Four items from the news this week, plus two more on the horizon. It was that kind of week.
In the News
The Deadline That Wasn’t, Then Wasn’t Again
Trump’s ultimatum to Iran — open the Strait of Hormuz or face renewed strikes on energy infrastructure — was due March 23. Then it was due today. Now it’s due April 6.
The structural problem with a deadline that gets revised is not just that it’s embarrassing. It’s that each revision updates the counterparty’s beliefs about the cost of the threatened action. The first deadline established a threshold. The first extension revealed that, at that threshold, the cost of following through exceeded the benefit — otherwise the deadline would have been kept. The second extension reveals the same thing about the revised threshold. Iran has now received, in sequence, three observations about the shape of Trump’s indifference curve between war and non-compliance. The fourth announcement will carry less information than the first because its informational content is now partly absorbed by the prior two.
Thomas Schelling’s insight was that the value of a commitment device comes precisely from its unconditional character: a threat that you might not carry out isn’t really a threat, it’s an opening bid. Two extensions don’t move the deadline. They reclassify the original statement as an opening bid, retroactively. The credibility of any subsequent announcement is not restored by making the announcement more emphatic. You can’t rebrand an opening bid as a commitment by raising your voice.
The Vote Not Taken
Reporter Ryan Grim of Drop Site News wrote this week that House Foreign Affairs Ranking Member Gregory Meeks has been delaying a vote on an Iran war powers resolution not because the votes aren’t there, but because they are. Grim’s framing — “Dems secretly want this war to continue because it hurts Trump” — is a motive story. The structural story is more useful, and doesn’t require that interpretation to be true.
Bachrach and Baratz’s second face of power is the power to determine what gets considered — not just the power to win votes, but the power to prevent votes from being held. The setter model, as this blog has discussed, is the formal version of that insight: when the agenda-setter controls what arrives on the floor, the most consequential choices are made upstream of any vote. What Meeks is doing — if Grim’s account is accurate — is a second-face move: keeping the status quo in place by denying the floor a chance to revise it. The procedural void he’s creating produces a kind of collective ambiguity: progressive Democrats can say the votes are there, leadership can say nothing has been foreclosed, and every party reads the non-vote as confirmation of its own position.
The blank space is the product. That’s what the second face of power looks like from the outside.
Five Minutes Before
Senator Chris Murphy went on the record this week about $1.5 billion in S&P 500 futures sold minutes before Trump announced a five-day pause on striking Iran’s power plants. A similar pattern preceded an earlier Trump social media post about “productive” talks with Iran. An unknown trader reportedly made $1 million from prediction-market bets about Iran’s military situation timed to the same events.
Murphy’s framing is legal — corruption, insider trading — and that framing may be correct. But there’s a distinct structural point worth separating out. In the standard model of price discovery, announcements move markets because they carry information that wasn’t already priced in. When the market moves before the announcement, it means the information has already diffused through some other channel. The announcement, when it comes, is then not transmitting information — it’s ratifying information that has already been incorporated. Whether that prior diffusion was a leak to connected traders, or simply that sophisticated observers were reading signals available to anyone, the result is the same: the presidential announcement is the last thing to know.
Crawford and Sobel showed that cheap talk can carry genuine information when sender and receiver have partially aligned interests. But a message sender whose private behavior — trading, in this case — is observable before the message arrives has effectively sent the message twice: once through the trade, once through the announcement. The second transmission doesn’t add information. It adds deniability.
ICE, TSA, and the Disaggregated Bill
The Senate passed a bill this week funding most of the Department of Homeland Security — and specifically excluding ICE. The bill passed. ICE funding remains a live dispute.
What’s worth noting is what the disaggregation did to the underlying game. When ICE and TSA were bundled in the same funding fight, Republicans held a specific piece of leverage: the airport chaos caused by unpaid TSA workers was visible, immediate, and politically expensive in a way that was hard to disentangle from the ICE funding standoff.1 The shutdown story and the ICE-at-airports story were running simultaneously — one because TSA workers weren’t being paid, the other because ICE was being deployed to fill the gap. Bundling made it structurally difficult for anyone to resolve one without resolving the other.
The disaggregated bill removes that linkage. TSA chaos is resolved; ICE funding now has to succeed or fail on its own terms, without the airport-chaos lever attached. Whether that’s a win for Democrats (who got TSA funded without conceding on ICE) or for Republicans (who can now fight about ICE in a setting with lower background noise) depends on whose preferred issues benefit from elevated salience. The setter’s choice to bring the disaggregated bill changed which game everyone was playing, mid-game.
In the Queue
Montana Has a New Variable
Seth Bodnar — former president of the University of Montana — entered the state’s Senate race this week as an independent, following Republican Steve Daines’s retirement announcement. Democrats are worried. They should be, but not for the reason usually given.
The standard spoiler framing treats third-party candidates as draining votes from whichever major-party candidate they’re closest to. That’s the right intuition, but it’s not actually a spoiler problem — it’s a coordination problem. Voters who collectively prefer any non-Republican outcome need to coordinate on a single non-Republican candidate. When they have heterogeneous beliefs about which candidate is viable, they can’t. The equilibrium fragments: some voters go with the Democrat (established infrastructure, party label), others go with Bodnar (independent appeal, crossover credibility), and the Republican wins with a plurality despite a non-Republican majority in the electorate. The problem isn’t that Bodnar is splitting the vote. The problem is that the plurality system doesn’t give non-Republican voters a mechanism to reveal and aggregate their second-choice preferences — which is precisely what Montana’s Senate dynamics this cycle have been about from the start.
The Fed’s Impossible Mandate
Oil is hovering near $120 a barrel. The Federal Reserve’s dual mandate — stable prices and maximum employment — was designed for a world where inflationary pressure and weak employment tend to move together. A negative supply shock decouples them: it pushes prices up and growth down simultaneously, which means the Fed’s two targets pull the policy rate in opposite directions. Raise rates to combat inflation; risk tipping a cooling labor market into contraction. Hold rates to protect employment; ratify inflation already running hot.
One instrument, two binding constraints. This is conservation of impossibility in its most mechanically literal form — not a metaphor, but a direct application. The Fed cannot satisfy both objectives simultaneously when the shocks driving them are supply-side. What it can do is choose which constraint it treats as binding — and that choice, which the institution’s formal language will describe as a technical judgment about “appropriate monetary policy,” is a political choice about who absorbs the cost. We’re watching which way that comes down.
1 For the full treatment of ICE at airports — and the chilling equilibrium that precedes enforcement deployments — see Stressing at the Airport? Just chill…ICE is on the way!
1 thought on “Dispatches from the Underground: March 27, 2026”
Comments are closed.