Private Order Institutions: The case of the Maghribi Traders

1
Private Order InstitutionsThe case of the
Maghribi Traders

• UBC - Econ 334
• Mauricio Drelichman

2
Last lecture recap

• The Manorial system was the set of institutional
  arrangements that structure rural economic life
  in the Middle Ages.
• Its key elements were
• Open fields.
• Serfdom.
• Input sharing (labor dues).
• North and Thomas argued that these institutional
  arrangements were efficient in the absence of a
  monetary market economy.
• Fenoaltea contended that the problems of the
  Manorial economy could be easily (and more
  efficiently) solved without resorting to serfdom
  and input sharing.

3
Bibliography

• Greif, A. Reputation and Coalitions in Medieval
  Trade Evidence on the Maghribi Traders. Journal
  of Economic History.

4
Impersonal trade

• We all conduct transactions - trade - with people
  we know and trust. We do so because we are
  reasonably certain that they will live up to
  their end of the bargain.
• This type of trade - personal trade - is
  sustained by
• Trust relationships.
• Natural groups (tribes / clans).
• Kinship ties (family).
• Fear of God.
• The development of a market economy, however,
  relies on impersonal trade - a large number of
  transactions with people we never met before and
  will likely never meet again.
• What are the problems of impersonal trade?

5
The prisoners dilemma

• In any economic transaction, at least one party
  will have the option to cheat.
• After enjoying a meal at a restaurant, you could
  walk out without paying the bill (failure to
  deliver payment).
• You might advertise a product in an online
  auction, collect the winning bid and never ship
  the item (failure to deliver goods).
• If there is no way to enforce the completion of
  the transaction or to punish a cheater, trade
  will not take place.
• In game theory, this problem is known as the
  prisoners dilemma.

6
The prisoners dilemma

• Two prisoners are accused of a crime.
• If both of them stay silent, each of them will
  receive a one year sentence.
• If both of them betray each other, each of them
  will receive a five year sentence.
• If one stays silent but the other betrays him,
  the prisoner who stayed silent receives a ten
  year sentence while the one who betrayed him will
  walk free.
• Each prisoner wants to minimize the time he
  serves, and does not care about the other
  prisoner.

7
Payoffs of the game for player 1
Prisoner 2
stay silent
betray
stay silent
Prisoner 1
betray
8
Payoffs of the game for player 2
Prisoner 2
stay silent
betray
stay silent
Prisoner 1
betray
9
One-shot Nash equilibrium

• If the game is played only once, no matter what
  prisoner 2 does, prisoner 1 is always better off
  betraying.
• Similarly, prisoner 2 is always better off
  betraying, no matter what prisoner 1 does.
• That means that both will betray and each will
  get 5 years.
• But if they both could agree to cooperate, they
  would only serve 1 year each!
• This is called a Nash equilibrium - both parties
  end up in a suboptimal situation because they
  cannot commit to play cooperatively.

10
Nash equilibrium vs. social optimum
Prisoner 2
stay silent
betray
stay silent
Prisoner 1
betray
11
Breaking the Nash equilibrium

• If the two players play the same game repeatedly,
  it is possible to break the Nash equilibrium.
  Each has an incentive to cooperate today so that
  their opponent will continue to cooperate in
  future periods.
• If the game is played only once, the only way to
  escape from the Nash equilibrium is to adopt a
  commitment device.
• Impersonal trade resembles the prisoners
  dilemma. You most likely meet your counterparty
  only once, and you only care about your gains.
• Institutions are commitment devices that allow
  traders to commit not to cheat on each other,
  therefore avoiding the Nash equilibrium and
  allowing impersonal trade to take place.

12
Agency problems

• One particular type of Nash equilibrium arises
  when agents are employed to oversee part of a
  business operation.
• The use of agents allowed capitalists and traders
  to realize efficiency gains.
• The use of traveling agents enabled capitalists
  to trade with different locations simultaneously.
• The use of agents residing in remote locations
  allowed for better knowledge of local conditions
  and saved on traveling costs.
• Agency, however, had its own problems
• The agent could run away with the capital or lie
  about the circumstances of a loss.
• The agent could lie about market conditions in
  remote locations.

13
Solving agency problems

• In the absence of a commitment device, agents
  would have always had an incentive to cheat on
  their principals.
• Some trading companies, particularly Italian
  ones, employed only family members as agents.
• Created a repeated game.
• Had a potentially large punishment if the agent
  was caught cheating.
• Greif examines a different type of commitment
  device private-order institutions.
• Public institutions are sustained and enforced by
  the state. Private-order institutions must be
  self-enforcing.

