11 · Ronald Coase · Elinor Ostrom · Douglass North · Hernando de Soto

Property & institutions

Earlier chapters take markets as given. This one asks what they run on. Ronald Coase, a British economist at Chicago, asked why firms exist (1937) and what happens when one person's activity harms another (1960); his answer in both cases was transaction costs. Garrett Hardin warned in 1968 that shared resources get destroyed; Elinor Ostrom, the first woman to win the economics Nobel, spent decades documenting communities that didn't destroy them. Douglass North argued that institutions, the rules of the game, explain why some societies grow rich, and Hernando de Soto showed how much wealth the poor hold in assets the law can't see. Each model below lets you change the rules and watch behaviour follow.

The Coase theorem: the rancher, the farmer and the cost of a deal

coase-theoremcoase-transaction-costs
A rancher's cattle stray onto a farmer's fields. Each extra steer earns the rancher less and tramples more crop. Who should have the right: the farmer to be free of cattle, or the rancher to let them roam? Coase's surprising answer is that if bargaining were free, it wouldn't matter for how many cattle there are. The interesting case is when it isn't free.
Legal right

The ranch and the farm

deal struck
ranchfarm#1#2#6#7#8herd 5 · efficient herd 5 · crop lost $15 · rancher pays farmer $30
Herd size
5
efficient
Total value (pie)
$130
best possible $130
Rancher ends with
$15
Farmer ends with
$115
Under the other rule
herd 5
same herd, different payments

Herd size as deal costs rise, under each assignment of the right

right to farmerright to rancher
02468$0$10$20$30$40transaction cost ($)herdefficientyou

Try: with transaction cost at $0, flip the legal right: the herd stays at 5 and only the payment changes direction. Now raise the transaction cost past $6 and flip again: the deal stops happening under one rule and the starting assignment decides the outcome.

Pigou vs Coase: smoke, laundries and who should adjust

pigou-externalitypigouvian-taxcoase-reciprocal-harm
Pigou's answer to a smoky factory was a tax equal to the damage. Coase pointed out that harm runs both ways: laundries could dry indoors, and sometimes that is the cheaper fix. Bargaining finds the cheapest fix if the parties are few; with many parties, deals get too costly. Compare four legal rules on the same numbers.

Factory and laundries

deal cost $3
factory · output 6.0dryers installed ($12)

Tax = damage per unit. Factory cuts output even if dryers were the cheaper fix.

Social surplus
$30.0
best possible $38.0
Cheapest fix
dryers
Output
6.0
of 10
Lost vs best
$8.0

Social surplus under each rule (same parameters)

No remedy
$38.0
dryers
Pigouvian tax
$30.0
dryers
Coase: right to emit
$38.0
dryers
Coase: right to clean air
$35.0
dryers

Try: at the defaults the laundries can avoid the harm for $12, so the Pigouvian tax cuts output for nothing (Coase's point: harm is reciprocal). Raise the dryer cost to $40: now cutting output is right, and with 3 laundries the Coasean deal finds it. Raise the laundries to 40 and the deal costs swamp the gain; the tax (or making the factory liable) does better than waiting for a deal.

If markets are so good at coordination, why do firms plan internally by command? Because using the market costs something. Activities that are specific to the firm are expensive to contract for; managing more activities gets harder as the firm grows. The firm ends where the two costs meet, and that line moves with technology.

