What this covers
Michael Munger and Russ Roberts examine constitutions not as sacred founding documents but as practical instruments for organizing coercion through consent. Munger argues that what matters in constitutional order is not the elegance of the text but the presence of shared norms, exit options, and mechanisms that make even losing parties accept outcomes as legitimate. The conversation draws heavily on public choice theory—particularly Buchanan and Knut Wicksell—to argue that the only defensible justification for coercion is prior agreement to the rules, not to any particular outcome. Along the way, Munger traces how the American system worked not primarily because of constitutional prose but because of federalism, the frontier as an exit option, and Anthony de Jasay's insight that a ruler's power is constrained by subjects' ability to leave.
The discussion moves through several distinct territories. Munger walks through the paradox that the Constitution itself violated the Articles of Confederation—the delegates simply rewrote the rules of amendment rather than respecting the existing status quo—and explores why unwritten constitutions like England's work through interpretive tradition rather than founding text. He engages with Hayek's warning against "men of system" designing social orders from scratch, and with Adam Smith's account of human nature as biologically tuned to enforce norms, even arbitrary ones. The pair consider whether a new constitutional convention would solve current problems (Munger is skeptical, citing Chesterton's Fence and the rent-seeking contest such a gathering would trigger) and wrestle with the harder question: whether constitutions are largely epiphenomenal, mattering only as focal points if people already believe in them.
Munger argues that constitutions matter not as sacred lists of outcomes but as 'rules about rules' that make coercion legitimate through prior consent, and that the durability of a society's order depends more on widely shared norms and the possibility of exit than on the genius of any particular founding document.
- Following Buchanan and Wicksell, the only justification for coercion is consent, and at the level of collective action this means consenting to the rule, not the outcome
- The status quo has special standing (Relatively Absolute Absolutes), so the burden of proof falls on those who want to change the rules
- Constitutions are largely epiphenomenal: what made America work was Federalism, exit, and the frontier as much as the document itself
Constitutions succeed by establishing shared focal points and rule-sets that enable losers to accept outcomes as legitimate.
- By 1980 Buchanan abandoned the view that the U.S. Constitution caused American success, holding instead that it was important as a focal point but not causal; words on paper everyone has agreed to (like 'all men are created equal' or the Magna Carta) matter even when not true or enforced, because they can be used as a bludgeon to hold people to what they claimed to believe.
“Buchanan had abandoned that view for a much more nuanced view that the American Constitution was interesting; it was probably important as a focal point; but it was not causal.”
- De Jasay argued there are only two possibilities for a constitution: either everyone believes in its embodied principles, making it unnecessary, or few do, making it ineffective; thus the constitution is largely epiphenomenal—what matters is people's attitudes toward it, and if no one believes in it there will always be a workaround.
“There's actually only two things that can be true. One is that everybody believes in the principles that are embodied in the Constitution--in which case it is unnecessary--or, no one believes in them or not many people believe them--in which case it is ineffective.”
- England effectively has a constitution—proven by the fact that it has functioning politics—because a set of values is widely enough shared (rule of law over rule of persons, traced to Edward Coke's reading of Magna Carta) that even the losers accept outcomes as legitimate, which is exactly de Jasay's condition for a real constitution.
“if there are a set of values that are widely enough shared, now we're in a situation of politics, where even the losers accept it as being legitimate. So, that's the reason that England does have a constitution. I can prove it, because they have politics.”
- Munger defines politics as a process of deciding on rules such that even the losers accept them as legitimate; when even losers accept the outcome we can organize society, but when losers do not accept it as legitimate, politics breaks down into anarchy and revolution—so constitutions exist precisely to enable politics.
“Politics is a process of deciding on rules, such that even the losers accept them as legitimate.”
- Because unanimous consent cannot work at the level of specific outcomes (some people are infra-marginal and can free-ride), Buchanan's answer to Rousseau is that legitimacy comes not from consenting to a particular outcome but from having consented to the rule by which the outcome was reached—so a constitution is properly understood as a set of rules about rules.
“The question is whether you consented to the rule by which the outcome was arrived at.”
