The Commons

Picture the commons in its oldest form: the village pasture, a stretch of grazing land open to all the households of the village and owned outright by none of them. Any villager may turn his cattle onto it; the grass is there for whoever brings a beast to eat it. Two facts about this pasture, taken together, contain the whole of what follows. The grass is rival — or, in the economist's harsher word, subtractable — for a mouthful eaten by one man's cow is a mouthful denied to every other's, and the resource can be used up. And the pasture is non-excludable: there is no fence, no gate, no toll, and by custom or by law no villager can be turned away, so the resource stands open. A good that is at once rival and open is what the modern literature calls a common-pool resource, and it is the unstable compound at the center of this volume.

The pairing comes clear when the four kinds of good are set side by side, sorted by the two questions the figure poses: is the good rival, and is it excludable? The familiar private good — a loaf, a coat, a car — is both rival and excludable, and the market governs it without difficulty, for a price can be charged and the unwilling kept out. Its opposite is the pure public good — national defense, a lighthouse, the light shed by a discovery — which is neither rival nor excludable: one man's protection or illumination subtracts nothing from another's, and none can be made to pay, so the market underprovides it. Between these poles lie the two mixed cases. The club good is excludable but not rival — a toll road, a cinema, a subscription — which can be sold to members and crowds only when it congests, and which the tenth plate takes up. And the common-pool resource, the subject here, is the fourth and most treacherous cell: rival like a private good, yet open like a public one. It carries the scarcity that demands rationing together with the openness that forbids it.

From that single misfit the tragedy follows in miniature, and the next plate will draw it in full. Because the grass is rival, each additional cow does a real harm — it leaves less for all the others; because the pasture is open, no one can charge that cow's owner for the harm, nor keep him out to prevent it. So the herder weighs only what is his to weigh: the gain of one more animal, which is wholly his, against its private cost, which does not include the grass it strips from his neighbors' beasts. He adds the cow. Every herder reasons the same way, and the pasture fills past what it can bear; the grass fails, and the cattle starve on ground that, husbanded, would have fed a smaller herd indefinitely. Each man, pursuing his own good by impeccable logic, helps to destroy the thing on which all of them depend.

This is the externality of the preceding volume, returned in a new and symmetrical form. The cost the cow imposes on the other herds is a real cost falling on third parties outside any transaction — an external cost, uncounted by the one who causes it. But where the eleventh volume's externality ran one way, from a polluter to a sufferer, here it runs every way at once: each herder fouls the pasture of all the others and is in turn fouled by them, so that every user is at once the author of the harm and among those who suffer it. The commons is the externality made reciprocal and multiplied across a multitude — which is why its diagnosis borrows the eleventh volume's tools while its cure runs into difficulties all its own. To mend an externality between two parties, one may tax the one or seat the two at a table; to mend an externality among five hundred symmetric herders, with no table large enough and no obvious party to tax, is the harder problem these plates pursue.

Set the parable down as a model, in the manner the discipline has refined since Hardin. A pasture supports cattle, and its total yield — call it beef, or milk, or simply output — rises as cattle are added, but by less and less with each beast, for the animals begin to compete for a fixed quantity of grass; past some number the pasture is so crowded that another animal lowers the total yield outright. Two curves describe this. The average product is the yield per animal, the total output divided by the number of cattle, and it falls steadily as crowding sets in — each cow, on a more crowded pasture, gives less. The marginal product is the addition to total output from one more animal, and because the average is falling the marginal lies below it and falls faster: the latest cow not only yields little itself but drags down the yield of all the rest.

Now ask what number of cattle each kind of decision-maker will choose, against a common cost — the price of acquiring and keeping a beast, drawn as a horizontal line. A sole owner of the pasture, who captures all the output and bears all the cost, reckons at the margin: he adds cattle only so long as one more adds more to total output than it costs, and stops where the marginal product meets the cost. That is the efficient herd, marked Q* in the figure, and at it the pasture yields its greatest surplus over cost — the rent of the land, the shaded rectangle. But the commoners face a different reckoning. No commoner captures the marginal product, for the loss a new cow inflicts falls mostly on the others' animals and only slightly on his own; what he captures is the average product, the yield his beast will fetch him. So he adds cattle as long as the average product exceeds the cost — and entry continues, by him and by every other, until the average product has fallen all the way down to the cost. That is the commons equilibrium, far to the right of Q*, with many more cattle than the pasture should ever bear.