14
Early Mediterranean Trade

• Encompassed a closely knit trading area from
  Southern Spain to Arabia.
• There were few trade barriers and rulers were
  generally supportive.
• Goods and transportation markets were
  competitive.
• There were many sources of uncertainty
• Market volatility.
• Transit time volatility.
• Information could only move as fast as ships.
• Risk of sea and people.

15
The Geniza Documents

• How can we learn about the particularities of
  Medieval trade? Greif finds a valuable source in
  the Geniza documents.
• In the Jewish tradition, documents containing the
  name of God could not be destroyed.
• A Geniza was a room in a synagogue where such
  documents were deposited.
• The Fustat (Old Cairo) Geniza documents have
  survived to the present, and contain a plethora
  of trading contracts.
• Among the Geniza documents, Greif exploits those
  pertaining to the Maghribi traders.

16
(No Transcript)
17
The Maghribi Traders

• The Maghribi traders were a group of Jewish
  traders spread throughout the Mediterranean basin
  in the eleventh century.
• They were largely middle-class, and engaged in
  small and mid-sized trading ventures.
• They were not ethnically or religiously different
  from other Jewish groups (which they could join).
  
• Their trading relationships were supported
  exclusively through an economic, private-order
  institution the coalition.

18
Trading relationships

• The Maghribi traders were sedentary.
• They acted alternatively as principals or agents
  in different ventures.
• If the trader risked his own capital, he acted as
  the principal.
• If the trader managed another traders capital
  and charged for the service, he acted as an
  agent.
• The legal system did not provide a framework to
  enforce contracts, as virtually all ventures
  spanned more than one sovereign territory.
• Trading relationships were based on trust.

19
Reputation mechanisms

• Kinship, religious and group mechanisms enable an
  agent to signal his honesty or fear of God.
• The Maghribi traders shared a religious faith.
  Greif argues that their ventures did not rely on
  this.
• Commitment mechanisms enable an agent to
  establish ex-ante that his most profitable course
  of action is to be honest.
• The Maghribi traders developed an economic
  institutions - the coalition - that enabled
  them to commit ex ante not to cheat on their
  business associates.

20
The coalition

• Every coalition member agrees to employ only
  other coalition members as his agents.
• This limits the supply of available agents, thus
  generating a premium compensation.
• All members agree to never employ a member who
  has cheated on any other coalition member.
• In addition, if a member cheats, all other
  members can cheat back on him without being
  penalized in turn (this is called a cheat the
  cheater strategy).

21
Enforcing the coalition

• Coalition members kept a fluid correspondence
  with other coalition members as part of their
  trading relationships.
• This ensured that each trader was timely informed
  of local market conditions in virtually every
  location where the coalition operated.
• The Maghribi traders used a per-venture
  accounting system, detailing the profit or loss
  of each individual trade.
• The abundant flow of information combined with
  the per-trade accounting system made it easy to
  spot cheaters.

22
Contractual forms in Medieval trade

• Italian merchants relied mostly on the commenda
  and the sea loan
• In a commenda, one partner supplies the capital,
  while the other travels. If the venture is
  successful, they split the gains. If
  unsuccessful, the capitalist bears most of the
  losses.
• A sea loan is a standard loan where the borrower
  bears all the risk, with the exception of some
  insurance provisions (for example, for shipwreck
  or piracy).
• Wealthy capitalists minimized the risk of
  cheating by hiring young partners with low
  outside options.
• The Magrhibi traders relied on partnerships,
  formal friendships, and factor relations
• These contractual forms are much more flexible,
  allowing the partners to alternatively fulfill
  the role of principal or agent.
• The penalty for cheating is external to the
  contractual relationship.

23
Life of the relationship

• Italian merchants created family firms, and sons
  inherited the liabilities of their fathers.
• Creating infinitely-lived firms increased the
  incentives to be honest with your firm (family).
• The Maghribi traders continuously entered and
  left short-lived partnerships with many other
  members of the coalition.
• Repeated games among the same individuals were
  not necessary to generate trust.
• Traders in the same family were normally not in
  partnership with each other.

24
The limits of the coalition

• The Maghribi traders retained their separate
  social identity within local Jewish groups only
  as long as they were engaged in long distance
  trade.
• They did not engage in partnership with other
  Jewish, Christian or Muslim members.
• Coalitions tend to be closed
• The cost of maintaining the flow of information
  increases with the size of the coalition.
• Outsiders require a higher premium if they
  cheat, they cannot by punished by the coalition.