IAP course held by Wes Chow and


The Benefits of Cooperation (Heath, Joseph, 2006)

no effort to categorize the benefits from cooperation. catallactic bias (benefits from exchange1)

defines efficiency in such a way that only gains from trade count as efficiency-promoting.

why is this bad? \(\leftarrow\) serious misunderstanding of the contemporary welfare state.

all the efficiency comes from the market. Therefore, the states main function will to promote or support the market. This includes, ensuring property rights, preventing collusion and other anti-competive practices, regulating the price when a monopoly happends, internalizing externalities (taxation).. everything to maintain the competiveness.

Everything else is “redistributive” -> social safety nets and pension insurance are all “redistributive”. Yet we can see this as mechanism to gain benefits.

The five categories of mechanisms in cooperation benefits

  1. economies of scale

    This is the benefit where adding another one to work gives disproportionate outcomes.

    example: two big rocks, two persons. One rock takes two people for a day to move it out of the way. If they don’t work together, the rocks will not be moved indefinitely. If they work together, they will have the path cleared in two days.

  2. gains from trade

    People have different needs and abilities. Benefits from rearrangement.

    example: The elder person may not require much food but may be disadvantaged by obtaining food from hunting but has knowledge how the meat should be treated. A second might need a lot of food

  3. risk pooling
  4. self-binding
  5. information transmission

Week 1


Reflections on the Commons

Refund Bonuses (Tabarrok)


The Use of Knowledge in Society

This was written in 1945. Right before Warren Weaver’s “Science and Complexity”. Complex System acknowledge the fact that small partial changes affect the whole in a unpredictable way.

  • part I

    Economic models often assume information is complete and readily available, but in reality, information is fragmented and dispersed.

  • part II

    Traditional planning centralizes information for decision-making, assuming this is more efficient than everyone acquiring complete information independently.

  • part III

    Hayek identifies two key types of knowledge: broad, scientific knowledge and specific, localized knowledge. The challenge is to disseminate this localized, fragmented knowledge effectively. His central concern for this article is this: (Hayek, Friedrich August, 2013)

    … the method by which such knowledge can be made as widely available as possible is precisely to which we have to find an answer.

  • part IV

    Economic problems arise when there are consequences of change, its the economic analysis job to handle this, and the n

    But these fragmented information is underrepesented.

    There is another type of knowledge is harder to recognize from a statistical standpoint. To be efficient, you will eventually need to rely on the “man on the spot.”

    Economic problems often stem from changes that impact the distribution and utilization of fragmented information, emphasizing the importance of ’the man on the spot’ who possesses contextual knowledge.

  • part V

    We can’t solve this in a centralized manner, it has to be decentralized. More than that, we need some mechanism that while it is decentralized, the information gets passed out that enough information arrives to the “man on the spot” with rich local information. “Enough information” that he can compare which is easier to fulfil his needs, you don’t need anything less or more.

    How to solve this problem, guess what its the price system. Individuals can compare the equal value point by combining the price system and her own preference (Marginal Rates of Substitution).

    If some where in the exchange chain a price difference occurs, this will propagate to the subsequent chain, in effect passing forward the information. It is decentralized and passes information.

  • part VI

    In fact the price system’s prime function is to pass on information. In this case fluid price system are efficient. This thing, the price system is so radical that you can’t imagine this is done by a central authority mannually. And we don’t even need to know more, or how this price system works.

    People might think this was magically invented because of our advancement of civilization through division of labour, but it’s the other way around, we happend to find this price system which lead to our advancment and derived the concept of dividing labour. All that we know is that no one person can come up with an alternative system.

  • part VII

    We can’t live without the price system, and it’s fortunate that this is outside of political debate. John Scumpeter got it wrong that the end consumers evaluation (demand) is the primary cause of how much it is produces. This is wrong even if we reword that the consumers demand impiles the production of a good. Because there is no implication (information) of the exchange chain, and it is impoosible to put all implication into one head.

    System equilibration analysis are useful, but we shouldn’t over credit it.

Review of Wealth of Nations

Week 2


Strength of Weak Ties

  • Strong ties -> Triads
  • Weak ties -> non-triad, bridges

Toward a Connected Society

Collective Dynamics of small-world networks

a variable p is set to re-route regular networks. 1 Different characterisics arise even you have very small p.

two metrics: L and C. L is the average the number of edge counts for all shortest paths to all two pairs. C is the defined as k

The Unbounded Savannah

Groups & Associations

Bowling Alone

A Theory of Groups and Organizations

Week 3


Optimized Democracy Lecture 1: Voting rules

  • Plurality

    highly problematic, -> used ranking to counter argument

  • ballot types
    • Rankings, Approvals, Scores/stars
  • Borda
  • Single Transferable Voting (alternative vote, ranked choice voting)
    • Cambridge and Oxford were the only district that supported to switch to STV.

      This is not Monotonic voting higher for a candidate, some times hurt that candidate.

  • Condorcet Paradox
    • cyclic preferences 1
    • condorcet consistent rules -> if there is a strong alternative that is consistently stronger in pairwise winner for every voter, that one should be selected
  • Llull’s rule
    • Condorset consistent
  • Dodgson

    close to condorcet, but he didn’t plagerasim

    “distance to the condorset” -> minimum score is the winner.

    This is NP-hard

    count how many swaps is needed to make an alternative a condorset winner.

  • Schulze’s rule
  • Awesome Example

    The aggregational rules disagree to each other.

Optimized Democracy Lecture 2: Voting models

Social Welfare function & Social Choice function

  • Simple axioms (properties that satisfy)
    • Anonymity

      The outcome will not change if we swap the voters.

    • Neutrality

      if we swap the options, the result will be swapped accordingly

    • Monotonicty

      pushing and option will not hurt that option.

    • May’s theorm
      • Is the three axioms independant?

        are they any redunduncy?

    • Extend May’s Theorm to multiple choices ?

      we take pairs of the options (a and b, b and c, and a and c) and check the majority of each case and see the winner.

      -> You face Condorset’s dillima. (cyclical preference)

      This will not give a SWF.



this holds true with liquid democracy

  • Plurarity and Borda Counts doesn’t hold condorcet consistency

Date: 2024-01-10 Wed 09:38