The gap between the two equilibria is the tragedy made quantitative. Because the marginal product lies below the average, the point where average product meets cost falls far to the right of where marginal product meets cost: the commons carries more cattle than is efficient, and the surplus the land could have yielded — the rent a sole owner would have kept — is not transferred to anyone but destroyed. As cattle pile in toward the commons equilibrium, each adds a little to the total and subtracts more from the rest, and the rectangle of rent is eaten away until, at the equilibrium, the average product just equals the cost and the rent is gone entirely. This is the precise and devastating result on which the whole economics of the commons turns: open access dissipates rent to zero. The resource yields, in the end, no more than its users sank into it; everything it might have produced over and above their costs — the whole reason the resource was worth having — is competed away in the scramble to use it. The pasture is destroyed, as the argument promised, not by greed but by arithmetic.

The pasture is a parable; the fishery is where the economics of the commons became a working science. A fish stock renews itself, and the rate at which it grows depends on its size: a tiny stock grows slowly, for there are few fish to breed; a stock at the sea's carrying capacity grows not at all, for it has filled its niche; and somewhere between, at an intermediate stock, the growth runs fastest. From this biology comes the central object of the model, the sustainable yield — the catch that exactly matches the stock's growth, and can therefore be taken year upon year without running the stock down. Plotted against fishing effort — the boats, the days, the nets at work — the sustainable yield traces a hump. A little effort takes a little, from a large and barely-cropped stock; more effort takes more, up to a peak; and beyond the peak the heavy fishing holds the stock so low that, for all the added boats, the sustainable catch falls. The peak is the maximum sustainable yield, the largest catch the sea will bear forever — beloved of fisheries biologists and, as the economics will show, not quite the right target.

Now lay the economics over the biology. Value the sustainable catch at the market price, and the hump becomes a curve of sustainable revenue; set against it the cost of effort, a straight line rising from the origin, since each boat-day costs about the same. Three points on the picture matter. Where the revenue curve stands highest above the cost line — where the gap between them is greatest — is the maximum economic yield, the effort a sole owner of the fishery would choose, for there the rent, the surplus of revenue over cost, is largest; and it lies, tellingly, to the left of the biological maximum, a lighter touch than even the largest sustainable catch. The peak of the hump is the maximum sustainable yield. And far to the right, where the revenue curve has fallen back to meet the cost line, sits the open-access equilibrium — for under open access boats keep entering as long as any rent remains to be had, and entry ceases only when revenue has been dragged all the way down to cost and the rent is zero. The open-access fishery is fished not to its economic optimum, nor even to its biological one, but past them both, to the point where it yields its crews no more than their costs and the stock is gravely overfished.

This is no blackboard curiosity. The great fisheries of the world have been fished down the right-hand slope of this picture within living memory. The northern cod off Newfoundland, worked for five centuries, collapsed in 1992 under a fleet that open access and then a captured regulator had let grow far too large; the stock has never recovered, and tens of thousands lost their living in a single season. The Peruvian anchoveta, the largest single fishery on earth, crashed in the early 1970s under the same arithmetic, compounded by a shift in the ocean itself. In each, the pattern of the figure played out to the letter: too many boats, drawn in by a rent that their own numbers then annihilated, fishing a renewable resource clean past the point of renewal. The rent was fished away to the last dollar, and then the fish were fished away too. What the model adds to the parable of the pasture is rigor and reach — the same logic of dissipated rent, now quantified, fitted to data, and written across the collapse of the world's seas.

Why does the herder not see the ruin coming and stop — and why can the herders not simply agree among themselves to graze less? The answer is that the commons has the structure of a game, and the game is a trap. Strip it to the bone. Each user chooses between two courses: to restrain, taking less than he might, or to overuse, taking all he can. Follow one user's reckoning. If the others restrain, the resource stays healthy, and he does best by overusing — he reaps the fruit of their restraint while bearing none of its cost, the classic free ride. If instead the others overuse, the resource is doomed whatever he does, and he had far better grab his share before it vanishes than restrain and be left with nothing. Either way — whatever the others do — overusing pays him more than restraining. Overuse is, in the language of the twenty-fifth volume, a dominant strategy; and since every user faces the identical reckoning, every user overuses, and the resource is destroyed, though all of them would have fared better had all restrained. This is the prisoner's dilemma, played not by two but by the many, and its grip is the deepest root of the tragedy: the ruin is no failure of foresight and no lapse into vice, but the strict consequence of each person's rational choice.

Mancur Olson, in The Logic of Collective Action of 1965, drew out the part that sheer number plays, and it is decisive. Conserving the commons is itself a collective good — a benefit that, once produced by the restraint of some, is enjoyed by all, the restrained and the unrestrained alike — and the provision of collective goods is undone by free-riding. In a small group, each member's restraint makes a visible difference to the whole, and his shirking is noticed and resented by people he must live among; the temptation to free-ride is checked by visibility and consequence. In a large group, each member's contribution is a drop in the ocean, his shirking is invisible, and the benefit of his restraint accrues overwhelmingly to others — so he shirks, and so does everyone, and the collective good goes unprovided. Olson's conclusion was stark: large groups, lacking coercion or some separate selective incentive that rewards contributors one by one, will fail to act in their common interest, however vast that interest may be. The figure shows the trap in a single frame. At every level of others' cooperation, the individual's payoff from overusing sits above his payoff from restraining, so the rational user always defects, and the group comes to rest in the overuse trap at the left — far below the shared prosperity it would have reached, at the right, had all restrained.

