Haute Lumière

Commerce · VI.11 · MMXXVI · daylight

La Bourse  /  Volume VI  /  Nº VI.11

The Limits of Self-Governance

Volume VI — Governance and the Commons

---

THE PLATE

---

THE LETTER

This chapter closes Volume VI, and it is the one to hand to the person in your organisation who thinks all of this is charming and none of it is serious.

Ten chapters have argued that groups of people can govern shared resources without an owner and without a policeman, and that when they do it they often do it better than either. That argument is right, the evidence for it is strong, and Elinor Ostrom won a Nobel Prize for assembling it. Volume VI has tried to give it the apparatus it deserves — the principles, the cases, the audit, the arithmetic of a long table.

Here is the part that has to be said with the same care.

Self-governance has a domain, and outside that domain it does not merely underperform — it cannot be constructed at all. There is a group size beyond which a reputational sanction has no force at any detection rate. There are conflicts of interest that no amount of good facilitation resolves, because the zone of agreement is empty. There are resources whose boundaries cannot be drawn at any expenditure, which means Ostrom's first design principle cannot be satisfied and therefore none of the others can begin. There are processes whose clock runs faster than any deliberative cycle, where the meeting that decides is always a meeting about something that already happened. And there is the case this volume has circled for ten chapters without landing on: a commons working perfectly, deciding well, and imposing the entire cost of its decision on somebody who was not in the room and has no way in.

For each of those five we give a case, a number, and where the line sits. Then we do the harder half, which is to say plainly what the state, the firm and the contract are actually for — not as regrettable necessities but as instruments that discharge functions a commons provably cannot — and to describe the interface between a self-governing unit and a coercive authority as it has actually worked in the places where it has worked.

You will find the strongest version of the case against commons governance in this chapter rather than outside the book. That is deliberate. A reader who has to leave a textbook to find its best critic has been handed a brochure.

— The Editors

---

DISCOVERY

What is already working: the interface, at its best

Begin with the positive core, and here it is not a commons and not a state. It is the seam between them, which is where almost every durable case in this literature actually lives.

Read Ostrom's long-enduring cases again with an eye for the seam and something becomes obvious. The Swiss alp at Törbel has a written charter from 1483 — and a charter is a legal document, filed, enforceable outside the village. The Japanese iriai commons of Hirano, Nagaike and Yamanoka ran for three hundred years — and iriai rights are recognised in the Japanese Civil Code, which is to say that the village's rule has a courthouse behind it. The Tribunal de las Aguas of the Valencian huerta meets on Thursdays at noon at the Apostles' Door of Valencia Cathedral, hears the eight historic canal communities, gives judgment orally and keeps no written record — and it is named in Spanish law as a customary tribunal whose rulings stand. None of the canonical commons is a commons alone. Every one of them is a self-governing body operating inside a jurisdiction that agreed not to overrule it.

The modern cases are built the same way and built on purpose.

Maine's lobster zones. The fishery had informal harbour territories for a century; Acheson's fieldwork found that fishermen in perimeter-defended territories took larger lobsters and more weight per trap than those in contested waters. What Maine did in 1995 was not to invent that. It passed a Zone Management Law creating seven zones, each with a council elected by licence-holders, each able to set trap limits and fishing times by a supermajority of its own members — inside a statutory floor the councils cannot touch: the double gauge, the V-notch, the escape vents. The state set the floor and the fishermen set the ceiling. Landings went from roughly 20 million pounds in the mid-1980s to a record 132.6 million pounds in 2016, a factor of 6.6.

Chile's territorial use rights. The 1991 fisheries law let artisanal caletas hold exclusive management areas for benthic species. Gelcich and colleagues, reporting in PNAS in 2010, found target-species densities significantly higher inside the management areas than in open-access grounds alongside. The state supplied the one thing the divers could not manufacture themselves — a legal right to exclude — and the divers supplied everything else.

