Tag Archives: the-setter-model

Dispatches from the Underground: 3/20/26

Posted on by
Reply

A new recurring feature, starting this week. A few items from the news — sometimes updates to arguments this blog has already made, sometimes new events the existing framework illuminates without quite rising to a full post — followed by a couple of things we’re watching that haven’t ripened yet. Everything here is shorter than a regular post, but hopefully not shorter than it needs to be.

In the News

The DNI and the Negotiator Walk Into a Senate Hearing

This week, Steve Witkoff — President Trump’s envoy to the ongoing conflict with Iran — told reporters he had asked Vladimir Putin whether Russia was providing military support to Iran. Putin said no. Witkoff said he took that at face value.

In the same week, and in the same approximate direction, DNI Tulsi Gabbard testified before the Senate and said she does not take Putin at his word — and offered to elaborate in a classified setting.

This blog has argued that a principal who publicly announces strategic ignorance doesn’t thereby escape accountability: a meta-principal is watching outcomes regardless of what the principal claims to know. Witkoff’s move is that argument’s mirror image. He isn’t strategically ignorant of something he could learn — he’s strategically credulous about something his own government has formally assessed as false. The meta-principal (Congress, the intelligence community, the public) is observing. Gabbard’s very public non-endorsement of Witkoff’s position is itself bilateral accountability in action: the DNI just went on the record, which means Witkoff now has a named co-principal who disagrees with him in an unclassified forum.

The structure is the same. The game is slightly more embarrassing.

The Surveillance System Watching Itself

Speaker Johnson is pushing a clean 18-month extension of Section 702 surveillance authorities before the April 20 deadline, using White House backing to override members of his own caucus — notably the libertarian wing — who are demanding judicial warrants before the government reviews communications involving U.S. citizens.

Two distinct structural stories are running here in parallel. The first is the setter model: the decision about which version of the bill comes to the floor — the clean extension or the warrant-amended one — is itself a policy choice, made before any vote is cast. Much of the most consequential agenda-setting, as we’ve discussed, happens in the determination of what never gets considered. The second is something we’ve been calling the self-referential problem: Section 702 permits the surveillance of foreign persons, but the system that does so collects communications of U.S. persons incidentally. Whether Congress can meaningfully oversee this program partly depends on whether it can inspect a system whose operations it cannot fully observe — and which the program itself affects (the endogenous base rates problem, in a different register, from this post). The rules are watching the rules watching themselves. Johnson’s maneuver doesn’t resolve that problem; it just decides, for another eighteen months, that we won’t look directly at it.

Moving the Debt Doesn’t Move the Problem

The Trump administration announced a three-phase plan to move the federal student loan portfolio — along with FAFSA management — from the Department of Education to Treasury.

The student loan portfolio sits inside Education not because Treasury couldn’t technically service it, but because the loans are coupled to a regulatory ecosystem that Education also runs: borrower defense adjudications, income-driven repayment oversight, accreditation enforcement, institutional eligibility determinations. Moving the asset while leaving the regulatory apparatus behind doesn’t simplify the system. It separates functions that were coupled for reasons that are now invisible to the people doing the separating. The debt moves to Treasury; the disputes, the court orders, the appeals, the eligibility questions, and the unhappy phone calls stay rooted in rules that Education still administers. When the jurisdictional collisions start, and they will, the question of who owns the problem will have a distinctly less legible answer than it does today. See also: conservation of impossibility.

In the Queue

The Bondi Subpoena

House Oversight Chair James Comer issued a subpoena to Attorney General Pam Bondi this week — and then immediately told reporters he “personally doesn’t see any reason for a deposition,” suggesting that members who voted for it were “embarrassed.” The subpoena is, in other words, a signal that is publicly announcing its own insincerity.

This inverts the core logic of “Signaling through Obstruction”: a vote or a procedural move has informational content because it is costly, and that cost is what credibly separates genuine preference intensity from performance. A subpoena accompanied by its author’s own interpretation that it shouldn’t be taken seriously is cheap talk in a subpoena-shaped container. Bondi can — and almost certainly will — treat Comer’s public statements as the authoritative gloss on how to respond.1 We’re watching whether this evaporates, escalates despite Comer’s best efforts, or produces the kind of procedural theater that generates a lot of floor speeches and no depositions.

Is a Prediction Market That Causes Its Predictions Still Predicting?

Arizona this week became the first state to allege that Kalshi — a federally regulated prediction exchange — has committed criminal violations by operating an unlicensed gambling platform. The legal question is interesting, but the structural question underneath it is more so.