Two consequences follow, and they organize the rest of the volume. The first is that exhortation is futile. To tell the herders to be less greedy, or the fishermen to think of the future, mistakes the nature of the problem: they are not greedy or short-sighted but trapped, and a trap is not escaped by good intentions. What must change is the structure of payoffs — and the three cures the coming plates examine are precisely three ways of changing it. Privatization changes the payoffs by making one party bear the full marginal cost of his use. Regulation changes them by fixing a penalty to overuse. And self-governance, the subtlest of the three, changes them by turning the one-shot dilemma into a repeated game among people who will meet again, so that the threatened loss of future cooperation makes restraint, at last, the rational choice. The second consequence is a thread that runs to the volume's end: because the trap tightens with numbers, the size of the group and the closeness of its members are not incidental but central. A handful of neighbors may slip the dilemma unaided; a planet of strangers may find it inescapable. The whole question of when a commons can be governed, and by what, turns out to be in large part a question of how many share it, and how well they know one another.

The first cure is the most direct, and the most drastic: abolish the commons by turning it into property. Carve the open pasture into fenced parcels, each with an owner; grant the fishery's catch to named holders; vest the aquifer's water in titled rights. The logic is the mirror of the tragedy. The commons failed because each user bore only a sliver of the cost of his use while reaping the whole of its gain; ownership repairs exactly this, for the owner of a parcel bears the entire cost of overgrazing it and keeps the entire reward of husbanding it, and so — consulting now a ledger in which the external cost has become internal — he grazes it to the efficient point and no further. The externality that each commoner imposed on all the others is internalized, in the eleventh volume's word, by the simple expedient of making one party the whole of the affected public. Where there is one owner, there is no one left to spill upon.

Harold Demsetz, in a celebrated essay of 1967, turned this static logic into a theory of how property rights come to exist at all. Rights are not free — to draw a boundary, record a title, and police a fence all cost effort — and so a resource held in common will be left in common as long as it stays abundant enough that the tragedy's losses are small. But let the resource grow scarce or valuable, and the losses from open access swell until, at some point, they exceed the cost of establishing property; and there, Demsetz argued, private rights emerge, summoned into being by the very scarcity that makes the commons ruinous. His illustration was the rise of private hunting territories among the Montagnais of Labrador: customary open access to the forest gave way to defended family lands precisely when the European fur trade made beaver pelts valuable enough that unrestrained trapping would have wiped the beaver out. Property, on this view, is no fixed institution and no natural right but an economizing response — it appears when, and where, and to the degree that internalizing an externality becomes worth its cost. The fence is built when the grass is at last worth fencing.

The modern instruments of this cure are subtler than the literal fence. Chief among them is the individual transferable quota, the device that has done most to rebuild collapsing fisheries: the regulator sets a total allowable catch and divides it into shares, each a private and tradable right to a fraction of the harvest. The holder of a quota now owns a stake in the stock's future — a healthier stock means a larger catch and a more valuable share — and so acquires the sole owner's motive to conserve; and because the shares may be bought and sold, the catch flows to whoever can take it most cheaply, restoring the efficiency that open access had destroyed. It is the cap-and-trade of the eleventh volume turned from a pollutant to a resource: a manufactured property right in something that had none. Tradable water rights do the same for rivers and aquifers, and the enclosure of common land did it, centuries ago, for the pasture. Where the instrument fits the resource it works, and works well — the fisheries managed by well-designed quotas are among the few to have recovered, because their owners, at last, want them to.

Two things the cure quietly assumes, and both matter. The first is that the resource can be bounded and defended at all — that a fence can be built around it cheaply enough to be worth the building. The second is that someone must be chosen to receive the property, and the choosing is enormous: to privatize a commons is to convert a thing that many used into a thing that one owns, and the question of which one — the incumbent fishermen, the lord of the manor, the first to file a claim — settles who is enriched and who dispossessed. Privatization solves the problem of efficiency by raising, in its sharpest form, the problem of justice; and the two problems, as the eleventh volume warned and this one repeats, are not the same problem. The fence that saves the grass also encloses it, and on one side of the fence stands an owner, and on the other, those who used to graze.