Namibia's conservancies. From 1996 the state devolved wildlife rights to registered communal conservancies. Roughly 86 of them now cover on the order of twenty percent of the country. Registration is conditional: a defined boundary, a registered membership, a constitution, a benefit-distribution plan. The conservancy governs; the state holds the registration and can withdraw it.

And the version a treasurer already recognises. A self-regulatory organisation is this exact structure wearing a suit. FINRA writes and enforces the rules for roughly 3,300 broker-dealers, funded by a levy on its own members — and every rule it makes is filed with the Securities and Exchange Commission under Section 19(b) of the Exchange Act, which may approve it, and under 19(c) may abrogate it. Self-governance, with four reserved powers held above it, in a market that moves money in microseconds.

Five cases, one pattern. In each, the self-governing body supplies local knowledge, cheap monitoring, legitimacy and graduated sanction; the coercive authority supplies boundary, standing, a floor, and speed. Neither could do the other's half. That is not a compromise between two ideologies. It is a division of labour between two instruments with different competences, and the rest of this chapter is about where the division falls.

---

THE ARITHMETIC

Five limits, with their numbers, and one honest positive

First, scale — and the limit is sharper than "large groups are harder."

Ostrom's monitoring principle works because sanction is graduated and social. The fourth and fifth principles are a pair: members monitor each other, and the first sanction is shame. A social sanction is worth something only to a violator who cares what the sanctioner thinks, and a person can only track so many relationships. Take Dunbar's figure of 150, and let the reputational cost of a violation dilute as the group exceeds it:

  S(n) = S_max x min(1, D / n)          deterrence needs   p >= g / S(n)

With a violation gaining £500 and full social sanction costing £1,500, deterrence needs a detection probability of 0.333 — easy, in a group where everyone is watching. Now grow the group. At 450 members the sanction has diluted to £500 and deterrence requires detection with certainty. At 1,500 it requires a probability of 3.33, which is not a number.

  n* = D x (S_max / g)

  S_max/g = 2:1   ->  300 members       5:1   ->    750
            3:1   ->  450                10:1  ->  1,500

Above roughly 450 appropriators, at these parameters, a purely reputational sanction cannot deter, even with perfect monitoring. Not "works less well" — cannot, because the inequality has no solution. Something with a fine behind it has to arrive, and a fine is coercion. Agrawal and Goyal's study of forest councils in Kumaon found the same thing from the other side: the relationship between group size and monitoring is curvilinear, and it is the middle-sized councils that fund a third-party guard, because the small ones do not need one and the large ones cannot agree to pay.

The monitoring cost moves the same way. Mutual monitoring scales with the pairs who must stay informed about each other — n(n−1)/2 — while a hired warden scales with the ground. At £4 a year to keep one pair informed against a warden service costing £6,000 standing plus £12 per appropriator, the crossover is at 59 appropriators. Past that the cheapest monitor is one nobody has to be friends with.

Second, heterogeneity — and the honest version, which is not the easy one.

The easy version says diverse groups cannot cooperate, and it is wrong. Varughese and Ostrom's study of eighteen Nepali forest user groups found heterogeneity's effect mediated almost entirely by institutional design; Poteete and Ostrom found the group-size literature genuinely ambiguous. Diverse groups cooperate all the time.

The hard version is different and it is Jack Knight's: institutions are not selected for efficiency, they are settled by bargaining, and where the zone of agreement is empty no procedure creates one. When one subgroup's use destroys the value of the resource for another — the trawler and the hook-and-line boat, the upstream and downstream irrigator, the farmer and the fish — there is no allocation both prefer to no agreement, and deliberation produces a very well-documented deadlock.

The Klamath basin is the case, and the number is the duration. Irrigation to the Klamath Project was shut off in 2001; at least 33,000 adult salmon died in the low, warm river in 2002; deliberative processes ran for two decades and produced a basin-wide agreement in 2010 that Congress declined to authorise and that expired in 2015. The four dams came down in October 2024 — the largest dam removal in history, on the order of $450 million — by way of a regulated utility, two state governments and a federal licensing commission. Twenty-three years of deliberation, resolved by an instrument none of the parties around the table possessed.

