Three items this week — all of them, in one way or another, about instruments that outlived their justifications.
In the News
The Toll Booth That Replaced the Threat
Last week I noted that Trump’s serial extensions of the Strait of Hormuz deadline were doing something structurally distinct from simply moving the goalposts: each revision updated Iran’s beliefs about the cost of the threatened action, until the original ultimatum had been reclassified — retroactively — as an opening bid. This week the story advanced in a direction that deserves its own paragraph.
Iran is now charging passage. The Gulf Cooperation Council reported what Lloyd’s List Intelligence called a “de facto toll booth regime,” with at least two vessels having paid for safe transit through the Strait — in yuan. Iran is not blocking the strait unconditionally; it is extracting rents from it. Trump, meanwhile, told reporters on April 2 that U.S. forces would hit Iran “extremely hard” over the next two to three weeks. He did not mention the April 6 deadline he set ten days ago at Iran’s request.
The formal point here is separable from last week’s credibility argument. Last week was about what a sequence of missed deadlines reveals about a threatener’s preferences. This week is about what happens when the strategic game form changes entirely. Iran moving from unconditional closure to a toll regime is a shift from a coordination game with two absorbing states (open / blocked) to an extortion equilibrium with a continuous rent variable. That shift is itself informative: a player who had the option to maintain a complete blockade and chose instead to monetize partial access has revealed something about where its indifference curve runs. And the denomination in yuan suggests Iran is already positioning toward the longer-run equilibrium it expects to inhabit after this conflict resolves — which is a story for another day, but worth flagging now.
The April 6 deadline, established “as per Iranian Government request” on March 26, expires Monday at 8 p.m. Eastern. As of this writing, the administration’s public posture has replaced it with a declaration of sustained bombardment. A deadline creates a specific decision node with a defined consequence; a declaration of sustained bombing for “two to three weeks” does not. The formal instrument has dissolved.
Liberation Day at One
One year ago, Trump stood in a Rose Garden and declared April 2 the day America would “begin to make itself wealthy again.” The tariffs he announced — the highest since the 1930s — were struck down by the Supreme Court in February 2026, 6–3, on the ground that invoking the International Emergency Economic Powers Act to impose open-ended reciprocal tariffs exceeded presidential authority. The IEEPA emergency mechanism, it turns out, was not designed to serve as a general-purpose trade policy lever. Within hours of the ruling, Trump reimposed a 10% baseline tariff using Section 122 of the Trade Act of 1974, which authorizes a temporary tariff capped at 15% for up to 150 days in response to a large and serious trade deficit.
This is conservation of impossibility in operation. The policy goal — protection for domestic producers, tariff revenue, negotiating leverage over trading partners — did not evaporate when the Supreme Court invalidated the instrument. It migrated to the next available instrument, which happens to be weaker, more constrained in magnitude, and subject to a hard expiration date. The new tariff expires in roughly five months. At that point the administration faces a choice it was not supposed to face: let the tariff lapse, find yet another statutory hook, or seek congressional authorization it has consistently avoided asking for. Each option is constrained in ways IEEPA was not. The impossibility was not eliminated; it was relocated, and compressed.
The refund arithmetic is worth dwelling on. Tariff revenue for the first five months of the fiscal year totaled approximately $151 billion — roughly four times the same period the prior year. The government now owes approximately $166 billion in refunds to some 330,000 businesses for tariffs the Court has established were unconstitutional collections. The plan for disbursing those refunds is due in mid-April. The gap between what was collected and what is owed is not large; what’s remarkable is the sequence: a revenue instrument was deployed, generated substantial receipts, was ruled unconstitutional, and now requires a refund apparatus to be designed and executed under a new administration deadline. Tariff policy has, in the span of twelve months, managed to acquire the structural features of a commitment device problem — a credible threat that was carried out and then reversed — on both ends simultaneously.
The Federal Government Joins the Race to the Courthouse
On April 2, the Department of Justice sued Connecticut, Arizona, and Illinois, challenging their efforts to regulate prediction market operators — principally Kalshi and Polymarket — under state gambling law. The CFTC has maintained throughout the ongoing multi-front litigation that it holds exclusive jurisdiction over CFTC-registered prediction markets under the Commodity Exchange Act; the federal suit escalates that position from amicus briefs and agency guidance to direct litigation against the states themselves.
The formal structure of this move warrants attention. Kalshi’s strategy throughout has been what gaming attorney Daniel Wallach has called “winning the race to the courthouse”: preemptively suing states in federal court, asserting federal preemption before state enforcement can take effect. Courts have split approximately 13–2 in favor of states in recent injunction and TRO decisions, suggesting the strategy is not working as well as it once did. The federal government suing states directly replaces the regulated firm as the nominal plaintiff, which changes the constitutional valence of the preemption question. A federal court that might be skeptical of a private firm invoking federal preemption to escape state regulation faces a different question when the federal regulator is itself in the caption asserting exclusive jurisdiction.
There were two preliminary injunction hearings today — in Arizona and Nevada — the outcomes of which I don’t yet have as of this writing. The Arizona hearing involves not just Kalshi’s injunction request but also whether the federal court should abstain under the Younger doctrine given the ongoing criminal proceedings Arizona has brought against the platform. Arizona filed the first-ever criminal charges against a CFTC-registered prediction market in mid-March, a move that, as the Morgan Lewis writeup put it, transforms the litigation from civil enforcement into something with “qualitatively different coercive pressure.” We’ll have more on all of this next week.
In the Queue
Three Posts on Prediction Markets
Starting Monday, the blog will publish three posts on prediction markets on three consecutive days. Monday’s post takes up recent reports that federal prosecutors are examining whether certain pre-announcement trading in prediction markets — in which participants appear to have had access to nonpublic information before markets moved — constitutes a violation of existing securities and commodities law. The formal argument there is about signaling breakdown, not cheap talk: the conditions under which prediction markets aggregate information versus the conditions under which they transmit it from informed parties to uninformed ones. These are not the same thing, and the legal exposure depends entirely on which one is actually happening. Wednesday’s post is about jurisdictional recursion — the question of who has authority to determine whether Kalshi’s contracts are swaps (federal) or wagers (state), and why that classification decision is not separable from the very statutes it is supposed to interpret. Friday’s post goes deepest: the determination of whether something is a “swap” or a “bet” under the law is constitutive rather than descriptive — the act of naming does not follow from the thing’s properties, it in part determines them. These three posts connect, and will reward reading in sequence.
Whatever Happens at 8 p.m. Monday
The April 6 deadline arrives Monday evening. The question, formally, is no longer whether the deadline will be honored — the administration’s current posture has already replaced the deadline logic with a campaign-of-sustained-strikes logic — but what the transition from deadline-based threats to open-ended-campaign-statements means for the shape of any eventual negotiation. A deadline creates a decision node. A sustained campaign declaration makes the payoff structure continuous and eliminates the commitment-forcing properties of the original form. If talks resume, the parties will be starting from a different game than the one that was on the table in March. We’ll come back to this.
With that, I leave you with this.