Where the firm ends

in-house 11 / 20
cost of buying it (market)cost of managing it (firm)
inside the firm0123405101520activities, from firm-specific to commoditycost per periodyour boundary
Least-cost firm size
11 activities
Total cost
22.1
least possible 22.1
Wasted by wrong boundary
0.0
Core R&D
make · 0.50
Product design
make · 0.62
Custom tooling
make · 0.74
Final assembly
make · 0.86
Quality control
make · 0.98
Key components
make · 1.10
Major-client sales
make · 1.22
Strategy
make · 1.34
Engineering support
make · 1.46
Brand marketing
make · 1.58
Machining
make · 1.70
Logistics
buy · 1.67
Payroll
buy · 1.53
IT support
buy · 1.39
Legal
buy · 1.25
Advertising
buy · 1.12
Recruiting
buy · 0.98
Catering
buy · 0.84
Cleaning
buy · 0.70
Office supplies
buy · 0.56

Try: step through the presets. Cheaper management (1980s) lets the firm grow; then cheaper market coordination (the internet) makes it shrink back and contract out payroll, IT and logistics. Drag the boundary off the crossing and watch the wasted cost.

A fishery open to all: each boat keeps every fish it catches but shares the cost of a smaller stock with everyone. Boats keep adding effort while fishing still pays, and the stock is driven past the point where it can recover. Then try the alternatives: one owner, tradable catch shares, or a community that sets and polices its own quota.

The fishery · year 0

healthy
stock 900 / 1000 · catch 0 this year
Fish stock
90%
collapse below 10%
Catch this year
0
max sustainable ≈ 158
Profit this year
0
price × catch − cost of effort
Profit to date
0
Boats fishing
12 / 12

Stock and catch over time

fish stockcatch × 4
05001k020406080100yearstock for max sustainable yieldcollapse threshold

Try: play open access and watch boats race for fish until the stock crosses the threshold and never comes back. Reset and pick a sole owner, ITQs or community rules. Then try switching from open access to quotas mid-run: in time it saves the fishery, too late it doesn't. Under community rules, turn monitoring down to zero.

Harold Demsetz (1967) argued that property rights appear when they start to pay. Among the Montagnais of Labrador, hunting grounds were open to all until the fur trade made beaver valuable; then families began marking and defending their own territories. The tragedy is not permanent: when depletion gets costly enough, people invent ways to exclude and to ration.

Ostrom's design principles: what long-lived commons have in common

ostrom-design-principlesostrom-polycentric-governanceopen-access-vs-common-property
Ostrom studied commons that had been governed by their users for centuries (Swiss alpine meadows, Spanish irrigation huertas, Japanese village forests) and those that failed. She found eight recurring features. Here 48 simulated commons face the same weather shocks; each principle you remove opens a known route to failure.

Design principles · 6/8

48 commons, same shocks · year 80

✕L
✕L
✕
✕
✕L
✕
✕L
✕
✕
✕
✕L
✕L
✕
✕
✕L
✕L
✕
✕L
✕L
✕
✕
✕
✕
✕
✕
✕L
✕L
✕
✕
✕
✕L
✕L
✕L
✕L
✕
✕L
✕L
✕
✕
✕L
✕
✕
✕
✕
✕
✕L
✕L
✕L

bar = resource level · ✕ resource collapsed · ⚡ rules broke down (users stopped cooperating) · L = large system

Still governed
0/48
Collapsed
48
Rules broke down
0

Share of commons still working

your principlesall eight
0%50%100%020406080year

Try: this starts with monitoring and graduated sanctions off: switch them on and watch survival jump. Then drop one principle at a time. No single principle is enough by itself (click "none", then add any one), which matches Ostrom's finding that robust commons had most of them.

De Soto's dead capital: a house you can't borrow against

desoto-dead-capitaldesoto-extralegalitydesoto-formal-property
Families in this neighbourhood own their homes but have no recorded title, so banks won't lend against them, and good business ideas go unfunded or are financed by moneylenders at 50%. Give them a title registry and see what the same houses can do. Then make the registry as slow as the real ones de Soto measured.
Formal property records

A neighbourhood of 48 homes · year 20

dashed house = no title (dead capital) · solid green = titled, usable as collateral · gold bar = business capital

Homes with title
0/48
Dead capital
$424k
100% of housing value
Bank credit this year
$0
Business capital
$64k
vs $64k with no titles
Profits this year
$17k