Third, boundaries that cannot be drawn at any price.

Ostrom's first principle is clearly defined boundaries — of the resource and of who may use it. It is first because nothing else works without it. Where the boundary is undrawable, the framework does not underperform: it cannot start.

The atmosphere is the standing example. There are on the order of 40,000 tracked objects larger than ten centimetres in Earth orbit and no owner of the orbit either. But notice the move that does work, because it is the constructive half arriving early: you cannot bound the atmosphere, and you do not have to. The EU Emissions Trading System regulates roughly 10,000 installations and reaches about 40 percent of the bloc's emissions — a monitoring ratio on the order of 45,000 citizens to one regulated point. The Montreal Protocol has 197 parties and has phased out about 99 percent of controlled substances, because ozone-depleting substances were made by a countable number of firms and the treaty carried trade measures. Where the same trick was unavailable, the record is what it is: current policies put the world on roughly 2.5 to 3 degrees of warming.

Fourth, the governance clock — and this is the cleanest line in the chapter.

A governing body is a controller, and a controller has a sampling rate. Nyquist and Shannon gave the condition: you must sample at least twice per cycle of the thing you are controlling. Applied to institutions:

                    T_g  <=  T_d / 2

  T_g  the deliberative cycle       T_d  the doubling or turnover time
                                          of the fastest state variable

The line falls between three days and seven years, and every resource a commons has ever governed well sits on the slow side of it. Silicon Valley Bank lost $42 billion on 9 March 2023 — 23.9 percent of deposits in one day, with $100 billion more queued for the following morning, 81 percent of the deposit base inside two days. No assembly convenes at that speed. The resolution came overnight from a receiver with statutory powers, and the alternative to that receiver was not a better meeting.

Fifth, the party with no standing, which is the limit this volume has been walking toward.

American land use is the most participatory institution in the country: public notice, open hearing, any resident may speak. Einstein, Palmer and Glick read thousands of comments across ninety-seven Massachusetts towns and found 63 percent of commenters opposed, and commenters older, whiter and far more likely to own their homes than their neighbourhoods. Hsieh and Moretti estimate that relaxing land-use constraint in three metropolitan areas to the level of the median American city would have left US output 3.7 percent higher — a figure their 2015 working paper put at 9.5 percent and their published 2019 paper revised down by a factor of 2.6. We quote the published number.

The meeting is not corrupt. It is working exactly as designed. The person who would have moved into the building is not a member of the commons, and the commons has no procedure that can hear them.

And now the honest positive, on the sceptic's own metric.

The forester's metric is standing forest per hectare and public expenditure per hectare. Nepal transferred forest to more than 22,000 community forest user groups covering on the order of 1.8 million hectares and 2.9 million households — and forest cover rose from 29.0 percent of land area in the 1994 assessment to 40.36 percent in the 2015 one, a gain of 11.36 percentage points, against a state forest department that had been losing the same ground.

Two honesties belong beside that number and neither of them removes it. The assessments do not use one definition: counting forest plus other wooded land, the series runs 39.6 percent to 44.74 percent — 5.14 points, so 45 percent of the headline gain is genuinely new woody cover and the rest is shrubland maturing into forest. That maturation is precisely what protection from grazing and cutting produces, so it counts, but it is not what "Nepal doubled its forest" implies and we will not print that sentence. And Nepali remittances run at roughly a quarter of GDP; the agricultural abandonment they fund regrows hillsides with no institution involved at all.

Both caveats priced, the community forests still beat the ministry on the ministry's own measure, at a fraction of the ministry's cost per hectare. That is the load-bearing positive in a chapter of limits, and it is why the limits are worth stating precisely rather than generally.

---

DREAM

What becomes ordinary when the seam is designed

Describe it in the present tense, because a dream in the future tense is a wish.

