Why Markets Need Controls
On the constitution of the marketplace, the alternative to it, and why a regulated market is a republic by another name
Keywords: markets; regulation; republic; externalities; monopoly; regulatory capture; central planning; constitutional rules
Here is a claim that will annoy nearly everyone, which is usually a sign it is worth making: a regulated market and a constitutional republic are the same invention, discovered twice, in two different centuries, by people who mostly never spoke to each other, in response to the identical problem. The problem is this. There exists a certain kind of machine — a decentralised aggregator that takes the scattered choices of millions of people and condenses them into a signal that coordinates the whole — and it is the most powerful social technology ever built, and it has a fatal flaw: left entirely to itself, it destroys the very thing that made it valuable. The market is one instance of this machine. Democracy is the other. And the solution, in both cases, turns out to be the same: a layer of fixed rules that the momentary crowd cannot override, placed there precisely so that the aggregator can keep working instead of eating itself. Regulation is to the market what the constitution is to the democracy. Controls are not the opposite of freedom. They are what makes the freedom last.
I have argued, in two prior pieces, what each machine does — that a price aggregates dispersed knowledge and then, in a bandwagon, aggregates only the crowd; that a vote transfers power peacefully and then, in a mob, aggregates only the passion of the moment. Set those two accounts side by side and the parallel is not loose or poetic. It is exact, joint by joint, and once you see it you cannot un-see it, and it settles the tired old argument about whether markets should be free or controlled by showing that the argument was malformed from the start.
The parallel, joint by joint
Take the two machines apart and lay the pieces next to each other.
The aggregator itself: in the market it is the price; in the polity it is the vote. Each takes what is dispersed — knowledge in one case, preference in the other — and condenses it into a signal that lets strangers act together without a commander. Each performs a miracle no central authority could: the price coordinates people on knowledge no single mind holds; the vote transfers power without anyone having to die for the transfer.
The runaway failure: in the market it is the bubble and the panic; in the polity it is the mob and the demagogue. In both, the aggregator’s feedback flips from negative to positive — rising prices drawing more buying, a swelling majority drawing more zeal — and the signal detaches from the world and locks onto the crowd’s opinion of itself, and the marble that was resting in a bowl finds itself on a dome.
The concentration failure: in the market it is monopoly; in the polity it is faction capture, the seizure of the whole apparatus by one party or interest. In both, the winner uses its winnings to prevent the next contest — the monopolist buys or crushes the rival that would discipline his price, the faction rewrites the rules so it can never be voted out — and in both the concentration kills the competition that made the aggregator honest.
The poisoned signal: in the market it is fraud; in the polity it is ballot-stuffing. Both are attacks on the integrity of the count itself — a lie about the quality of the good, a lie about the tally of the votes — and both, unchecked, destroy the one thing the aggregator must have to function at all, which is that the signal reflect something real.
And the fix, in both cases, is a fixed layer: in the market, regulation — the rules of the game; in the polity, the constitution — the rights beyond the vote. A set of commitments held stable, put beyond the reach of any single transaction or any single election, so that the aggregator is protected from its own tendency to concentrate, poison, and stampede. Leave the market unconstrained and it collapses into an oligarchy of winners. Leave the democracy unconstrained and it collapses into the tyranny of the majority. Same machine. Same failure. Same cure.
Why the market cannot fix itself from the inside
The market-lover’s instinct at this point is to say: but the market is self-correcting; bubbles pop, frauds are exposed, monopolies are eventually undercut, and so the fixed layer is unnecessary, or worse, an intrusion. This instinct is half right, and the half where it is wrong is the half that matters.
The market corrects its belief errors well. A bubble is a mistaken belief about value, and reality eventually arrives to contradict it, and the price falls. That kind of self-correction is real, and it is why one should never replace the market wholesale. But there is a whole family of failures the market cannot correct from inside, not because it is malfunctioning but because it is functioning exactly as designed, and a machine working as designed does not self-correct — there is nothing for the correction to grip.
Consider fraud. A market runs on information; the price is only as good as the truth of the claims behind it. But a lie about quality can be more profitable than quality itself, and so, absent a rule against it, fraud does not get competed away — it gets rewarded, until trust collapses and no one will trade with anyone, at which point you do not have a freer market, you have a dead one. Anti-fraud law is not an imposition on the market. It is a precondition of it, exactly as the rule against ballot-stuffing is a precondition of an election rather than a limit on it. Consider monopoly: the market’s own logic, followed to its end, tends toward the elimination of competition, because crushing your rival is more profitable than out-serving him, and the monopoly that results kills the price signal that made the market worth having. The market does not correct this; it produces it, and only a rule from outside — antitrust — can make the market preserve the competition it would otherwise devour. Consider the externality, the sharpest case: the factory that dumps its filth in the river prices what it sells but not what it destroys, because the poisoned town was never a party to the sale. The price is honest, efficient, and systematically wrong, and it will stay wrong forever, because the market is built to ignore exactly the cost that is landing on the people downstream. No self-correction can fix this, because nothing is broken. The machine is pricing what it trades and ignoring what it dumps, which is precisely its design. Only a rule — a tax, a cap, a liability — can put the missing cost back into the number.