Business capital in the neighbourhood

your settingsno titlesfree, instant titles
$0$200k$400k$600k05101520year

De Soto's estimates (2000)

extralegal real estate, US$ billions
Haiti
$5.2bn
Peru
$74bn
Philippines
$133bn
Egypt
$240bn
years to get legal title or permission, roughly
Haiti
19 yrs
65+ steps
Peru
7 yrs
207 steps
Philippines
19 yrs
168 steps
Egypt
10 yrs
77 steps

Try: switch the title registry on with a cheap process: most homes get titled and bank credit flows to families with good projects. Now push the cost towards 289 days (de Soto's Lima experiment for registering one small workshop) and watch households rationally stay informal.

The evidence is mixed in an instructive way. Titling in Buenos Aires and Peru raised investment in homes, tenure security and (in Peru) work outside the home, since people no longer had to stay home to guard their claim. The jump in bank credit was smaller than de Soto expected: collateral helps only if courts can actually enforce it, which brings us to North.
Forty traders on a ring. Trading with distant strangers brings bigger gains from specialization, but strangers may cheat and searching costs more. Kin and neighbours are honest because reputations stick. Courts make strangers trustworthy; a predatory ruler skims the gains. Each trader picks the partner distance with the best expected payoff.

Who trades with whom · round 0

personal exchange
press play
expected payoff of a trade, by distance to partner
kin & neighboursdistant strangers →
Market reach
2 steps
only kin & neighbours
Expected gain per trade
1.04
Cheated last round
–
Income per head
0.00
avg of last 20 rounds

Income per head, round by round

012010203040round

Try: in the village, everyone trades only with neighbours: safe but small gains. Raise contract enforcement past about 70% and trade jumps to distant strangers with much bigger gains. Then raise expropriation: even with good courts, a ruler who takes the gains shrinks the market back.

“Institutions are the rules of the game in a society.” (Douglass North, 1990). North and Weingast argued that after the Glorious Revolution of 1688 the English crown could no longer seize property or default at will, and its borrowing costs fell. The model shows the same tension: a state strong enough to enforce contracts is also strong enough to expropriate.
  • takeawayTransaction costs decide. With free bargaining, who holds a right changes only who pays. Real deals cost something, so the law's first assignment of rights, and firms' boundaries, shape what actually happens.
  • takeawayThe tragedy is open access, not sharing. Owners, tradable quotas and well-designed community rules all conserve; Ostrom showed that users themselves often design the best rules.
  • takeawayRules make markets. Secure, recorded property and enforced contracts let people trade with strangers and borrow against what they own; without them, wealth stays small, local and dead.

Key concepts · 20

Ronald Coase · Elinor Ostrom · Douglass North · Hernando de Soto
  1. Coase theorem

    If rights are clearly defined and bargaining is costless, people bargain to the efficient outcome whoever holds the right; the law only changes who pays whom.

    Coase never claimed transaction costs are zero. His point was the reverse: because they are positive in real life, how the law assigns rights does change what happens, so it should be judged by the outcome it produces.

    Ronald Coase, The Problem of Social Cost (1960)

    Rancher and farmer (Coase 1960)Sturges v Bridgman (1879): confectioner and doctor
    ↑ see it in the visualization
  2. Transaction costs

    The costs of using markets: finding a trading partner, negotiating, writing a contract, checking it is kept and enforcing it.

    They are the friction behind most of this chapter. They explain why firms exist, why some harms are never bargained away, and why institutions that make strangers trustworthy are worth so much.

    Coase, The Nature of the Firm (1937); The Problem of Social Cost (1960)

    ↑ see it in the visualization
  3. The reciprocal nature of harm

    A conflict of uses has two sides: stopping the factory harms the factory just as smoke harms the laundry.

