The Exit

The volume has anatomized market power — the monopolist who sells too dear, the monopsonist who buys too cheap, the barriers that shelter them, the cartels that contrive their power, the law that would break it. Through all of it has run a question the author set at the outset and the plates have circled: what is a fair price worth, when one side has nowhere else to go? The flagship takes up that question directly, and in answering it finds that power, for all its varied anatomy, is at bottom a simpler thing than any of its forms suggested — and that a fair price is not the kind of thing the volume, or the reader, has been assuming it to be.

Strip market power of its forms and ask what remains — what the monopolist and the monopsonist and the colluding few all have in common — and the answer is this: in every case, the party on the other side has no alternative. The monopolist has power over his buyers not because he is one but because they have no other seller to turn to; were there another, his power would vanish, however large he remained. The monopsonist has power over his sellers for the identical reason — they have no other buyer. The cartel has power because, while it holds together, it has foreclosed the alternatives its members would otherwise have offered one another's customers. The barrier to entry is precisely the thing that prevents an alternative from arising. Market power, in every guise, is the absence — for the party who must trade — of somewhere else to go. It is not a property of the powerful at all, but a property of the powerless: their lack of an alternative is the whole of the other's power.¹

This dissolves the distinction the textbooks draw between selling power and buying power, and restores them as one. We are taught to treat monopoly and monopsony separately, the one a seller's power over buyers, the other a buyer's power over sellers, as though they were different phenomena needing different theories. They are not. Both are the same fact — one side has nowhere else to go — seen from opposite ends. When the buyers have no other seller, we name the seller a monopolist and say he sells too dear; when the sellers have no other buyer, we name the buyer a monopsonist and say he buys too cheap; but in both the powerful party is simply the one whose counterpart is trapped, and the price is bent toward whoever is not trapped, by exactly the measure of the other's entrapment. The volume on the wage met this power in one of its forms — the worker with a single employer, trapped, underpaid — and could see only that one organ; seen whole, the underpaid worker and the gouged customer suffer one thing, the absence of an alternative, and monopoly and monopsony are revealed as the two faces of the single power the gloss names: too few on one side, which is only ever a way of saying that the other side has nowhere else to go.²

Here the inversion turns, and it reaches back to recast the whole of the price families that preceded this one. We took the competitive price — the price that ruled when sellers were many and buyers were many — to be fair because it was disciplined by numbers, the press of many rivals forcing it down to cost. But numbers were never the substance of the thing; alternatives were. The competitive price is fair not because the sellers are many but because each buyer can always turn to another, and each seller to another buyer — because everyone, always, has somewhere else to go. It is that standing possibility of departure, that ever-available exit, that disciplines the competitive price: the seller dares not gouge because his customer would simply leave for a rival, the buyer dares not lowball because his supplier would simply sell elsewhere. The many rivals matter only as the guarantors of the exit; it is the exit itself, the freedom to walk away to an alternative, that does the disciplining. Competition, rightly understood, is not a count of firms but a condition of universal exit — a state of the world in which no one is ever trapped with a single counterparty, and so no one can ever be made to take an unfair price. And market power is simply the failure of that condition: the closing of the exit on one side, the trapping of a party with a counterpart he cannot escape.³

Now the author's question can be answered, and the answer is harder than the question expects. What is a fair price worth, when one side has nowhere else to go? The answer is that there is no fair price then — that fairness in price is not a thing that survives the closing of the exit. For a price is fair, if the word means anything in a market, when it is the price that free parties, each able to walk away, agree upon; fairness is a property not of the good, nor of its cost, nor of any number written on a tag, but of the bargaining position from which the price emerges — and when one side cannot walk away, the bargaining position is not a fair one, and no price struck from it is a fair price. The trapped party does not pay a fair price held slightly too high; he pays whatever the untrapped party chooses, limited only by what he can bear before he ceases to trade at all. To ask what a fair price is worth when one side is trapped is therefore to ask a question that cancels itself: the trapping is the absence of the very conditions under which prices can be fair. There is only the price of being trapped, which is no fairer or less fair than a ransom — and the whole apparatus of competition, of entry, of antitrust, and of invention exists, when one sees to the bottom of it, to do one thing: to give the trapped party somewhere else to go, and so to restore the conditions under which a price can be fair at all.⁴