If the commons cannot, or should not, be carved into private property, the second cure leaves it shared but sets a sovereign over it. Hardin himself, for all that his name clings to the tragedy, did not prescribe privatization; he prescribed the state. His phrase was "mutual coercion, mutually agreed upon" — the recognition that, since no user will restrain himself while the others do not, and since exhortation cannot break the dilemma, the users may agree to be coerced, each consenting to a binding limit on all precisely because he cannot trust the rest to limit themselves unbound. The Leviathan of Hobbes is summoned for a Hobbesian reason: in the state of nature that is the open commons, each devours what all depend upon, and only a power standing above them, able to compel, can impose the restraint that none can impose alone. The state supplies, by coercion, the discipline that the missing price and the missing owner did not.

The instruments of this cure fill the statute books of every fishing nation and every environmental ministry. The total allowable catch, parcelling a fixed harvest among the fleet; the licence, capping the number who may enter; the closed season and the closed ground, the marine reserve and the national park, that put a time or a place beyond reach; the limits on gear, on mesh, on engine size, that blunt the efficiency of effort; the permits to extract, to pump, to emit. In each, the regulator stands in for the missing owner: he estimates the efficient level of use — the maximum economic yield of the fishery, the safe yield of the aquifer — and then rations access to it by command, allocating the scarce right to use where no price and no property allocated it before. Done well, regulation reaches the same conservation a sole owner would have chosen, while keeping the resource public and its users many.

For a whole class of commons, moreover, regulation is not one option among three but the only one to be had. The great resources that cannot be fenced and are shared by far too many to self-govern — the high-seas fisheries beyond any nation's waters, the air, the migratory stocks, the river that crosses borders — admit of no private owner and no face-to-face community; for these, the coercion of states, acting alone within their territories or together by treaty, is the sole instrument left. The whole apparatus of international environmental law — the fishing conventions, the whaling moratorium, the protocols on pollution and on the ozone layer — is the Leviathan extended, by halting and imperfect agreement, to commons that overflow every single sovereign. Where the resource is too large for the fence and too peopled for the community, the state, or a league of states, is the last resort and often the only one.

The dichotomy held the field for two centuries: a commons must be saved by the market or by the state, by the fence or by the Leviathan. Elinor Ostrom broke it — and she broke it not with a theorem but with evidence. A political scientist by training, schooled in economics, she did what the theorists of the tragedy had not: she went and looked at actual commons, hundreds of them, across the world and across the centuries, and found, again and again, the thing the dichotomy declared impossible. Communities of users were governing their shared resources themselves, neither privatizing them nor surrendering them to any state, but crafting their own rules of access and use, monitoring one another, punishing violators, and sustaining the resource for generation upon generation. The Swiss villagers of Törbel had governed their high alpine pastures and forests by communal statute since the thirteenth century; the Japanese villages their common woodlands, the iriai, for as long; the huerta of Valencia had watered its fields by a customary law of turns, adjudicated by a Water Tribunal that still convenes each week at a cathedral door; the lobstermen of Maine policed their own harbor grounds. None of these was Hardin's lawless free-for-all, and none was a fenced estate or a state fishery. They were commons, in the precise sense — common property, governed by their commoners — and they worked.

Why should self-governance succeed where the dilemma of the fourth plate proved it must fail? Because the dilemma of the fourth plate was the wrong model: it was a game played once, among strangers, and a commons is neither. A commons is a game played over and over by the same people, who will face one another again tomorrow and in the year to come; and in a repeated game, as the twenty-fifth volume shows, cooperation that is impossible in a single round becomes sustainable, held in place by the shadow of the future. If a user who cheats today can be punished tomorrow — shamed, fined, excluded, cut out of the cooperation he depends upon — then so long as he values the future stream of cooperation above the one-time prize of cheating, restraint becomes his rational choice, and the commons holds itself together with neither fence nor sovereign. But this self-enforcement carries a precise condition, and the figure draws it: cooperation survives only if the users are patient enough, and the patience required rises with the size of the group. In a small, stable, face-to-face community, where every user's conduct is seen and his cheating at once known and answered, modest patience suffices and cooperation holds; as the group grows large and anonymous, the threshold of patience climbs toward an impossible perfection, and below any attainable patience the cooperation collapses. Ostrom's enduring commons are, to a one, the small and the bounded — which is no accident but the very mechanism. They are the groups for which the repeated game can be won.

For this work — gathered in Governing the Commons of 1990 and a lifetime of fieldwork around it — Ostrom was awarded the Nobel Memorial Prize in 2009, the first woman to receive it. The honor marked more than a career; it marked the overturning of a settled idea. Where the discipline had recognized two solutions to the tragedy and had argued endlessly over which to prefer, Ostrom established a third and showed it was no curiosity but a workhorse, the means by which a great share of the world's forests and fisheries and irrigation systems had in fact been governed all along, beneath the notice of a theory with eyes only for the market and the state. She did not claim the community always works — the next plates are largely her own catalogue of when it does not. She claimed something more carefully revolutionary: that the community is a third mode of governance, standing in its own right beside the other two, so that the question is never simply whether to privatize or to regulate, but which of three instruments, or which blend of them, fits the resource and the people at hand. The dichotomy that had counted two was dissolved into a trichotomy, and the discipline of the commons was remade.