    The question is not 'who is to blame?' but 'which arrangement produces the most value?'. Sometimes the cheapest fix is on the victim's side (indoor dryers, moving away), and a rule that only restrains the 'polluter' can destroy value.

    Coase, The Problem of Social Cost (1960)

    ↑ see it in the visualization
  4. Externality

    A cost or benefit of an activity that falls on people who are not party to it, such as smoke from a chimney.

    Arthur Pigou argued that markets over-produce activities with external costs. Coase reframed the problem as one of missing or unclear property rights plus the cost of bargaining over them.

    A. C. Pigou, The Economics of Welfare (1920)

    Factory smokeSpark-throwing railways and farmers' crops
    ↑ see it in the visualization
  5. Pigouvian tax

    A tax on each unit of a harmful activity equal to the damage it does to others.

    It needs no negotiation, which makes it attractive when the people harmed are many. Coase's objections: the regulator must know the damage, and a tax that ignores cheaper avoidance by the victims can reduce total value.

    A. C. Pigou, The Economics of Welfare (1920)

    Carbon taxesCongestion charges (London 2003)
    ↑ see it in the visualization
  6. The nature of the firm

    Firms exist because directing work by command is sometimes cheaper than buying each piece on the market; a firm grows until organizing one more activity inside costs as much as buying it.

    It explains why the economy is islands of planning in a sea of markets, and why firm boundaries move when communication and management technology change.

    Ronald Coase, The Nature of the Firm (1937)

    Ford's River Rouge plant (vertical integration)Outsourcing and the gig economy
    ↑ see it in the visualization
  7. Asset specificity

    How much an investment is worth only within one particular relationship, such as a die made for one car model.

    Oliver Williamson built on Coase: the more specific the asset, the more a contract partner can hold you up once you've invested, so such activities tend to be brought inside the firm.

    Oliver Williamson, Markets and Hierarchies (1975)

    General Motors and Fisher Body (1926)
    ↑ see it in the visualization
  8. Tragedy of the commons

    When a resource is open to all, each user gets the full gain from taking more and bears only a fraction of the depletion, so the resource is over-used, sometimes to collapse.

    Hardin's 1968 essay made it famous (the idea goes back to Aristotle and William Forster Lloyd). His remedies were private property or government control; Ostrom later showed a third way.

    Garrett Hardin, The Tragedy of the Commons, Science (1968)

    Grand Banks cod collapse (1992)Atlantic bluefin tunaGroundwater overdraft
    ↑ see it in the visualization
  9. Open access vs common property

    Open access means no one can be excluded; common property means a defined group owns a resource jointly and can exclude outsiders.

    Hardin's tragedy is really a tragedy of open access. A community that owns and governs a commons can manage it well, as many have for centuries.

    Ostrom, Governing the Commons (1990)

    Swiss alpine meadows of TörbelJapanese village forests (iriaichi)
    ↑ see it in the visualization
  10. Emergence of property rights

    Property rights appear when the gains from internalizing a cost or benefit grow larger than the cost of defining and enforcing the right.

    Demsetz's example: when the fur trade raised the value of beaver, the Montagnais of Labrador developed family hunting territories. Property is an evolving answer to scarcity, not a fixed legal fact.

    Harold Demsetz, Toward a Theory of Property Rights (1967)

    Montagnais beaver territories (18th c.)Barbed wire on the American plains (1870s)
    ↑ see it in the visualization
  11. Individual transferable quotas

    A total allowable catch split into shares that fishers own and can trade.

    A share in a fixed total turns a race for fish into ownership of a future stream of catch, so fishers gain from a healthy stock and the least efficient boats sell out.

    New Zealand (1986)Iceland (1990)Alaska halibut: the 'derby' ended in 1995
    ↑ see it in the visualization
  12. Ostrom's design principles

    Eight features shared by long-lasting, self-governed commons: clear boundaries, rules fit to local conditions, users make the rules, monitoring, graduated sanctions, cheap conflict resolution, recognition by outside authorities, and nested layers.