Consent justifies coercion only when people voluntarily subject themselves to agreed rules they can enforce against violators.
- Following Wicksell, the only justification for coercion is consent: a key to liberty is the ability to voluntarily subject yourself to coercion if you violate the terms of an agreement, because without a means of enforcing agreements we are not actually free to make binding commitments.
“what Wicksell had said--and it was interesting--was that the only justification for coercion was consent.”
- Collective goods like mosquito or flood control can be funded through conditional voluntary contracts—'I will only pay if everybody else pays'—enforced by a third party against non-contributors, which begins to look like taxation but is voluntary in the sense that all consented to be taxed on condition everyone else pays too.
“'I will only pay if everybody else pays. And, we will contract with some third-party enforcement agency that anyone who doesn't contribute this fee ... will be punished.' That's beginning to look a lot like taxes, but it's voluntary taxes”
- The 'love it or leave it' theory of tacit consent—being bound by a country's rules simply because you were born into it and don't leave—is not real consent, as Hume showed with the analogy of a man dumped asleep on a ship a hundred leagues out to sea who cannot meaningfully be said to have consented to the captain's rules by not jumping overboard.
“David Hume said: That's like saying you're taking on board a ship asleep. And, a hundred leagues out at sea, you wake up and you say, 'I don't want to be on this ship.' And, the captain said, 'Knock yourself out. Go ahead. Just jump over the side and be devoured by monsters of the sea.'”
- Even in a stateless arrangement, enforcement of consented-to agreements would be provided by a specialized third party rather than by the parties themselves, because the Adam Smith logic of division of labor means specialists in enforcement and violence will be cheaper and better at it—provided their action is justified by prior consent.
“division of labor means that they're going to be better at it. They'll have practice. They'll have tools and equipment. So, the usual Adam Smith story about division of labor.”
Even flawed institutions shaped by evolutionary selection processes outperform top-down design by men-of-system who wrongly believe they understand complex adaptive systems.
- Schmidtz's Desert Town shows that a known, fair, but arbitrary rule like a stoplight is better than getting out to argue who has the more urgent errand at each intersection: the stoplight caps your wait at two minutes whereas arguing makes the minimum wait seven, so we are all better off with an arbitrary rule we accept—provided everyone accepts it, since one defector running the light causes accidents.
“a stoplight means the longest I have to wait is two minutes. Getting out and arguing about all the rules means that the minimum I have to wait is seven minutes.”
- It is not only romantic young radicals who wrongly want to tear everything down and rebuild; Hayek argued in The Counter-Revolution of Science that social scientists become 'men of system' who design elaborate systems they want implemented in every detail, and these too are wrong because we do not understand institutions—which are complex adaptive systems shaped by an evolutionary selection process—very well.
“Hayek would say that many people who see themselves as social scientists come up with these elaborate--they are men of system. They are like Smith's men of system.”
- James Buchanan won the Nobel Prize in large part for his work on clubs: club goods are excludable but non-rival (a pool doesn't get crowded until many people use it), so collectively citizens can privately create public goods like tennis courts and pools through voluntary associations governed by small-c constitutions you consent to, can be expelled from, or can quit.
“James Buchanan actually won the Nobel Prize in large part for his work on clubs ... club goods are those that--in my neighborhood, there's four tennis courts and two pools.”
- A core premise of Public Choice is Alistair Cooke's 'homely but true' observation that politicians are very much like you and me—self-interested rather than angelic or smarter—so we should not assume their benevolence as the justification for the state.
“the homely but true observation that politicians, after all, were very much like you and me. That is, that they're self-interested and that we should not assume that they're angelic or they're smarter.”
Constitutional conventions risk catastrophic failure because consensus is difficult, outcomes become rent-seeking contests, and new rules are invented by ignoring existing ones.
- A new U.S. Constitutional Convention is a terrible idea for three reasons: the current Constitution is not so terrible that the better alternatives clearly outweigh the risk; reaching a consensus takes a long time and a convention would, like the original, simply ignore existing rules as binding; and putting everything up for grabs creates a rent-seeking contest in which powerful interests exert back-channel influence.