What recourse, then, has the trapped party, if no fair price is to be had where he stands? Hirschman, who gave the idea of exit its name, paired it with another: voice. When we cannot leave a thing that fails us, we may yet stay and protest — complain, organize, demand redress — and where exit is foreclosed, voice is what remains. The trapped customer who cannot switch can petition, agitate, sue, vote; and antitrust itself is a kind of collective voice, the public's organized protest against a power its members cannot individually escape. But there is a deeper recourse than voice, and it is the one a later plate presses: the manufacture of new exits by invention. For alternatives are not fixed; they are created. The monopolist who has foreclosed every existing exit cannot foreclose the exit that does not yet exist — the new product, the new method, the new rival who will give his trapped customers, at last, somewhere else to go. This is the deepest meaning of creative destruction: that the relentless invention of new alternatives is the ultimate enemy of every entrenched power, prying open exits that no law could force and no incumbent could bar. The prize of being, for a time, the only door is precisely what lures the next innovator to build a second door beside it — and so the very profit of market power funds, in the end, the alternatives that destroy it.⁵

The account must not be pressed into a complacency it does not license. That power is the absence of an alternative, and that alternatives are continually created, does not mean that power is harmless or that it always passes — for the closing of exits is real while it lasts, the trapped pay dearly in the meantime, and some power works precisely to foreclose the alternatives that would end it: the incumbent who buys the rival before he can grow, who locks customers into a network they cannot leave, who lobbies for the barrier that forbids the entrant, who lets his service rot because his captives cannot flee. Such power does not wait to be swept off by creative destruction; it suppresses the destruction, holding its exits shut by main force, and against it the slower correctives are no help — only voice, only the law, only the deliberate prying-open of the door. The dynamic view licenses patience with the power that earns its place by offering, for a time, the best door, and that a rival may yet outbuild; it licenses nothing toward the power that survives only by nailing the other doors shut. The whole difficulty of policy, the reason the pendulum of antitrust swings and never settles, is that these two powers wear the same face — both are firms grown large and profitable — and telling the one that should be left alone from the one that should be broken is the hard, unfinished work that the simple fact of bigness can never decide.⁶

With the power, the third family stands complete, and its argument can be seen whole. The price, the great coordinator of the families before it, fails in four ways, and the failures have now mapped them all. It cannot convey what one side hides — the asymmetry, where private knowledge corrupts the price as a signal. It cannot count what no one owns — the externality, where costs fall outside the boundary of property and so outside the price. It cannot defend what is shared — the commons, where a good that none can be charged for and all deplete lies wholly beyond the price's reach. And it cannot discipline what it cannot contest — the power, where a party too large to be checked by competition escapes the discipline that would have forced its price to cost, and bends the price instead. Three of the four corrupt what the price carries; the fourth corrupts the discipline that keeps the price honest — for competition, the volume has shown, is nothing but the universal availability of exit, and market power is its failure. Four edges of the price, four regions where it cannot be trusted to coordinate alone, and four reasons the market, magnificent as it is, is not the whole of economic life.⁷

And so the failures end, and with them the first great movement of the table. They began by honoring the price — the wage, the rent, the rate, the premium, the spread, each a price doing its coordinating work — and they end by mapping the four frontiers beyond which that work breaks down: hidden knowledge, absent ownership, shared goods, and escaped competition. At each frontier the volume found the same thing: that where the price fails, it must be supplied or supplanted by something that is not a price — a device to convey what it cannot signal, a right to enclose what it cannot count, a covenant to govern what it cannot defend, and a rule to discipline what it cannot contest. These — the device, the right, the covenant, the rule — are not the market, but the scaffolding the market stands on and the soil it grows from; and they are the subject of everything the table has yet to build. The institutions that supply them — the firm, the contract, the money, the tax, the rule — are a family still to come. The behaviors that animate them — the choosing, the trusting, the strategizing, the norm-keeping — are the family that ends the table, and the foundation beneath all the rest. The price was the table's first subject and its most powerful; the failures are the family that found its limits; and beyond those limits lies the larger half of the science, where the institutions are raised and the behaviors play out that the price alone could never command. A fair price, this last failure has shown, requires that the one who pays it be free to walk away; and whether he is free — whether the door is open or held shut — depends, in the end, not on the price at all, but on the institutions we build and the games we choose to play.⁸