If self-governance works under conditions, what are they? Ostrom answered not by deduction but by comparison: she set the commons that had endured for centuries beside those that had collapsed, and asked what the survivors shared. The answer was a set of design principles — not a formula guaranteeing success, for she insisted to the end that there is no panacea, but a list of features that recurred among the long-lived commons and were missing or weak in the failures. Eight in number, they read as the anatomy of an institution that can win the repeated game. The boundaries of the resource and of the group of users are clearly drawn, so that it is known what is being governed and who may use it. The rules of use are matched to local conditions — to this soil, this stream, this stock — rather than imported from a template. Those bound by the rules take part in making and revising them, so that the users govern themselves and are not governed from without. The resource and the users' conduct are monitored, by the users themselves or by monitors accountable to them. Violations meet graduated sanctions — a warning first, then a small penalty, escalating only for the stubborn or the repeat offender — so that enforcement deters without shattering the community. Cheap and accessible mechanisms stand ready to resolve the disputes that arise. The users' right to organize and make their own rules is recognized by the outside authorities, who do not override it. And in large systems governance is nested — local units within larger ones, layer upon layer — the polycentric arrangement the final plate returns to.

These principles are no arbitrary checklist; each repairs a specific way the commons fails. Clear boundaries solve the founding problem of open access — they close the entry, converting Hardin's free-for-all into a bounded common-property regime with a known membership, the very thing the first plate distinguished from a true commons. Monitoring and graduated sanctions are the machinery of the repeated game made real: cooperation holds, the seventh plate showed, only if the cheat is seen and answered, and these are how a community sees and answers him — credibly, since the sanctions are enforced, and durably, since they escalate gently rather than expelling a neighbor at his first lapse and so dissolving the very community whose future punishments give the arrangement its force. Collective choice and the fit to local conditions harness the users' own knowledge — the knowledge a distant regulator lacks — and secure their consent, so that the rules are kept as theirs and not resented as another's. Recognition by the state keeps the institution from being swept away by the very Leviathan of the sixth plate. The design principles are, in sum, the engineering of trust and enforcement — the concrete devices by which a group of users manufactures, out of repetition and visibility and proportionate punishment, the cooperation the bare dilemma said they could never have.

The seventh plate's optimism has a boundary, and the eighth plate's principles, read backward, draw it. The conditions that let a community govern its commons are the conditions whose absence dooms it, and each design principle casts a shadow. Where the users are too numerous to know or to monitor one another, the repeated game's threshold of patience climbs past reach. Where they are anonymous and mobile — strangers passing through, harvesters who will never return — there is no reputation to stake and no future punishment to fear, and the shadow of the future that disciplines the village falls on no one. Where they are heterogeneous, split by wealth or interest or value, they cannot agree on rules, and the agreement a homogeneous village reaches with ease a divided multitude cannot reach at all. Where the resource or the technology changes faster than institutions can evolve — a new market that makes a local resource immensely valuable overnight, a new gear that multiplies each harvester's reach — the slow growth of custom is simply outrun. Where the resource is hard to observe, so that overuse goes unseen, or where its boundaries lie open and its users uncountable, monitoring fails, and with it the whole apparatus of restraint. The commons that cannot be self-governed is, in short, the one that is large, open, anonymous, divided, fast-changing, and unwatchable — the mirror image, point for point, of Törbel and the Valencian huerta.

These adverse conditions reach their extreme in a single class of resource, and it is the gravest the volume must face: the global commons. The atmosphere into which all the world pours its carbon and once poured its ozone-destroying gases; the high-seas fisheries beyond every nation's reach; the deep ocean, the polar regions, the orbital space about the earth — these are commons whose community of users is all of humanity and all its states, and that community is too vast, too various, too anonymous, and above all too far beyond any overarching authority for either the village's self-governance or the sovereign's coercion to apply. There is no world community to shame a defecting nation as a village shames a defecting herder; there is no world state to set and enforce a global quota; and the atmosphere can be neither fenced nor owned. This is why the climate, which the eleventh volume named the externality of externalities, is the hardest problem in the discipline's gift: it is the tragedy of the commons at planetary scale, a reciprocal externality among nearly two hundred sovereigns and eight billion people, with every condition that makes a commons governable absent and every condition that makes one tragic present. The treaties — Kyoto, Paris — are halting attempts to build, by negotiated consent, the missing institution that neither market nor Leviathan supplies.