These failures do not self-correct because they are not errors in the ordinary sense. They are the machine doing its job, and its job, unconstrained, includes lying profitably, concentrating relentlessly, and ignoring everyone not at the table. The fixed layer is not there to override the market’s judgement. It is there to repair the specific places where the market has no judgement to offer — where, left alone, it points confidently in the wrong direction.
What the fixed layer must be: a constitution, not a plan
But here the market-lover’s fear becomes legitimate and important, and I want to honour it rather than wave it away, because there is a right kind of control and a disastrous kind, and confusing them has ruined economies as thoroughly as no controls at all.
The distinction is exactly the one that separates a constitution from a dictatorship. A good political constitution fixes the rules of the game — who may vote, how power passes, what no majority may do — and then it leaves the actual outcomes of the game entirely open. It does not dictate who wins the election; it guarantees that there will be a fair election whose winner is not predetermined. A constitution that fixed the winners would not be a constitution; it would be a fraud wearing one’s clothes. The same distinction governs the market. Good regulation fixes the rules of the game — no fraud, no monopoly, disclose what you sell, bear the cost of what you dump — and then leaves the outcomes of the game open. It does not set the prices, allocate the goods, or decide who succeeds. It guarantees a fair contest whose result is not predetermined, and then it gets out of the way of the result.
The catastrophic kind of control is the kind that reaches past the rules and seizes the outcomes — that sets the prices, commands the production, allocates the goods. This is not a constitution for the market; it is the abolition of the market, and it fails for the reason every attempt at central planning fails: it destroys the price signal, and once the signal is gone the planner is blind. Without prices, there is no way to compute what should be made, in what quantity, from which inputs, for whom — the knowledge that the price silently aggregated simply vanishes, and no committee, however well-meaning, however well-computed, can reassemble it, because most of it was never written down and never could be. The planner does not know less than the market by a little. He knows less than the market by an amount that cannot be closed, ever, because the knowledge he needs does not exist in any collectible form. And so central planning delivers, with grim reliability, shortage where there should be plenty and waste where there should be thrift, and then — because the failure cannot be admitted — coercion, to force the blind plan onto a reality that keeps contradicting it.
So there are three options, and only one of them works, and the same three appear in the market and in the state. Leave the aggregator wholly unconstrained — laissez-faire, or pure direct democracy — and it eats itself: the market concentrates into an oligarchy of winners, the democracy stampedes into the tyranny of the majority. Abolish the aggregator and replace it with command — central planning, or rule by an unremovable elite — and you go blind: the planner loses the price, the philosopher-king loses the vote, and both lose the very knowledge and legitimacy the aggregator existed to produce. Only the narrow, unglamorous middle endures: constrain the aggregator with fixed rules that govern the game and leave the outcomes free. A constitution for the marketplace. A republic for the state. The same discovery, made twice.
The steelman: controls get captured, and become the disease
Now I owe the strongest case against controls, because it is strong, and a defence of regulation that has not felt its force is propaganda.
The case is this: controls do not stay pointed at the failures they were built to fix. They get captured. The regulator, over time, is captured by the very industry he regulates — because the industry has the money, the lobbyists, the expertise, and the patience, and the public has none of these and is not paying attention — and the rules, once captured, are turned inside out: the anti-monopoly regime becomes a moat that protects the incumbent monopolist from competition, the licensing rule that was meant to ensure quality becomes a barrier that keeps out anyone who might challenge the established firms, the safety regulation becomes a compliance cost that only the giants can afford and that therefore strangles every upstart in the cradle. The control that was supposed to preserve competition becomes the incumbent’s chief weapon against it. And there is worse: the regulator suffers from exactly the knowledge problem that afflicts the planner — he knows less than the market he is regulating, his rules ossify while the world moves, and his interventions misfire in ways he cannot foresee and will not admit. And regulators, being human, serve themselves and their future employers in the industry, not the abstract public they are nominally protecting. So controls ratchet only ever upward, never getting simpler, each new rule laid atop the last, until the fixed layer is not a lean constitution but a suffocating thicket that protects the powerful, strangles the new, and serves chiefly those clever enough to game it. On this account, the cure is worse than the disease, and the honest response to market failure is not more controls but fewer, and a grim patience with the failures the market has not yet corrected.
This is a serious argument, and large parts of it are simply true. Regulatory capture is real and pervasive. Rules do ossify, misfire, and metastasise. The regulator does often know less than the market and serve someone other than the public. Any defender of controls who denies these things is lying or has not looked.
