№ XI
The twelve plates
Every market rests on a promise: that the price of a thing captures its worth, so that a person pursuing his own gain is led, as if by an unseen hand, to serve the general good. The externality is the broken promise. It is the cost — the factory's smoke, the carbon in the air — that falls on someone who was never party to the bargain, and that the price does not count. This volume anatomizes the spillover, and the long effort to force it back into the price.
who pays for the smoke, when the price of the goods does not?
Every defense of the market comes down, in the end, to a single promise: that a price tells the truth about a thing. Pay what others freely ask, the promise runs, and you will be paying exactly what your wants cost the world to satisfy — so that each person, pursuing nothing but his own advantage, is led as if by an unseen hand to weigh his desires against their real cost and to serve, without intending it, the good of all. The promise has a condition buried in it, so obvious it is easily missed: the price must capture the whole of what a thing costs to provide. The previous volume of this group examined one way the promise fails — when one side of a market grows too strong, and sets the price rather than takes it. This volume examines another, subtler and more pervasive: the case where the price is struck in a perfectly free and open market, and is wrong all the same, because part of what the thing costs is paid by someone who was never at the table.
The classic instance is a smoking factory. It makes cloth, which it sells, and smoke, which it does not. The cloth carries a price that covers the wool and the wages and the wear on the looms, and the factory, like any sensible firm, produces up to the point where one more bolt would cost it more than it fetches. But the smoke costs something too — it soils the wash hung in the next street, blackens the buildings, settles in the lungs of those who breathe it — and that cost the factory does not bear, for the air is free to foul and no one sends an invoice for its fouling. So the factory reckons only its private cost, the part that falls on itself, and ignores the social cost, the larger sum that takes in the harm to everyone else. Weighing too little against its revenue, it makes too much cloth, and too much smoke besides. The market has not led it to serve the common good; it has led it to injure the public and call the injury efficiency — because the one figure that should have told the truth, the price, was silent about the part of the cost it shifted onto strangers.
The first economist to anatomize this with rigor was Arthur Pigou, who in 1920 gave the gap its lasting name: a divergence between private and social cost, between what an act costs the one who chooses it and what it costs the world. His remedy followed from the diagnosis with great clarity. If the trouble is that the factory does not feel the cost of its smoke, then make it feel it — lay a tax on the smoke equal to the harm it does, and the factory, now obliged to count that harm as its own expense, will pull its output and its fouling back to the measure society would have chosen. The tax does not forbid the smoke; it prices it, forcing the external cost into the private reckoning, so that the invisible hand, given true figures at last, resumes its proper work. There is a catch in it, which later plates must face: to set the tax rightly the state must first know the size of the harm, and the harm — a darkened sky, a shortened life, a warmer century — is the very thing hardest to put a number on. The same logic, run in reverse, yields a subsidy for the good whose benefits spill onto others — for schooling, for invention, for the vaccine. This is the Pigovian tax, and the carbon tax debated today is its direct and most consequential descendant.
For forty years that was the orthodoxy, until in 1960 Ronald Coase published a paper that accepted Pigou's analysis entire and then turned it inside out. Coase began with a correction that sounds at first like mere logic-chopping and proves to be a hinge: the harm, he said, is reciprocal. We speak of the factory harming the laundry, and so it does; but to silence the factory is to harm the factory, and the cloth it would have made is wanted no less than the clean wash. To ask only how to stop the factory from injuring the laundry is to assume that the laundry's claim on the air outranks the factory's — which is the very question in dispute. The right question is not who is at fault but which use of the air is worth more. And then came the stroke that made his name: if the right to the air is clearly assigned to someone, and the parties can bargain at low cost, they will reach that more valuable use on their own — the laundry paying the factory to fit a filter if clean air is worth more, the factory paying the laundry to bear the smoke if cloth is worth more — and they will reach it whoever was handed the right to begin with. The efficient outcome needs no tax. It needs only an owner and a bargain.
This is the Coase theorem, among the most cited and most misread ideas in the discipline, and its sting is real: the externality, on this telling, is not a flaw inherent in the market but a sign that some right has been left undefined, or that bargaining over it is too dear. Yet Coase drew no laissez-faire moral from it, however often the careless have since. His whole argument turned on the proviso the theorem is usually stripped of — that bargaining be cheap — and he knew that in the cases that matter most it is anything but. When a factory's smoke falls on a single neighbor, the two may strike a deal; when it falls on a city of a million lungs, the million cannot assemble to negotiate, each holds out for more or waits for another to pay, and the bargain that would serve them all never forms. The lesson is not that the state should always act, nor that it never should, but that the right remedy turns on the cost of transacting: leave the small and local spillover to the parties and the courts, and reach for the tax or the rule only when the affected are too many to treat. And some spillovers defeat even that division — the carbon poured into a sky shared by every nation and every generation to come, where the harmed number in the billions, own no common government to assign the rights, and cannot, being unborn, sit at any table at all. The quarrel between Pigou and Coase is, rightly understood, less a war than a division of cases.
Beneath the dispute over remedies lies the matter this table everywhere pursues. A price, it has argued throughout, is a verdict on a hidden contest, mistaken for a fact of nature; and the externality is the shape that verdict takes when one of the contestants is not in the room. The price of coal-fired power looks like the cost of keeping the lights on, a plain technical fact — but it is a judgment, rendered by silence, that the cost of the smoke and the carbon shall be charged to the asthmatic child and the unborn generation rather than to the buyer and the seller who struck the deal. The market has not weighed those costs and found them small; it has not weighed them at all, because the parties who bear them had no voice in the price. Every remedy this volume examines — Pigou's tax, Coase's bargain, the tradable permit, the regulator's standard — is at bottom a device for dragging the silenced cost back into the figure, for making the verdict count the claims it had left out. And the deepest difficulty is that the silenced claims are everywhere, since nearly every act we perform sheds some cost or benefit onto someone who never consented — so that the concept, pressed to its limit, would draw the whole of life under the price's eye. The twelve plates trace the spillover from its definition and Coase's theorem through the smoke and the gift, the congested road and the warming air, the permit and the holdout, the shadow price and the unequal burden and the regulator's standard, to the edge where the idea at last overreaches. Where the previous volume asked what happens when one party to a bargain is too strong, this one asks what happens to those who are no party to it at all. The argument remains open.