And yet it is not hopeless, and the proof is the Montreal Protocol. The thinning of the ozone layer was a global commons problem of just this shape — a shared atmospheric resource degraded by emissions diffuse and worldwide — and it was, very nearly, solved: the world agreed, in 1987, to phase out the offending chemicals, and the ozone layer is healing. What made ozone tractable where carbon is not is itself instructive. The science was clear and dramatic; the substitutes for the banned chemicals were cheap and ready; and the parties that mattered were few — a handful of chemical firms and the governments that housed them. Climate enjoys none of these advantages: its science, though settled, is diffuse and slow in its effects; its substitutes touch the whole of the energy economy; and its parties are everyone. The lesson of Montreal is therefore double-edged. It shows that the global commons can be governed, that the tragedy is not destiny even at planetary scale — and it shows, with equal force, how demanding the conditions of that governance are, and how few of them the hardest commons enjoys.

Return to the four-cell matrix of the first plate, and to the cell the volume has so far passed over: the club good, excludable but not rival. A swimming pool, a golf course, a toll road, a cinema, a subscription service — these are goods that one member may enjoy without diminishing another's enjoyment, up to a point, and from which non-members can readily be barred. Because they are non-rival, they are shared like a public good; because they are excludable, they can be sold like a private one, and so, unlike the pure public good, the market provides them without difficulty — by a fee at the gate, a toll on the road, a price for the membership. James Buchanan gave them their theory in a paper of 1965, "An Economic Theory of Clubs," and the theory turns on a question the pure cases never raise: not whether to provide the good, but to how many. For the club good is non-rival only up to a point, and that point — the onset of congestion — is where it shades into the commons.

The optimal size of a club balances two opposing forces, and the figure draws their resolution. On the one side, each new member shares in the cost of providing the good, so that the cost falling on each existing member declines as the membership grows — a gain that pulls toward a larger club. On the other, each new member adds to the crowding — the pool more thronged, the road more clogged, the course more booked — so that the enjoyment of each existing member declines as membership grows past the point where congestion sets in: a loss that pulls toward a smaller club. The net benefit per member, the gain from shared cost set against the loss from crowding, therefore rises, reaches a peak, and falls; and the optimal club is the size at the peak, where the saving in shared cost from the last member admitted just balances the congestion he brings. A club smaller than this leaves cost-sharing gains untaken; a club larger crowds away more than the new members save. There is a right number of members, and it is neither one nor everyone.

The deeper service of the club good is that it reveals the boundaries of the whole taxonomy to be soft, and rivalry itself to be a matter of degree. A road with few cars upon it is a pure public good: my driving on it subtracts nothing from yours, and it is non-rival and, absent a toll, non-excludable. The same road at rush hour is a common-pool resource: now my driving slows yours, the road is rival, and we are back in the congestion externality of the eleventh volume, each driver imposing delay on all the others. The good has not changed; the crowding has changed it, sliding it from one cell of the matrix into another. And the remedy for the congested road is exactly the club's: charge for entry, exclude at a price, size the membership — which is why the congestion charge that the eleventh volume met in the streets of London is, seen from here, the club solution to a common-pool problem, a toll that turns an overgrazed public road into a rightly-sized club. A great many of what we call commons problems are at bottom congestion problems — the crowded park, the jammed bandwidth, the overloaded server — and a great many of their solutions are at bottom clubs.

The volume has treated the commons as a thing prone to tragedy and the fence as its cure. This plate sets against that picture three inversions, in which the logic turns over: cases where the commons is no tragedy, and cases where privatization is itself the disease. The first inversion is the comedy of the commons, named by the legal scholar Carol Rose in 1986. Some resources grow more valuable the more they are used. A public square, a market, a fair, a harbor, a street thrown open to all comers — the worth of these lies in the gathering, in the crowd itself, so that each additional user, far from subtracting from the rest, adds to what all enjoy. For such resources openness is not the road to ruin but the path to greatest value, and the right policy is not to fence but to fling wide; restricting access destroys the very worth the commons creates. The network goods of the eleventh volume are the modern instance — a telephone system, a language, a marketplace whose value swells with every user — and they are commons whose proper ending is comedy, not tragedy.

