation.
So, without further ado…
Competitive Enterprise Institute: “End IRA Subsidies Now”
One of the greatest fallacies in policymaking is this assertion: “If it doesn’t work, we’ll just repeal it.” Once a bill becomes law, it becomes entrenched, difficult to repeal — particularly if it provides some benefit to an industry or demographic able to threaten repeal-curious politicians with electoral defeat. As President Calvin Coolidge said, “It is much more important to kill bad bills than to pass good ones.” It is far easier to kill bad bills than to kill bad laws.
The Inflation Reduction Act (IRA) — passed in 2022 without a single Republican vote in the affirmative — is proving difficult in 2025 for a Republican-controlled Congress to repeal. Once an ill-conceived proposal, the IRA and its subsidization of green energy are now a fountain of taxpayer-sponsored jobs — many of which are in GOP districts. The political economy of the bill has wrought its intended effect, write Jacob Tomasulo and Ryan Young of the Competitive Enterprise Institute (CEI): “The IRA has projects in as many districts as possible for the same reason that defense contractors have operations in as many congressional districts as possible.”
TPA has noted the IRA’s perpetually growing price tag and economic distortions before. The CEI analysts note the fallacy of its job-creation scheme: “It is not about how many jobs there are, or what sector they are in, but the value they create.” An economy such as America’s in 2025, which does not lack for available jobs, requires the right workers to find the right jobs. It does not require the “creation” of job openings for the sake of having more job openings.
Tomasulo and Young continue:
IRA subsidies do not create net value. They take already-existing money out of some pockets and put it into others in a zero-sum game. They do not make people more productive. They do not create new wealth or new jobs. Instead, they redistribute money and jobs to people with better political connections, not necessarily people with better ideas. The labor force would be the same size even if the IRA had never passed.
Nonetheless, the constituents whose votes will determine whether incumbent Republicans return to Washington, D.C. are not likely to make such distinctions. Should the IRA be repealed, voters who lose employment on inefficient heretofore subsidized work will blame the politicians who caused those jobs to be lost. It is an eternal — not to mention grubby — feature of politics.
In 1774, in a row over his constituents’ taste for protectionist measures, the free trader Edmund Burke told the elders of Bristol that an elected representative owes his constituents close attention to their opinions and interests. “But his unbiassed opinion, his mature judgment, his enlightened conscience, he ought not to sacrifice to you, to any man, or to any set of men living,” Burke stated. “These he does not derive from your pleasure.” He continued: “Your representative owes you, not his industry only, but his judgment; and he betrays, instead of serving you, if he sacrifices it to your opinion” (emphasis added).
Burke’s courage cost him his seat. But if good policy is ever to prevail over rent-seeking, today’s Republicans must rediscover some of that virtue. Mature judgement and enlightened conscience are the surest guides to prudent lawmaking, despite the risks that attend their exercise.
The App Association: “Antitrust at a Crossroads: Protecting Innovation in the AI Era”
There is an irksome assumption held by those who favor a hypertrophic competition-policy regime. To wit, the notion that the rearrangement of markets by lawmakers and bureaucrats has little chance of threatening competition, innovation, dynamism, and prosperity. Private businesses (they will argue) stifle competition at every turn, but state interventions — no matter how empirically groundless — never will.
A new paper from The App Association examines exactly how recently competition policy proposals, statutes, and lawsuits threaten the development of artificial intelligence (AI). “The choice before policymakers is stark: adopt a risk-based, evidence-driven framework that targets proven harms, or freeze the engine that currently makes the United States the most fertile ground on earth for AI innovation,” the paper reads.
American lawmakers have plunged enthusiastically and recklessly into the work of regulating AI. In 2025, state legislatures considered more than 1,000 bills concerning the technology. American antitrust enforcement, too, threatens innovation. The paper states:
Ironically, the antitrust interventions aimed at AI markets tend to seek restrictions on AI markets’ most significant advantages: vertical integration, a healthy market for acquisitions, and dynamic marketplace management. Integration between chips, models, and apps often cuts costs and speeds learning. Acquisitions supply the liquidity and incentives founders need to keep betting on the next breakthrough. [Curated online marketplaces] are a means of distribution that can help connect developers, deploying AI, with their target customers via trusted channels. Antitrust interventions to eliminate these attributes would raise prices, undermine privacy and security, and push capital away from smaller, younger startups and toward more established firms hampering innovation.
In Europe, laws like the Digital Markets Act (DMA) have smothered the continent’s already gasping tech sector. “Framed as a tool to rein in dominant platforms and unlock competition, the DMA instead creates friction, fragmentation, and uncertainty, especially for the startups it was meant to empower,” the paper argues. “The DMA illustrates what happens when regulators outlaw conduct without a workable alternative and then struggle to show what ‘better’ looks like.”
As TPA has written previously, “Voters ought to hear the phrase, ‘It’s so European!’, as a dire warning — not an endorsement.”
Cato Institute: “The Misleading Panic over Misinformation”
Modern America tends towards the pathologization of everything. This observation is hardly original. Ordinary nerves are inflated into an anxiety disorder; the bout of the blues becomes depression; a 12-year-old boy’s excess of energy in English class secures him an ADHD diagnosis and the accompanying prescription — that sort of thing.
Something similar has occurred with respect to online speech. The terms of art in this case are misinformation, disinformation, and malinformation. Whatever their merit in the academic context, these terms have, in common use, come to refer to something very simple: being wrong. Human reason is ever fallible, and man’s motives, often suspect. They have, to crib from James Madison, diverse and opposing opinions, passions, and interests — yielding conflicting value judgments and conflicting accounts of fact.
Being wrong is an inexorable aspect of being human. The liberal means (classically liberal, that is) of combatting disagreeable speech involves more speech: if someone is wrong, argue against her. In the marketplace of ideas, arguments compete with one another to prove their worth, submitting the final determination of truth to a candid world.
The pathologization of being wrong furthers the notion that some opinions ought to be stifled — either by speech intermediaries, such as social media platforms, or by the state. Instead of viewing the public discourse as an unending exercise in epistemological discovery and refinement, some seek to deploy technology to excise the problem of dubious opinions and factual inaccuracy from public discourse.
A new paper from the Cato Institute’s David Inserra outlines the reasons anti-misinformation campaigns founder. Inserra notes that many disagree with respect to the definition. “Perhaps the most central challenge in defining misinformation and disinformation, though, is whether to cover out-of-context, disputed, incomplete, or otherwise misleading information that is not demonstrably false,” he argues. “The veracity of information might be thought to lie on a spectrum.”
He continues:
On one side, you have information that is provably true because of scientific experimentation, sufficient evidence, and valid, well-caveated logical arguments. On the other hand, you have provably false information because the evidence, data, and logic can show anyone why a statement is false. In between these two extremes, you have a lot of information that is not demonstrably false or true and about which people disagree. This information may present an incomplete set of data or facts as the truth, and without proper context. Information may be framed by narratives that present it in a certain light. Information may be partially right and partially wrong. Information may not always be provable or disprovable, either because the issue is still developing or it represents opinion. Some experts would include hyperpartisan news, rumors, clickbait, or even parody and satire as also potentially misleading misinformation in ways that run significantly counter to long-established norms of free expression.
The effort to transform ordinary human fallibility into a disorder to be eliminated by elite, top-down management of the marketplace of ideas should worry anyone one who values free speech. Any society should concern itself with the pursuit of truth, but censorship is a vehicle unlikely to carry us to that destination.
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.