Economics of organization

The question is difficult, at many levels. In short, I will argue that it all boils down to the nature of shared understandings, beliefs, and values, i.e., to culture.

Hall and Jones (1999) find that "More than half of the effect of social infrastructure that is indicated by Figure 1 remains after controlling for endogeneity."

Ellingsen och forskning tycks verkligen tala för att det finns ett samband.

Tabellini (2010) "Does culture have a causal effect on economic development? The data on European regions suggest that it does."

etc.

• Management practice affect output of firms,
• management styles are national (e.g. US vs Sweden).

Inherited management is bad management (especially eldest son). Als government-owned firms are badly managed.

Human capital intensive firms are well managed.

Lighter labor market regulation is associated with stronger monetary incentives.

Product market competition is associated with good practices.

Methodological individualism: individuals act. We need to understand individuals' goals and beliefs.

Social structures: Individuals act in context. We need to understand shared conceptions (culture) and how they change (cultural entrepreneurs).

Game theory: Need a formal theory of social interaction. However, models are used simply as illustrations to keep our concepts clear.

• Här måste vi göra antaganden om "complete and perfect information". Så om player 1 känner till player 2 payoffs, så kan man genom backward induction anta att player 2 kommer att välja att "accomodate" eftersom det är sub-game perfect equilibrium (1 > -1 för player 2).

Assumptions:

• Players are maximizing their utility
• Players have perfect and complete information.

Method: If we look for the equilibrium of this game, considered as a whole, we find that Up-Left is a Nash equilibrium (red). However, it’s not a perfect equilibrium. In order to find the subgame-perfect equilibrium, we must do a backwards induction, starting at the last move of the game, then proceed to the second to last move, and so on. In this particular case, we know that player 2 will choose Left if player 1 goes Up, and Right if player 1 goes Down, since these are the moves that maximise his utility. Because there is complete information (and therefore each player’s payoffs are known), player 1 knows these choices in advance, and will therefore choose to go Down, because the payoff will be greater. Therefore, Down-Right is the perfect subgame equilibrium (green).

• Note: LL means L in every case (unconditional L), LR means L if T and R if be (conditional R and L), RL means R if T and L if B (conditional L and R), RR means R in every case (unconditional R).
• This is the same game as a tree.

Here we do not assume perfect information but rather trust an assigned probability.

In a perfect Bayesian equilibrium Player 1 enters if probability that the opponent (Player 2) is weak, p > 1/2. Player 2 fights if strong and accomodates if weak. If player 2 is strong he will get a payoff of 2 (rather than -1) if he fights.

• -1,-1 is the optimal outcome, but rational behaviour will make each individual deviate from this strategy.

Interpretation: The prisoner's dilemma is a paradox in decision analysis in which two individuals acting in their own self-interest pursue a course of action that does not result in the ideal outcome. The typical prisoner's dilemma is set up in such a way that both parties choose to protect themselves at the expense of the other participant. As a result of following a purely logical thought process, both participants find themselves in a worse state than if they had cooperated with each other in the decision-making process.

Solution: Tit-for-tat strategy in an iterative prisoner's dilemma game would solve the situation. We could then chose to replicate the other person's strategy. If prisoner A cooperated last time, we will cooperate again, and so forth. This is in line with reciprocal altruism in biology and a reason why we are in fact cooperating more than conflicting in society.

See also Ostrom, Basu (recordkeeping, reciprocity, reputation).

• so gamma, δ, is how sensitive we are as individuals to inequalities.

• And where:

The inequality, m^- is defined as the difference of one person maximizing her own material payoff and one other person minimizing.

• where δ is the society's sensitivity to inequality.

So in this case, the material outcomes are less valued the more they come from inequality. That is, if 1% owns everything and 99% are living off of nothing, that is not perceived as as valueable to society as 99% sharing most and 1% being very rich. etc.

• Suppose players behave unselfishly because they feel guilty otherwise.
• Preferences are affected by the culture’s (see: recurring figure) values and norms (civility):

• Dictator situation. Endowment is 100; the amount kept is k.

• Player 1 is the divider, so:
• s_1 = m_1 = k
• and 
• m_2 = 100 - k

• The payoff for person 1 = k
• The payoff for person 2 = 100 - k

• 1. Prescriptive norms
• How people ought to behave. Deviating from this might cause "social pressure costs", associated with gamma, γ.