“That's a terrible idea. There's a lot of terrible ideas. That's one of the terrible-est ideas.”
- Buchanan shifts the burden of justification from the level of rules to the change in rules: the status quo is the one outcome that requires no one to vote for it to be binding, so what requires consent is altering the Relatively Absolute Absolutes, even though some inherited rules (racism, patriarchy) are indefensible—making this the weakest part of his theory.
“Instead of saying that we have to justify the level, he says we have to justify the change in the rules. So it takes our consent to justify a change in the rules because the status quo has a special status.”
- The U.S. Constitution was literally unconstitutional: the Articles of Confederation required 13 of 13 states (Wicksellian unanimous consent) to amend, yet the Philadelphia delegates simply decided to ratify by a nine-of-13 rule, ignoring the existing status quo rules—and a new convention would do the same, inventing its own rules rather than respecting the current ones.
“The Constitution is literally unconstitutional because it was ratified using a nine-out-of-13 rule when the existing Constitution--the Articles of Confederation--required 13 out of 13.”
Constitutions can meaningfully constrain process and powerful interests despite losing substantive regulatory reach, and written rules remain superior to precedent-alone systems.
- Chile is currently rewriting its 1980 constitution—imposed at bayonet point by a non-democratic regime and since reformed—and the objection to it is not that it doesn't work but that it does work, with reformers believing that if they start over they can do better.
“They're not objecting to it because it doesn't work. They're objecting to it because it does. And, they think that if they start over, they can do better.”
- De Jasay's central concept is exit: one's ability to coerce another is sharply limited by the other's viable ability to exit, so American Federalism (competition among states you could escape to) and the American frontier were more important than the Constitution itself—and the combination of frontier and Constitution is what made America great.
“his central concept is exit. If you can exit ... my ability to coerce you is going to be limited sharply by your viable ability to exit.”
- Countries without written constitutions, like England and Israel, substitute an expanded interpretive tradition—analogous to Torah and Midrash, or England's common law—where analogical reasoning over a body of precedent settles cases, ideally so well that the courts only hear the genuinely unsettled cases, unlike Louis XIV's tree-side rulings that carried no precedential value.
“Countries that don't have constitutions tend to have an expanded interpretive tradition ... It's basically Torah and Midrash.”
- Even if most day-to-day legislation now evades constitutional constraint—so that only the First and Second Amendments still bite—the Constitution can be effectively binding at rare but crucial moments of process, as when Mike Pence concluded he could not unilaterally declare Trump the winner despite pressure, demonstrating that rules about rules constrain process even when substance is unconstrained.
“I think they don't always, because Mike Pence wanted to talk his way out of it and he didn't. He said, 'You know, I can't do that. I cannot declare Donald Trump the President.'”
- The U.S. Constitution succeeded partly by luck: the Articles of Confederation were so bad that they created an enormous 'win-set'—many possible arrangements were better—so the framers could pick one of many superior contingent points, meaning the Constitution's content was contingent rather than the product of unique godlike genius.
“the Articles of Confederation were so bad that they created a large--what political scientists call a win-set. So, there's many things that were better than the Articles of Confederation”
Rule-enforcement and small-norm violations matter fundamentally because civilization's veneer is thin and defectors cascade social breakdown.
- The foundation of social order rests on enforcing even the simplest rules, like jaywalking, in the spirit of broken-windows policing—because the veneer of civilization is far thinner than people assume, and tolerating small violations puts a society 'just inches away from theft' and broader breakdown.
“It's like the broken windows in New York. If you are going to jaywalk, we're just inches away from theft.”
- Drawing on Adam Smith's sense of propriety, humans have a 'constitution' too: we are biologically constituted as rule-following punishers who, on seeing a norm violation, are suffused with a non-rational chemical cocktail driving us to attack the violator—though the specific content of the enforced norms (jaywalking, or in 1940 a Black person entering a bar) is socially constructed.
“What happened was her body was suffused with a cocktail of chemicals that said, 'Attack. He's violating the rules.' And so, she acted on that irrational impulse.”