In the jurisdiction that has absorbed this, devolution comes with a floor and a clock, and everyone can read both. A charter granting a group the right to govern a resource states, on its first page, three things: the statutory minima the group may not go below, the class of decision reserved to a faster authority, and the party whose interest the group is constitutionally unable to see, with the name of whoever speaks for them. Nobody experiences this as distrust. It is read the way a pilot reads a minimum safe altitude — as the number that makes the rest of the flight discretionary.

Assemblies size themselves against the arithmetic. A group approaching the reputational ceiling does not discover this at a crisis; it nests, or it hires a warden and writes a fine, and it does so as a matter of ordinary maintenance, because the ceiling is a published figure with the group's own parameters in it.

Decision rules are chosen rather than inherited. A group that wants unanimity computes what unanimity costs it in meetings per decision, writes the figure into the constitution beside the rule, and keeps the rule if it still wants it. Most do not. The ones that do are protecting something specific and can say what.

Emergency powers exist, are used, and expire. The fast authority acts inside hours on the things that move in hours, and its action carries an automatic sunset and a mandatory review in the slow forum, so the group's response to speed is not to give up the slow decisions with it. The ratchet that used to run one way — every crisis leaving a permanent enlargement of the central power — has a return spring, and the spring is written into the same clause as the power.

And the excluded party has a door. A tenant who does not live in the town yet, a cohort not yet born, a downstream village in another watershed, a species: each has somewhere to be represented, not by the group's good intentions but by an office with standing and a budget. Groups do not resent this. They have noticed that the charters with an outside advocate are the ones that survive a change of government, because they are the ones nobody outside can characterise as a cartel.

---

DESIGN

Count the principles

Here is the structure, and here is the cut.

Ostrom's design principles are eight. After Cox, Arnold and Villamayor-Tomás tested them against ninety-one studies and split three, they are eleven. The number of them that requires a group to weigh a cost it imposes on somebody outside its boundary is nought.

Go and check, because that is the claim. Clearly defined boundaries. Congruence between rules and local conditions. Collective-choice arrangements for those affected — those affected within the group. Monitoring by or accountable to the appropriators. Graduated sanctions. Conflict-resolution mechanisms that are rapid and low-cost — for members. Minimal recognition of the right to organise, which is a claim the group makes against the outside. Nested enterprises, which looks up at the larger system's capacity to support the group, not at the group's obligations to it.

Eleven of eleven are about the group's capacity to govern itself. None gives a non-member a voice.

That is not an oversight in the framework and it is not a criticism of it. It is the framework working. Principle one converts an open-access resource into a governable one by defining who may use it, and defining who may use it is identical to defining who may not. The boundary is the engine. The same act that makes the commons capable of governing makes it constitutionally blind to everyone the boundary excludes, and the better the institution is, the more effective an excluder it is. A weak commons leaks. A strong one does not.

Read as a set, the eight principles are a specification for a very effective cartel, and the only thing that distinguishes a beloved commons from a condemned one is whether the excluded party had anywhere else to go.

From which the constructive half follows without embarrassment, because it is now a derivation rather than a concession.

What the state is for. Three functions, none of which a commons can discharge. It gives standing to the absent — the non-member, the minority inside the group, the unborn, the mute — by force, which is the only mechanism that works on a body with no reason to listen. It enforces where reputation has run out, above the ceiling, with a sanction that does not require the sanctioned person to care about the sanctioner. And it acts on the fast clock: Bagehot's rule for a central bank in 1873 — lend freely, at a high rate, on good collateral — is a rule written in advance precisely so that a decision taken in hours does not have to be deliberated in hours.