The second inversion is deeper, and turns on a property the volume has held fixed: rivalry. A common-pool resource is rival — that was the whole engine of the tragedy, the grass one cow eats being grass another cannot. But information is not rival. A fact, a formula, a melody, a line of code, once made, may be used by one without leaving any less for another; my knowing a theorem subtracts nothing from your knowing it. A knowledge commons is therefore immune to the tragedy of overuse — one cannot overgraze a fact, and the more who draw on an idea the more is built upon it. The open commons of knowledge — the sciences, free and open-source software, the great collaborative encyclopaedia, the public domain — flourish with use rather than failing under it; they are commons in the fullest sense, shared and unfenced, and yet they do not deplete. Here the eleventh volume's positive externality reigns where its negative one cannot, and the danger, as the twelfth volume on information presses, is not overuse at all but its opposite: under-provision, since a non-rival good is hard to be paid for and so may go unmade, and over-enclosure, since the temptation to fence what need not be fenced — by patent, by copyright, by paywall — can starve a commons that thrives only open.

The third inversion is the exact mirror of the tragedy, and it was named by the legal scholar Michael Heller in 1998: the tragedy of the anticommons. The commons is overused because too many hold a right to use it and none a right to exclude; the anticommons is underused because too many hold a right to exclude it and none may act without the others' leave. Fragment a resource into many separate ownership rights, each carrying a veto, and the resource is not overgrazed but gridlocked — left idle because no one can assemble the permissions to put it to use. Heller's first example was the empty storefronts of post-Soviet Moscow, where a single shop might require the assent of a dozen agencies each owning some shred of the right to it, so that the windows stayed dark while flimsy kiosks crowded the pavement before them. His gravest is the patent thicket of the eleventh and twelfth volumes' world — an invention that touches dozens of patents, any one of whose holders may block it, so that useful products go unbuilt for want of a deal among the veto-holders. The anticommons teaches the lesson the fifth plate could not: that the cure for the commons can overshoot into the opposite disease, and that too much private property gridlocks a resource as surely as too little lays it waste.

The figure sets the two failures side by side, and from it the unifying lesson falls out. Begin with a resource held by few and used near its efficient level, and then multiply the holders of rights in it. If the rights are rights to use, with none to exclude — the commons — use climbs above the efficient level into overuse, the tragedy of the early plates. If the rights are rights to exclude, with none able to act alone — the anticommons — use sinks below the efficient level into underuse, Heller's gridlock. The optimum lies between, where rights are neither so absent that all overuse nor so fragmented that none may act. And the moral is one the volume has been circling from the start: property is not a scalar, a quantity of which more is always better. It is a structure — a particular bundling and placing of the rights to use and to exclude — and a structure can fail by excess as by want. The fence of the fifth plate is no universal good, to be applied without limit; it is a tool with an optimum, and beyond that optimum it does not cure the tragedy but inverts it.

Gather the volume. It began at the village pasture, where two properties met that the textbook good keeps apart — rivalry and openness — and made of their meeting the common-pool resource, the unstable compound at the center of all that followed. From the pasture came the tragedy, drawn in full: the herder who reckons the average and ignores the marginal, the rent grazed away to nothing, the ruin worked not by greed but by arithmetic. The fishery turned the parable into a science, and the dilemma of the many turned it into a game whose dominant move is to overuse and whose trap tightens with numbers. Against the failure stood three cures — the fence that privatizes, the Leviathan that regulates, and the governed commons that Ostrom found communities had been quietly keeping all along — each with its charter of conditions and its catalogue of limits: the stranger and the global commons marking where self-governance fails, the club and the congestion charge refining the concept, the comedy and the anticommons showing the tragedy neither universal nor always cured by the fence.

Beneath it all lies a single insight, and it is the one the argument promised. The commons is the failure of the unowned shared resource. Because the resource is rival, each user's withdrawal is a real cost to all the others; because it is non-excludable, no price can charge him for that cost and no gate can keep him from imposing it; and so each user, weighing only his private gain against his private cost, takes more than the whole can bear, and the sum of these rational withdrawals is a ruin that no one willed and all must suffer. It is the externality of the eleventh volume made reciprocal and many-sided — where the externality ran one way, from a polluter to a sufferer, the commons runs every way at once, each user at the same instant the author of the harm and among the harmed. The tragedy of the commons is the externality written in the plural, and the failure it names is the failure of a thing that, belonging to all, is in the only sense that governs conduct the responsibility of none.

Against that failure the discipline sets three instruments, and the volume's hard-won wisdom is that none is a master key. The market can mend the commons by the fence — privatize it, and one owner internalizes what the many externalized. The state can mend it by the Leviathan — regulate it, and public coercion supplies the restraint the users cannot. The community can mend it by self-governance — and here was Ostrom's revolution, the demonstration that users themselves, neither privatizing nor surrendering to the state, govern their commons by rules of their own, turning the one-shot tragedy into a repeated game that neighbors can win. For two centuries the discipline named only the first two and quarreled over which to prefer; Ostrom established the third and dissolved the dichotomy, and the answer she left is not a formula but a judgment — there is no panacea, the right instrument depends on the resource and the people who share it, and often the truest answer is polycentric, a layering of market and state and community, local nested in national nested in global, no single one sufficient and all of them at work.