When a prediction market shows a candidate at 8% odds, it isn’t passively estimating probability. It’s shaping the behavior of donors, reporters, staffers, and the candidate herself — and it may be triggering the very withdrawal it was predicting.2 Honest and Effective covered the strategic-withdrawal-as-kingmaking problem; Can a Game Know Its Own Rules? covered the general structure of penalties that open games rather than close them. The combination here is: a market that causes the outcomes it forecasts has become something other than a measurement instrument. It’s an agenda-setter with a ticker. We’ll have more to say about this one.

1 For an earlier, blog-level treatment of the same dynamic — forced votes as constituent signals rather than genuine policy moves — see “Make Me an Offer I Can’t Refuse (to Reject).”

2 On classifiers that reshape the populations they classify, see The IRS Is Here to Help. So Is ICE., and the endogenous base rates discussion therein.

Also published today: a follow-up to All Statistics Are Local on the CPI, tariffs, and whose inflation the headline number is actually measuring — Your Basket May Vary.

Honest and Effective (Or, “Montana Has a Type”)

Posted on by
1

On March 5, at 4:52 p.m. Mountain Time, Kurt Alme filed to run for the United States Senate in Montana. At 4:55 p.m., incumbent Senator Steve Daines withdrew from the same race. At 5:00 p.m., the filing deadline closed. At 5:02 p.m., Daines endorsed Alme.

Jon Tester, Brian Schweitzer, Steve Bullock — none of them had time to file. The window that would have allowed any of them into the race lasted, in practice, approximately zero seconds. By the time anyone knew to run, the door was shut.

Daines later explained his reasoning without embarrassment: “The primary reason was there was no competitive Democrat in the race. Keep Tester out and Bullock out and Schweitzer out.”

This is agenda manipulation, executed with precision, announced after it was irreversible, and explained to the press the following morning. I want to talk about all three of those things, because the third one is actually the most interesting.1

The Setter Model

Political scientists have a name for the strategic position Daines occupied: the agenda setter. The setter doesn’t win by having the best alternative. The setter wins by controlling which alternatives are on the table when the choice is made. This insight goes back at least to Kenneth Shepsle’s work on institutional structure in the 1970s, and it is one of the most durable results in formal political theory: if you control the agenda, you often control the outcome, regardless of anyone’s preferences over the underlying options.

Daines didn’t change anyone’s preferences. Tester didn’t become less popular between 4:52 and 5:00 p.m. What changed was the feasible set — the universe of options available to Montana voters in November. Alme was in it. Tester wasn’t. And because filing deadlines are, by design, irreversible, the manipulation couldn’t be undone once it was discovered.

The textbook setter-model move has a particular structure: the setter proposes, everyone else responds, and the setter’s advantage comes from moving first. Daines ran a variation on this. He didn’t propose Alme — he cleared the field for Alme, which is subtly different. The move wasn’t “choose Alme over Tester.” It was “choose from {Alme} rather than {Alme, Tester, Bullock, Schweitzer}.” The reduction of the choice set is the proposal. (Ed: Twelve years of this blog and you’ve finally found the perfect example. You’re welcome, Montana.)

The Announcement

Here is what makes the Montana case formally distinctive, and worth thinking about carefully.

Classic agenda manipulation — the kind the Gibbard-Satterthwaite theorem formalizes — typically works because it is concealed. The manipulator pretends to be playing straight. The strategic move is hidden inside an apparently sincere action. A voter who ranks candidates A > C > B votes for A, knowing A will lose, to prevent B from winning — but they don’t advertise this.

Daines advertised it. The morning after the deadline, he told reporters exactly what he had done and why. This was not a confession extracted under pressure. It was a press briefing.

Why announce? Because the announcement came at the precise moment when knowing about it couldn’t help anyone. The feasible set was already closed. Transparency, in this case, costs nothing — because the information arrives after the last moment at which it could have been acted upon. Daines got to be honest and effective, which is an unusual combination in politics, and it’s only available when you’ve timed the manipulation correctly.

Bachrach and Baratz called this the second face of power: not the ability to win visible conflicts, but the ability to determine which conflicts become visible in the first place. The public announcement is not a contradiction of the strategic move. It is the strategic move, completing it.

Montana Does This

I should note, for readers who have been with this blog for a while, that Montana has a history of attracting this kind of formal attention.

In October 2014, I wrote three posts about a field experiment in which Stanford and Dartmouth researchers — Adam Bonica, Jonathan Rodden, and Kyle Dropp — sent mailers to 102,780 Montana voters bearing the official Great Seal of the State of Montana. The mailers placed candidates in nonpartisan Supreme Court races on an ideological scale anchored by Barack Obama and Mitt Romney. The Commissioner of Political Practices found this violated state law. Stanford apologized. Political scientists argued about research ethics for several weeks.