What the firm is for. Coase's answer still holds: the firm exists where the cost of transacting across a market exceeds the cost of direction inside a hierarchy. Put it in this chapter's terms and it is the same finding — a firm is a device for making a decision when deliberation is too expensive, and the price of the device is that somebody is subordinate. That price is real and should be said out loud rather than dressed. It buys speed, unified command, a residual claimant who bears the downside, and the capacity to hold capital at a duration no assembly will vote for. Hansmann's rule follows: ownership settles on whichever class faces the lowest sum of ownership costs and contracting costs, and where interests are heterogeneous that class is rarely the workers — which is the honest reason the cooperative is rare, and it is not hostility.

What the contract is for. Williamson's answer: it lets strangers cooperate once, on a specific asset, without a relationship and without a reputation. A commons requires repeated play and a shared future; a contract is what you use when there is neither. It is the instrument of the one-shot, the distant and the unlike, and an economy that could not do business with strangers would be a small one.

And what the interface is. Four reserved powers, held above a self-governing unit, and only four:

Everything else is the group's. That is subsidiarity with a floor, and it is the only arrangement in the record that has held for a century in more than one country.

---

DESTINY

How the seam holds when nobody is watching it

Three things sustain an interface, and only three.

The floor is statutory and the ceiling is not. A floor set by agreement drifts down; a floor set in law requires somebody to go and change the law in public. Maine's V-notch is in statute and the trap limit is in the zone council, and that is the right way round.

The reserved powers are used occasionally. A power never exercised is a power that has lapsed in practice, and the first time it is exercised after thirty years it will be litigated as an outrage. Exercise it small, early and visibly.

The advocate for the outside is paid and independent. An unpaid representative of the absent party is a volunteer, and volunteers are polite.

Now the honest part. Here is where the interface fails.

It fails by capture of the self-regulator — the standing charge against every SRO in finance, and it is not baseless; a body funded by its members and staffed from them will discover, without anyone deciding to, that its enforcement is gentle. It fails by the floor being lowered in a technical amendment nobody reads, which is why the floor should be harder to move than the rules above it and usually is not. It fails by the emergency power that never sunsets, where each crisis leaves a permanent enlargement and the slow forum wakes up one year with nothing left to decide. It fails by local capture: Bardhan and Mookherjee's finding is that there is no presumption local government is less captured than national, and where local inequality is high and local accountability weak, devolution moves the rent rather than removing it. And it fails — most often — when the group defends its boundary against the very party the state chartered it to accommodate, and the state, which wrote the charter, finds it has no appetite to enforce the one clause that matters.

Araral's scepticism belongs here in full rather than in a footnote: the long-enduring cases are small, rural, homogeneous and low-value, they are survivors, and nobody knows the denominator. We do not know how many commons were attempted and failed, so we do not know the base rate, and a literature of survivors tells you what survival looks like, not how likely it is. That charge is correct, it is unanswered, and it is the reason this chapter gives thresholds rather than encouragement.

---

DELIGHT

The pleasure of a bounded remit

There is an underrated relief in knowing exactly what is not yours to decide.

Anyone who has sat on a body with unbounded responsibility knows the particular exhaustion of it — every question arrives, every question is yours, and the meeting cannot end because nothing has been ruled out of scope. A charter with a floor and four reserved powers takes most of that away. The group is not diminished by the floor. It is relieved of the questions it was never going to answer well, and what is left is the part it is good at: the ground it knows, the people it knows, the sanction that works because everybody is present.

And there is a quieter pleasure in the boundary stone itself, once it is understood. A well-drawn line is an act of care toward the people inside it and an act of consequence toward the people outside, and a group that has named the second one out loud walks differently. The meetings are shorter. The arguments are about thresholds rather than about who is a good person. Somebody at the table is paid to say and what does this do to the village downstream, and after the first year nobody finds that adversarial; they find it is the question that made the charter survive the election.

The long table is better with a floor under it. It always was.

---

OPERATIONALIZE THIS

At the level of finance

A governance principle with no budget line is a preference. Here is the instrument, in the form a general counsel and a treasurer will both recognise, because it already exists in three industries and two ministries.

The structure: a delegated management charter with a statutory floor, four reserved powers, a member levy and a reversion trigger.