The volume reaches into the table at every turn. It is, before all, the child of The Externality, whose missing market it suffers in the plural, and whose remedies — the cap-and-trade permit, the congestion charge, the carbon price — are revealed here as commons remedies, property manufactured for a thing that had none. It rests on The Game, for the tragedy is an n-person prisoner's dilemma and its cure a repeated game sustained by the shadow of the future. It is twin to The Norm, for the community's monitoring and its graduated sanctions are social norms enforcing cooperation, and Ostrom's self-governance is the norm volume's mechanism turned upon a resource. It leans on The Rule, for property law and regulation and the Leviathan are the formal entitlements that law-and-economics designs. It turns on The Rate, for whether a private owner conserves or liquidates depends on the discount with which he weighs the future — the intergenerational commons, the climate, hanging on that number. And it bears the mark of The Power, in the local elite who captures a commons and the concentration that enclosure breeds; of The Asymmetry, in the non-rival knowledge commons that cannot be overgrazed but can be under-provided; of The Rent, whose unearned return is here the resource rent that open access dissipates and enclosure captures; and of The Surplus, for that dissipated rent is destroyed surplus, a deadweight loss the sea and the pasture pay.

And with this volume a family closes. The Commons is the fourth and last of The Failures, and with it the third group of the table is complete. Set the four side by side and the architecture of the group stands revealed. The Power was the failure of number — too few hands on one side of the market. The Asymmetry was the failure of knowledge — what one party knew and another did not. The Externality was the failure of the missing market — a cost or a worth with no price to carry it. The Commons is the failure of the unowned shared resource — a thing rival and open, exhausted because it is no one's to keep. Four ways a competitive and lawful market fails with no villain in sight: not through fraud or force, but through structure — the structure of a market cornered, a market blinded, a market missing, a market unowned. The Failures join The Institutions and The Behaviors as the third of the table's six families to be finished entire, and they are, perhaps, its dark heart: the volumes that explain why the elegant machinery of price, which the first eight volumes anatomized, does not always deliver what it promises.

For the deeper unity of the four is that each is a failure of the market's coverage — of its completeness, its reach. The competitive ideal that the early volumes built assumes a world of markets complete and universal: every good owned, every cost priced, every party informed, every seller one among many. The Failures are the four great ways the world falls short of that ideal — a market cornered by the few, a market blinded by unequal knowledge, a market missing where a cost goes uncounted, a market unowned where a resource lies open — and each volume of the group is the study of one shortfall and the imperfect art of its repair. To read them together is to learn the discipline's most adult lesson: that the market is neither a machine that always works nor an idol that always fails, but an institution magnificent within its competence and lost outside it, and that the work of economics is to know, precisely, where the edge of that competence lies.

What, then, is the verdict on the economics of the commons? On the side of triumph, it is among the discipline's most fruitful and most humane achievements. Gordon and Schaefer built the working model of every fishery on earth; Hardin, for all his confusions, named a problem that organizes the whole of environmental policy; Demsetz explained why property itself comes into being; and Ostrom, above all, overturned a false and confining dichotomy and won the Nobel for showing that ordinary people, given the room, govern their commons with a wisdom the theorists had denied them. The field is the foundation of natural-resource economics — of fisheries and forestry and the management of water — and it bears with mounting urgency on the gravest commons of all, the climate. More than any other body of work in the table, it restored the community to a discipline that had recognized only the market and the state, and so enlarged the very idea of how human beings might govern themselves.

And on the side of humility, every cure the volume offers is partial and conditional. Privatization dispossesses and may not conserve; regulation is captured and ill-informed and displaces what worked; self-governance does not scale and can mask an injustice of its own. The models are stylized, the design principles a diagnosis and not a recipe, and the hardest commons — the global atmosphere, with no community to govern it, no world state to regulate it, and no way to fence it — may exceed every instrument the volume holds, as the eleventh volume's plate on climate already confessed. The deepest question of all, how to weigh the resource against the present need and the future against the now, is the discount-rate question the third volume could not settle and this one cannot either. So the volume ends where it began, with a verdict rather than a fact: the commons is the failure of the unowned shared resource, governable by the market, the state, or the community, no one of which is a panacea and the choice among which depends on the resource and the people who share it; the discipline names the failure with precision and supplies a plural, partial, and hard-won set of cures, and leaves to judgment, case by case, which institution can tend what no one owns. The plates hand on to The Game, whose dilemma is the tragedy's engine; to The Norm, whose informal rules are its quietest cure; to The Rule, whose formal law is its loudest; to The Rate, on whose discount the future of every shared resource turns; and back to The Externality, whose plural this volume has been. What men hold in common they may sometimes govern, often save, and never manage by formula. The argument remains open.