The mechanism was completely different — that was about information injection into a nonpartisan election, not agenda manipulation before a partisan one. But the underlying structure rhymes: someone used a procedural instrument (a state seal; a filing deadline) in a way its designers almost certainly did not envision as a strategic option, and the debate that followed was really about whether technically-legal-but-clearly-strategic behavior violates something that the rules cannot name. Montana, it seems, is where these questions go to get tested empirically.

Gibbard-Satterthwaite, Again

The Gibbard-Satterthwaite theorem tells us that any non-dictatorial voting rule over three or more alternatives is manipulable — there always exists some situation in which some agent can achieve a better outcome by acting strategically rather than sincerely. The theorem doesn’t tell you how manipulation happens. It just tells you it’s always possible.

What Daines did is Gibbard-Satterthwaite at the agenda-setting stage rather than the voting stage. He didn’t manipulate his own vote. He manipulated the choice set before anyone voted. This is, if anything, a more powerful form of manipulation than the theorem’s canonical examples, because it operates upstream of the election itself. By the time voters show up in November, the strategic work is long done.

Some Republicans were genuinely annoyed. Representative Eli Crane called it “messed up.” Roll Call moved Montana from Solid Republican to Likely Republican, calling the move “an unforced error.” Independent Seth Bodnar announced a candidacy and said Daines had “so little respect for Montana Republicans that he withdrew at the last minute to coronate his handpicked successor.”

All of this is correct. None of it is surprising. The Gibbard-Satterthwaite theorem doesn’t say manipulation is nice. It says manipulation is inevitable, given any sufficiently rich rule system and an agent with the right position and timing. Filing deadlines are, by design, a powerful institutional instrument — and powerful institutional instruments attract strategic agents. Daines had been chair of the National Republican Senatorial Committee. He knew the instrument well.

One final note: Democratic Representative Chuy Garcia of Illinois made essentially the same move last fall — withdrawing to clear a path for his chief of staff — and was censured by the House for it. The structural move was identical. The partisan valence was reversed. The formal theory doesn’t notice the difference.

So What Would You Do About It?

Readers who are not formal theorists — and I am told these exist — sometimes complain that political scientists are better at identifying problems than solving them. This is fair, so let me briefly sketch what a fix might look like, and why each fix has its own problem.

The most natural remedy is an automatic extension: if an incumbent withdraws within some window before the deadline, the deadline extends by an equivalent period, giving challengers time to respond. Clean, intuitive, directly targeted at the specific maneuver. It also creates a new strategic question — namely, how far outside that window can you time your withdrawal to still shape the field without triggering the extension? You’ve moved the manipulation point, not eliminated it. As a recent post argued — at some length — and another argued more briefly, the gap doesn’t close. It migrates.

A second option is a filing bond — candidates or withdrawing incumbents post a financial guarantee, forfeited if the withdrawal is found to have violated the spirit of the rules.2 This has some precedent in election law, and it does raise the cost of strategic manipulation. It also raises the cost of entirely sincere withdrawal, which is not nothing. And “found to have violated the spirit of the rules” is doing an enormous amount of work in that sentence — which is precisely the problem the previous posts were about.

The deeper point, which I will return to, is that filing deadlines are a blunt instrument designed to solve a coordination problem, not a strategic one. They were not built with Daines in mind. Retrofitting them to handle strategic withdrawal is possible, but every retrofit creates a new surface for strategic agents to work on. The manipulation migrates; it doesn’t disappear.

More on this — including what institutional designs have actually worked, and why the ones that work tend to look nothing like what naive reformers propose — another time.

And with that, I leave you with this.

1 March 5 was a busy day for filing deadlines. On the same day, Georgia State Senator Elena Parent — who represents the district where I live — announced her retirement one day before Georgia’s March 6 filing deadline, clearing the path for her endorsed successor, Representative Saira Draper. Same mechanism, smaller stage, closer to home. I have feelings about this, which I will keep to myself for now, as I have already written about procedural legitimacy this week and Maggie says my style doesn’t translate to cocktail parties.

2 The idea of posting a bond to deter strategic behavior in a civic context is not purely theoretical. City Schools of Decatur recently demanded that ten residents post a surety bond of $10-15 million — roughly half the cost of the building in dispute — as a condition of pursuing an appeal of a bond validation ruling. The residents, who described themselves as “parents, retirees, and taxpayers,” declined. I mention this only because the instrument travels, and because the week has had a theme.