You are not inventing this. You are taking the self-regulatory organisation — FINRA under the Exchange Act, Chile's AMERB, Maine's zone councils, Namibia's gazetted conservancy — and pointing it at your resource.

The mechanics.

• The grant. An exclusive right to manage a defined resource for a defined term. Five to ten years. The right is the consideration; everything the group gives is paid for with it.
• The floor. Statutory minima the charter may not go below, drafted so they are harder to amend than the rules above them. Name them individually. A floor described as "applicable law" is not a floor.
• The four reserved powers. Approval, abrogation, substitution, standing — as in the table above, each with its trigger written and its sunset stated. The substitution power carries an automatic expiry and a mandatory report to the slow forum; without those two clauses it is not an emergency power, it is a transfer.
• The advocate. A named, funded, independent representative of the party the boundary excludes, with a right to be heard and a right to appeal. Budget it at one to three percent of the resource's annual value and put it outside the group's control, because a line the group can cut is not independent whatever the charter says.
• The levy. The group funds its own monitoring. This is the financial heart of the instrument and the reason it clears.
• The reversion trigger. Defined, measurable, automatic: a breach of the floor, a failure to file, a monitoring shortfall below a stated threshold, and the right reverts. Chile's management areas revert on failure to submit a management plan, and the clause is why the plans arrive.

The balance-sheet treatment. The charter is an intangible with a defined useful life equal to its term, amortised over it, and impaired in full on a reversion trigger — which makes the trigger a real number in the accounts rather than a sentence in an appendix. The levy is an operating cost. The reserved substitution power is a contingent liability and should be disclosed, because a counterparty relying on the group's rules is exposed to a power the group does not hold.

The counterparty. The chartering authority first — a ministry, a regulator, a municipality, or inside a firm, the board committee that owns the policy. Take the structure to a lender only once one charter term has completed with the right renewed, because now you are financing a demonstrated institution rather than a proposal.

The number that decides it. One figure, on the front page:

      state enforcement cost at the required detection rate
    −  ( member levy  +  residual oversight  +  advocate )
   ------------------------------------------------------------  >  0
             and       levy per member  <  value of the right

Worked, on a resource of 12,000 hectares where a monitor covers 60 hectares a day, a violation gains £500 and the statutory fine is £1,500. The required detection probability is 0.333, so the state needs 66.7 monitor-days a season at £180 a day — £12,000 a year, or £1.00 a hectare. Under the charter, 240 members supply half a monitor-day each at an opportunity cost of £22 — £2,640 — plus £4,000 of residual oversight: £6,640 a year, or £0.55 a hectare. A saving of £5,360 a year, 44.7 percent, and the levy is £27.67 per member against a right worth £520 — 5.3 percent, one twentieth.

The soft term is the detection probability and it is named rather than buried, because it is the term that can be gamed to nothing. A charter whose members file monitor-days they did not walk costs £6,640 and deters no one. That is what the reversion trigger is for, and it is the clause to negotiate hardest.

The first ninety days.

---

APPRECIATIVE QUESTIONS

Twelve, for a room

Discovery — what is already working

1. Think of a rule this group made itself that has held for years without anyone enforcing it. What made it stick, and who was watching when it did?
2. When has an outside authority done something for us that we genuinely could not have done for ourselves? What exactly did it supply?
3. Where have we successfully handled a conflict with someone outside this group — and what was it about how we did it that worked?

Dream — what becomes possible

1. If everyone affected by our decisions had a way to be heard, what would our meetings sound like, and what would get better first?
2. Imagine our charter is the one other groups copy. What is on its first page?
3. If we knew exactly which decisions were ours and which were not, what would we do with the time we got back?

Design — what we build

1. What is the floor beneath us — the things we could not vote to do even if we all wanted to — and would two of us write that list the same way?
2. Who is harmed by a decision we make well, and what would it take for that person to have a seat and a budget here?
3. At what size does our sanction stop working, and what would we put in its place the month before we get there?