The crack: the answer to bad controls is good controls, not none
But watch, once more, what the argument establishes and what it does not. It establishes that controls can fail toward a specific tyranny — capture, ossification, the strangling thicket that serves the incumbent. It does not establish that you should therefore have no controls, because no controls returns you directly to the other set of failures — fraud, monopoly, externality, panic — which are not hypothetical either. You have not escaped the problem by abolishing the fixed layer. You have merely chosen which failure to suffer, and traded the captured regulator for the unconstrained monopolist, who is not an improvement.
The answer is the same answer the republic gives to the dead-hand objection, and it is the whole point of the comparison. A constitution can be captured too — by entrenched interests, by unaccountable judges, by the clever lawyers of the powerful — and the response is not to abolish the constitution but to design it better: keep it lean, keep it aimed at the rules of the game rather than the outcomes, keep it hard to change but not impossible, and keep it a hinge rather than a weld. The market’s constitution demands exactly the same discipline. Make it minimal — a few structural rules against the failures that genuinely do not self-correct — fraud, monopoly, externality, systemic panic — and nothing more. Make it structural — aimed at preserving the mechanism, not at dictating the results. And make it revisable — hard enough to change that it is not rewritten at every lobbyist’s whim, easy enough that it can be pruned when it ossifies or is captured. The reply to “controls get captured” is not “have no controls,” which gives you monopoly and fraud, nor “have total control,” which gives you the plan and the blindness. It is “have a good constitution for the market” — few, fixed, structural, revisable — for exactly the same reason, and by exactly the same craft, that a nation needs a good constitution rather than either a mob or a king.
The teeth: freedom without rules is not freedom, and rules without freedom is not order
Two errors bracket this subject, and by now their shape is familiar, because it is the same shape that has closed every one of these essays.
The libertarian error is that controls are tyranny, markets are self-regulating, and freedom means the absence of rules. But a market with no rules is not free; it concentrates. The strong buy or crush the weak, the liar out-earns the honest man until trust dies, the monopolist inherits the earth, and the pollution lands on people who never agreed to breathe it. What you get from the absence of rules is not the freedom of everyone but the freedom of the winners to end everyone else’s — the oligarchy of the victors, which is simply the market’s version of the mob: the unconstrained aggregator devouring the competition that gave it worth. The anti-fraud rule is to the market what the anti-ballot-stuffing rule is to the election. It is not a limit on freedom. It is the precondition of it, and a man who cannot tell a rule that enables the game from a rule that fixes the game has not understood either.
The statist error is the mirror: the market fails, therefore replace it with the plan, the expert, the command. But the plan destroys the aggregator, and an aggregator destroyed does not give you a wiser allocation; it gives you a blind one, enforced, because the knowledge the price carried cannot be reconstructed by any committee, and the shortage that follows can only be papered over with coercion. The cure for market failure is not the abolition of the market any more than the cure for democratic failure is the abolition of the vote. It is a better constitution for the market — the same move, the same middle, the same refusal of both abysses.
Both errors make the identical mistake, and it is the mistake this whole sequence has been circling: both treat the aggregator as a single thing with a single verdict — an oracle to be left utterly alone, or a failure to be replaced by command — when it is a mechanism that does one job superbly and another disastrously, and therefore needs neither worship nor abolition but a constitution: a fixed layer of rules that lets it keep doing the job it is miraculous at while forbidding it the failures it is prone to. That is what a regulated market is. That is what a republic is. They are the same answer to the same problem, and the problem is that the most powerful machine humanity has built for coordinating itself will, if left entirely alone, coordinate itself straight into ruin — and will, if replaced by a commander, go blind — and the only thing that has ever threaded that needle is a constitution: few rules, fixed, structural, revisable, held beyond the reach of the crowd of the moment, precisely so the crowd can keep governing itself.
Ask not whether the market should be free or controlled. That question is malformed, and it has wasted a century. Ask instead the only question that was ever real, the one that decides whether a rule belongs in the constitution of the marketplace or has no business there at all:
Does this rule preserve the game, or fix the score?
Is it so?
State. Classify. Done.
The scaffolding is standard and checkable: the socialist-calculation argument that without market prices an economy cannot rationally allocate resources, because the necessary knowledge exists only in dispersed and largely tacit form; the standard economics of externalities, monopoly, public goods, and fraud as failures the price mechanism does not self-correct; the theory of regulatory capture and the public-choice account of regulators as self-interested actors; the tragedy of the commons and the demonstration that commons can nonetheless be governed by well-designed rules; and the founders’ constitutional design of fixed rules that govern the game while leaving its outcomes open. Nothing here endorses a party or a programme; the argument concerns the structure of the machine, not anyone’s side of it.