The fence can multiply until no one can open the gate. A resource that would be exhausted by unrestricted use can also lie idle beneath overlapping powers to forbid its use. The people who hold those powers may all want the resource to be productive. Each may prefer a completed bargain to an empty building, an abandoned project or a method no one employs. Yet each can wait for a larger share of the gain, and each can impose terms that make the whole undertaking less attractive. The tragedy has turned inside out: the commons was vulnerable because a user could proceed without bearing the full consequence; the anticommons is vulnerable because too many permissions may be needed before anyone can proceed at all.
The relevant fragmentation concerns the powers attached to a resource, rather than the number of names attached to its ownership. A company can have thousands of shareholders and a settled procedure by which one authorized decision commits it. Two people can instead hold separate rights whose exercise is indispensable to the same use, with neither able to settle the terms for both. Heller's anticommons identifies multiple effective rights to exclude from a scarce resource. The rights can overlap in one object or govern pieces that must be assembled for a particular purpose. What matters is the relation among them: who can refuse, whose agreement binds whom, and whether any rule can resolve the competing claims.
The earlier Mirror located this danger on the far side of enclosure. A boundary can make someone answer for a resource; dividing the power to cross it can make everyone else's permission another object of negotiation. A workshop may need access through two passages controlled separately. A proposed use may require the consent of several parties whose rights protect different interests. The arrangement creates a complementarity: obtaining one permission is useful only if the others can also be obtained. This differs from having several possible workshops or routes among which a user can choose. Additional alternatives may strengthen the user's position. Additional indispensable permissions can strengthen each holder's power to obstruct the same choice.
A small pricing example shows how individually reasonable demands can jointly suppress use. Suppose customers need two independent permissions to use a facility, and each holder charges a fee per use. Let demand be twelve minus the combined fee, with zero demand once the combined fee reaches twelve. The facility already exists, capacity is ample, another use consumes no additional resources, and there is no congestion or outside harm. The fees are simultaneous, each holder seeks the largest receipt, and neither bargains with the other over a joint price. If one charges four, the other's demand is eight minus its own fee. A fee of four then yields the largest receipt: four uses at four each. In this interior equilibrium, both charge four, users pay eight in total, and only four uses occur.
Each holder overlooks part of the cost that its demand imposes on the other. Raising one fee reduces the number of users who buy both permissions. The holder weighs the resulting loss of its own receipts, while the companion's lost receipts fall outside its account. If the permissions were priced together, their owner would choose a total fee of six, attracting six uses and collecting thirty-six. The separated holders collect only thirty-two between them. Each insists on its own margin and together they lose even part of the return they could have shared. With a third indispensable, independently priced permission, a positive-use equilibrium has each holder charging three: the total charge rises to nine, and use falls to three.
Coordinating the permissions improves this allocation without making the resulting price socially ideal. At four uses, the combined value to users is forty: eight remains with them after payment and thirty-two reaches the holders. At six uses, the total value is fifty-four: users retain eighteen and the unified owner receives thirty-six. The gain of fourteen comes from two additional uses whose value exceeds their stipulated resource cost. The fees themselves move money between people. A single owner still charges for access that costs nothing at the margin, excluding further valuable uses. Under this example's deliberately narrow assumptions, free access would permit twelve uses and yield total value of seventy-two. Removing a fragmented veto and removing monopoly restriction are different improvements.
The possibility of a larger joint return gives the holders a reason to agree, but leaves the work of agreement to be done. They must identify the rights, establish who can transfer or license them, decide how to divide the receipts and make a promise that survives a change of owner. One holder may believe that waiting will reveal a more profitable use. Another may distrust the account through which the joint proceeds would be distributed. A dispute about who deserves the gain can prevent a gain that every party would otherwise welcome. Heller accordingly treats bargaining, norms and coordinated control as possible ways out, while retaining the transaction costs and strategic obstacles that can make those ways difficult.
Even a real improvement can be too small, or too poorly distributed, to finance its own coordination. Suppose joining the two permissions in the example requires continuing administrative work costing five. The change still creates a net total gain of nine: fourteen in additional use value less five in new work. But the holders' gross receipts rise by only four, from thirty-two to thirty-six. If they alone must pay the five, their combined net receipt falls to thirty-one. The beneficiaries include users whose additional gains do not automatically arrive in the holders' account. A different financing arrangement might support the improvement. Declaring that everyone could gain is not yet a procedure for collecting the contributions and securing the promises that make it happen.
A right to refuse may also protect something that an aspiring user has left out of the price. The neighbors' quiet, the integrity of a watershed or the continued existence of a place can matter even when the proposed project earns a profit. Return to the facility example without the added administrative cost. Expanding use from four to six added fourteen in value because no outside harm was assumed. If that expansion instead caused additional harm of twenty, its combined effect would be a loss of six. The calculation changes because a previously excluded cost has entered. A resource left unused can be wasted; it can also be preserved for a reason that the would-be user has no incentive to count.
The anticommons therefore supplies a question to investigate rather than a count of objections that proves obstruction. Heller expressly recognizes that overlapping exclusion can sometimes be useful, that coordination can prevent the feared paralysis, and that actual underuse is an empirical matter. Buchanan and Yoon likewise qualify the extension of their pricing model to public agencies: officials may pursue purposes other than revenue and may refuse permission only for specified reasons. A finding that two consents are required does not establish that either is needless. The inquiry must trace the interests each protects, the costs of obtaining it, the possibility of a joint decision and the consequences of proceeding without it.
Useful reform can connect decisions without erasing every independent claim. A common licensing office can combine complementary permissions under an agreed allocation of receipts. A clear order of decision can keep a later approval from reopening a matter already settled. A procedure for resolving conflicting claims can reduce the advantage of indefinite delay while requiring that a displaced interest be heard and addressed. These arrangements differ in whom they empower and in the mistakes they risk. Consolidating authority can relieve a bottleneck and create a more powerful gatekeeper at the same time. The governing task is to make the consequences answerable, including the consequences of the new concentration of control.
The original question's two extremes do not identify a universal middle number of owners. A widely shared resource can be governed through workable rules; a resource under one owner can damage its neighbors; two indispensable permissions can block uses that a coordinated arrangement would allow. The productive boundary depends on what the resource is, what its use does to others, how decisions are connected and what it costs to change them. Even the meaning of productive use can be contested when livelihoods, places and future options stand behind the competing claims. A precise account of the blockage makes that disagreement more intelligible. A general preference for more ownership, or for less, cannot do the same work.
A right to protect the resource can become a price for allowing it to be used.
The gridlock begins when separately reasonable powers cannot be brought into a decision adequate to the whole undertaking. The remedy must recover the value that fragmentation prevents while accounting for the interests that exclusion was meant to protect. Sometimes that calls for joining rights; sometimes for making a shared decision more dependable; sometimes for accepting that the proposed use ought not proceed. The next plate considers a further reversal: knowledge whose use by one person leaves it available to another, even while its creation and preservation depend on scarce work. The argument remains open.