Destiny — how it holds

1. What decision would arrive too fast for us to make, and who would we want to make it — with what sunset?
2. What would the first sign be that our floor had been lowered, and who would notice it first?
3. What would have to be true for this charter to be renewed by people who have never met any of us?

---

WORKS CITED

Ostrom, E. (1990). Governing the Commons: The Evolution of Institutions for Collective Action. Cambridge University Press.

Ostrom, E. (2010). "Beyond Markets and States: Polycentric Governance of Complex Economic Systems." American Economic Review, 100(3), 641–672.

Ostrom, E., Burger, J., Field, C. B., Norgaard, R. B. and Policansky, D. (1999). "Revisiting the Commons: Local Lessons, Global Challenges." Science, 284(5412), 278–282.

Dietz, T., Ostrom, E. and Stern, P. C. (2003). "The Struggle to Govern the Commons." Science, 302(5652), 1907–1912.

Cox, M., Arnold, G. and Villamayor-Tomás, S. (2010). "A Review of Design Principles for Community-based Natural Resource Management." Ecology and Society, 15(4), 38.

Hardin, G. (1968). "The Tragedy of the Commons." Science, 162(3859), 1243–1248.

Hardin, G. (1994). "The Tragedy of the Unmanaged Commons." Trends in Ecology & Evolution, 9(5), 199.

Olson, M. (1965). The Logic of Collective Action: Public Goods and the Theory of Groups. Harvard University Press.

Buchanan, J. M. and Tullock, G. (1962). The Calculus of Consent: Logical Foundations of Constitutional Democracy. University of Michigan Press.

Tullock, G. (1967). "The Welfare Costs of Tariffs, Monopolies, and Theft." Western Economic Journal, 5(3), 224–232.

Niskanen, W. A. (1971). Bureaucracy and Representative Government. Aldine Atherton.

Araral, E. (2014). "Ostrom, Hardin and the Commons: A Critical Appreciation and a Skeptical View." Environmental Science & Policy, 36, 11–23.

Bardhan, P. and Mookherjee, D. (2000). "Capture and Governance at Local and National Levels." American Economic Review, 90(2), 135–139.

Knight, J. (1992). Institutions and Social Conflict. Cambridge University Press.

Baland, J.-M. and Platteau, J.-P. (1996). Halting Degradation of Natural Resources: Is There a Role for Rural Communities? Clarendon Press / FAO.

Varughese, G. and Ostrom, E. (2001). "The Contested Role of Heterogeneity in Collective Action: Some Evidence from Community Forestry in Nepal." World Development, 29(5), 747–765.

Poteete, A. R. and Ostrom, E. (2004). "Heterogeneity, Group Size and Collective Action: The Role of Institutions in Forest Management." Development and Change, 35(3), 435–461.

Agrawal, A. and Goyal, S. (2001). "Group Size and Collective Action: Third-party Monitoring in Common-pool Resources." Comparative Political Studies, 34(1), 63–93.

Agrawal, A. (2001). "Common Property Institutions and Sustainable Governance of Resources." World Development, 29(10), 1649–1672.

Agarwal, B. (2001). "Participatory Exclusions, Community Forestry, and Gender: An Analysis for South Asia and a Conceptual Framework." World Development, 29(10), 1623–1648.

Agarwal, B. (2009). "Gender and Forest Conservation: The Impact of Women's Participation in Community Forest Governance." Ecological Economics, 68(11), 2785–2799.

Chhatre, A. and Agrawal, A. (2008). "Forest Commons and Local Enforcement." Proceedings of the National Academy of Sciences, 105(36), 13286–13291.

Persha, L., Agrawal, A. and Chhatre, A. (2011). "Social and Ecological Synergy: Local Rulemaking, Forest Livelihoods, and Biodiversity Conservation." Science, 331(6024), 1606–1608.