    They came from comparing real cases that lasted for centuries with ones that failed. They are not a blueprint but a checklist of what working arrangements tend to have.

    Elinor Ostrom, Governing the Commons (1990)

    Valencia's huerta water tribunal (meets weekly since the Middle Ages)Philippine zanjera irrigationMaine lobster harbour gangs
    ↑ see it in the visualization
  13. Polycentric governance

    Many overlapping centres of decision (local, regional, national) instead of one top-down authority.

    Local users know the resource and can adapt the rules; higher levels handle what crosses boundaries. Ostrom shared the 2009 Nobel Prize for showing this works better than either pure market or pure state solutions in many settings.

    Vincent & Elinor Ostrom; Ostrom, Nobel lecture (2009)

    ↑ see it in the visualization
  14. Dead capital

    Assets the poor already own, mainly homes and land, that cannot be used as collateral, sold widely or invested because they lack formal legal title.

    De Soto estimated it at about $9.3 trillion worldwide in 2000. The problem isn't a lack of savings but that the savings are held in a form the legal system can't see. Later studies (e.g. in Peru and Argentina) found titling raised investment and security, with smaller credit effects than he predicted.

    Hernando de Soto, The Mystery of Capital (2000)

    Egypt: ~$240 bn of extralegal real estatePeru's titling program (1990s)
    ↑ see it in the visualization
  15. Extralegality

    Living and working outside the formal legal system because getting into it costs too much.

    De Soto's team spent 289 days, six hours a day, legally registering a one-worker garment workshop in Lima. When legality costs that much, informality is rational, and the cost is paid by the poor.

    Hernando de Soto, The Other Path (1986)

    Lima workshop experiment (1980s)Informal settlements around Cairo and Manila
    ↑ see it in the visualization
  16. Formal property system

    Public records that describe who owns what in a standard way, so assets can be pledged, divided, sold and traced by strangers.

    In de Soto's account, the paper representation is what turns a house into capital: it gives the owner an address, accountability and a way to borrow against it.

    Hernando de Soto, The Mystery of Capital (2000)

    ↑ see it in the visualization
  17. Institutions as rules of the game

    Institutions are the humanly devised constraints that structure interaction: formal rules (laws, constitutions), informal norms, and how they are enforced.

    North argued they are the main reason some societies grow and others don't: they set the payoff to producing versus predating, and they shape what organizations form and learn.

    Douglass North, Institutions, Institutional Change and Economic Performance (1990)

    ↑ see it in the visualization
  18. Impersonal exchange

    Trade with strangers you may never meet again, as opposed to personal exchange among kin and neighbours kept honest by reputation.

    Big gains from specialization need large markets, which means dealing with strangers. That requires third-party enforcement of contracts, which most societies in history never achieved.

    North, Institutions, Institutional Change and Economic Performance (1990)

    Medieval Maghribi traders' coalitionsEnglish common law courts
    ↑ see it in the visualization
  19. Credible commitment

    A state's ability to bind itself not to seize property or default on debts.

    North and Weingast argued that after 1688, Parliament's control over taxes made the English crown's promises believable, and interest rates on government debt fell sharply. A state strong enough to protect property is also strong enough to take it.

    North & Weingast, Constitutions and Commitment (1989)

    Glorious Revolution (1688)Founding of the Bank of England (1694)
    ↑ see it in the visualization
  20. Inclusive vs extractive institutions

    Inclusive institutions protect property and open markets and politics to many; extractive ones concentrate power and income in a narrow elite.

    Acemoglu and Robinson use paired cases (the two Nogales, the two Koreas) to argue that institutions, not geography or culture, mainly explain why nations fail. Critics question how far the categories can be measured.

    Acemoglu & Robinson, Why Nations Fail (2012)

    North vs South KoreaNogales, Arizona vs Nogales, Sonora
    ↑ see it in the visualization