What is it to own a firm? The question sounds simple until the nexus view complicates it: if the firm is only a meeting of contracts, and the owners are merely one party to the knot, alongside workers and lenders and suppliers, what do the owners have that the others lack? Grossman and Hart gave the deep answer, and it turns on the fact that has shadowed every plate of this volume — that no contract can foresee everything.
Begin with that incompleteness, now made central. A contract tries to specify what each party will do in every circumstance, but the circumstances are endless and the future is dark, and so every real contract is incomplete: it falls silent on countless contingencies it did not foresee or could not cheaply describe. The supply agreement does not say what happens if a new technology makes the product obsolete; the employment contract does not say which project the engineer will take up next quarter. Most of the time this silence is harmless, but when an unforeseen circumstance arrives, someone must decide what to do — and the contract, being silent, does not say who. Here is the gap that ownership fills. For the law's answer is that whoever owns the asset holds the right to decide its use in any matter the contract did not settle. Ownership is the default authority over the unspecified — the residual right of control, the say over everything the contracts left open.1
This redefines ownership in a way at once strange and exact. We usually think to own a thing is to have its use and its fruits; the property-rights theory says, more precisely, that to own a thing is to hold the residual control over it — the right to decide its disposition in all the cases no contract addressed. The fruits follow from this, but the control is the essence. And it illuminates what was murky in the nexus view: the owners of a firm are distinguished from its other contractors not because they alone have a contract, but because they hold the residual rights the others' contracts did not claim. The bondholder has a contract for fixed repayment and nothing more; the supplier, a contract for delivery and payment and nothing more; but the owner has whatever is left over — the residual profit after the fixed claims, yes, but more fundamentally the residual authority to direct the assets in everything the other contracts did not specify. To own the firm is to be the one who decides when the contracts run out — and since the contracts always run out, somewhere, the owner's residual authority is the firm's ultimate power of decision.2
If ownership is residual control, then the make-or-buy question — should two activities be one firm or two, integrated or traded — becomes the question of who should hold the residual rights over which assets, and Grossman and Hart answered it with precision. Recall the hold-up problem: a party who must make an investment specific to a relationship fears being squeezed afterward, and so underinvests. Residual control determines who can do the squeezing — for the owner of the assets decides the unforeseen, and so holds the whip hand when the unforeseen arrives. Therefore, the theory says, ownership should be placed with the party whose specific investment matters most — for that party, secure as owner against being held up, will invest fully, where another, exposed as a non-owner, would hold back. Integration is good when one party's investment is so much more important that giving it the residual control unleashes its investment more than it chills the other's; it is bad when both parties' investments matter, for then whoever is made the non-owner will underinvest, and the firms are better kept separate, each owning its own critical assets. The boundary of the firm, which Coase drew by transaction cost and Williamson by asset specificity, Grossman and Hart drew most sharply of all: put the residual rights where the investment that most needs protecting lives.3
The theory carries a crucial corollary that tempers any rush to integrate: ownership cannot solve every problem at once, because residual control is a single thing that can be held by only one party. Give the residual rights to me, and my specific investment is protected, but yours — for you are now the non-owner, exposed to my authority over the unforeseen — is chilled. There is no arrangement in which both parties feel fully secure, for ownership protects whoever holds it at the expense of whoever does not; and so integration always cures one hold-up by creating another, protecting the owner's investment by exposing the non-owner's. This is why the world is not one giant firm: integration is not free, but trades one party's incentive for another's, and the gain is worth the loss only when one investment so dominates that protecting it is worth chilling the rest. The firm boundary is thus a boundary of incentives — drawn to protect the investments that matter most, at the unavoidable cost of exposing the investments that matter less — and the art of it is to own what you must to call forth your own critical investment, and to leave to others, in separate firms, what they must own to call forth theirs.4
With ownership understood as residual control, the volume's three accounts of the firm's boundary stand revealed as one deepening argument. Coase saw that the boundary falls where the cost of the plan meets the cost of the price, but left the costs unspecified. Williamson found the chief cost in asset specificity and the hold-up it invites, and predicted that specific transactions would be integrated. Grossman and Hart explained what integration actually does — it reallocates the residual control rights, the authority over the unforeseen — and so why it helps, by protecting the owner's specific investment, and why it hurts, by exposing the non-owner's, and thus exactly when it is worth doing. The three are not rival theories but a single insight refined: that the firm exists because contracts are incomplete and the market therefore costly; that the incompleteness bites hardest where assets are specific; and that the firm's distinctive instrument against it is ownership, the residual right to decide what the contract did not. The boundary of the firm is the boundary of residual control — drawn, asset by asset, to put the power of the unforeseen where it will call forth the investments most worth having.5
The owner, then, is the holder of the residual: not merely the one who takes the profit left after all claims are paid, but the one who decides everything the contracts did not — and since no contract can foresee everything, that residual authority over the unforeseen is the whole substance of ownership and the firm's ultimate power. Grossman and Hart's insight completes the volume's account of the firm's boundary, revealing make-or-buy to be the question of who should own which assets, answered by placing the residual rights where the most important specific investment lives. And it sets the stage for the largest question of all. We have seen why firms exist, where their edges fall, what they are made of, and what it is to own one. It remains to ask what the firm is, in the widest frame — and the answer, the flagship now argues, is that it is the thing the great ideological war of the last century declared the loser: planning, alive and indispensable, dwelling within the very market that was supposed to have rendered it obsolete. The next plate is the flagship. The next plate is the plan.