Porter-Bolland, L., Ellis, E. A., Guariguata, M. R., Ruiz-Mallén, I., Negrete-Yankelevich, S. and Reyes-García, V. (2012). "Community Managed Forests and Forest Protected Areas: An Assessment of Their Conservation Effectiveness Across the Tropics." Forest Ecology and Management, 268, 6–17.

Nagendra, H. (2007). "Drivers of Reforestation in Human-dominated Forests." Proceedings of the National Academy of Sciences, 104(39), 15218–15223.

Acheson, J. M. (1988). The Lobster Gangs of Maine. University Press of New England.

Acheson, J. M. (2003). Capturing the Commons: Devising Institutions to Manage the Maine Lobster Industry. University Press of New England.

Steneck, R. S., Hughes, T. P., Cinner, J. E., Adger, W. N., Arnold, S. N., Berkes, F., Boudreau, S. A., Brown, K., Folke, C., Gunderson, L., Olsson, P., Scheffer, M., Stephenson, E., Walker, B., Wilson, J. and Worm, B. (2011). "Creation of a Gilded Trap by the High Economic Value of the Maine Lobster Fishery." Conservation Biology, 25(5), 904–912.

Gelcich, S., Hughes, T. P., Olsson, P., Folke, C., Defeo, O., Fernández, M., Foale, S., Gunderson, L. H., Rodríguez-Sickert, C., Scheffer, M., Steneck, R. S. and Castilla, J. C. (2010). "Navigating Transformations in Governance of Chilean Marine Coastal Resources." Proceedings of the National Academy of Sciences, 107(39), 16794–16799.

Coase, R. H. (1937). "The Nature of the Firm." Economica, 4(16), 386–405.

Alchian, A. A. and Demsetz, H. (1972). "Production, Information Costs, and Economic Organization." American Economic Review, 62(5), 777–795.

Williamson, O. E. (1985). The Economic Institutions of Capitalism. Free Press.

Hansmann, H. (1996). The Ownership of Enterprise. Harvard University Press.

Hirschman, A. O. (1970). Exit, Voice, and Loyalty: Responses to Decline in Firms, Organizations, and States. Harvard University Press.

Weber, M. (1919). Politik als Beruf. Duncker & Humblot. Translated as "Politics as a Vocation."

Bagehot, W. (1873). Lombard Street: A Description of the Money Market. Henry S. King.

Dunbar, R. I. M. (1992). "Neocortex Size as a Constraint on Group Size in Primates." Journal of Human Evolution, 22(6), 469–493.

Dunbar, R. I. M. (1993). "Coevolution of Neocortical Size, Group Size and Language in Humans." Behavioral and Brain Sciences, 16(4), 681–694.

Shannon, C. E. (1949). "Communication in the Presence of Noise." Proceedings of the IRE, 37(1), 10–21.

Hsieh, C.-T. and Moretti, E. (2019). "Housing Constraints and Spatial Misallocation." American Economic Journal: Macroeconomics, 11(2), 1–39.

Einstein, K. L., Palmer, M. and Glick, D. M. (2019). Neighborhood Defenders: Participatory Politics and America's Housing Crisis. Cambridge University Press.

Ogilvie, S. (2019). The European Guilds: An Economic Analysis. Princeton University Press.

Ostrom, E., Parks, R. B. and Whitaker, G. P. (1978). Patterns of Metropolitan Policing. Ballinger.

Efficiency Valuation Organization. International Performance Measurement and Verification Protocol (IPMVP), Core Concepts. Successive editions.

Note on figures. The reputational ceiling, the governance clock, the Buchanan–Tullock decision-cost optimum, the monitoring crossover, the Nepali forest series and the charter's decision inequality are all computed in lib/verify/VI_11.py and reproducible there, with every input labelled as sourced, illustrative or assumed. The Nepali assessments of 1994 and 2015 do not use one definition of forest; both series are computed and both are quoted. Where this chapter uses a figure that a later paper revised — Hsieh and Moretti — the published figure is used and the revision is stated.