• 2. Descriptive norms
• How people tend to behave.

In this course, we are using norms = prescriptive norms.

We can use prescriptive norms as an explanation to why people are not going for the maximizing/rational approach. Because there might be social pressure costs/guilt associated with deviating from the norm (prescriptive norm), people might go for other options that might also be more optimal.

So, for example, cooperation might be the norm, meaning that getting a payoff from cheating might be less attractive. This will increase the level of cooperation in a society even if the individual preferences might be not to cooperate.

Norms that aim at cooperation seem to be correlated with better payoffs.

Norms might reach new equilibria in f.x. a stag hunt game or a prisoner's dilemma.

Self-serving bias, seems to be hard to overcome which might lead to standstills in bargainings.

• Solution: 
• Change positions, hire external consultants.

This could be the case in a game of conflicting interests where we cannot go any further:

• A game of conflicting interests:

There there are two Nash equilibria in order to reach them, one player has to be convinced to become the "sucker".

If both are equally confident, they will both go for their highest payoff: 4, ending up in [0,0] (BR). But if one player can somehow communicate credibly that they will go for one strategy and "tie themselves to the pole", they might reach an equilibrium.

This requires an external force/decider, a Schelling sort of bargaining power from limiting your own freedom or by perhaps having a good track record and/or more confidence from previous situations, and that can be communicated to the other player.

If one player has some authority over the other we are also able to reach a Nash equilibrium.

A reason why we might find a conflicting interest game like this is that both players might suffer from self-serving bias/overconfidence (Explaining Bargaining Impasse: The Role of Self-Serving Biases, 1997).

Otherwise the outcome here is very uncertain, and we would find a mixed strategy equilibrium.

Credible promises are based on that we can remember past actions of cooperation. That makes us able to reach efficient equilibria in the next stage (illustrated by a help game (stage 1) + conflicting interest game (stage 2)).

Remembering past cooperation can help us reach equilibria in situations where otherwise impossible and/or improbable.

Lesson 1: Credible promises must be consistent with equilibrium play in the future.

• Lesson 2: Fixing Stage 1 required thinking about how, at Stage 2,
• behavior could be conditioned on what happens at Stage 1. The
• agreement concerns what we are going to be remembering in the future.

See that an efficient equilibrium is to cooperate in stage 1 and [B,L] in stage 2. Then the total payoff is 3 for player 1 and 4 for player 2. So if they can trust eachother's promises much can be gained.

• We can also make a point that keeping records, reputation etc. might help the credibility of the promise and thus increase cooperation. This seems to be in line with Basu et al. "Recordkeeping alters economic history
• by promoting reciprocity". Quid pro quo-behaviour.

• To demonstrate the potentially large influence of institutions on human welfare, we begin by spending some time on state–of–the–art macroeconomic
• evidence. We also briefly discuss the microeconomic question of whether
• there are some business practices that are universally better than others. If
• so, which are they?

• Relates to N. Bloom and J. Van Reenen (2010) Why Do Management Practices
• Differ Across Firms and Countries?

and 

R. Levine (2005): Law, Endowments and Property Rights

• The purpose of this section is to set the stage by (i) presenting a simple
• classification of fundamental obstacles to cooperation (cognitive, expectational, and motivational) and (ii) introducing game theory.

• Game theory is an excellent tool for analyzing situations that are well-
• defined and well understood by the participants. It can also be useful for
• analyzing very simple situations that are not well understood by participants.
• It is still an open issue to what extent game theory will also be successful in
• analyzing complex situations that are not (yet) well understood by partici-
• pants.

Self-serving bias in bargaining may lead to inefficient outcomes.

Failing to trust eachother. (Expectational obstacles)

Failing to remember agreements or cooperation in earlier stages (cognitive obstacles).

Who is setting the rules of the game? How does this affect motivations to follow the rules? Elinor Ostrom has some good points here about motivational obstacles to cooperation.

• We pay special attention to games with multiple stages.
• After all, most interesting situations unfold over time! A central objective
• is to pinpoint the difference between credible threats and promises and non-
• credible threats and promises.

A non-credible threat is a term used in game theory and economics to describe a threat in a sequential game that a rational player would actually not carry out, because it would not be in his best interest to do so.

For a simple example, suppose person A walks up, carrying a bomb, to another person B. A tells B he will set off the bomb, killing them both, unless B gives him all his money. If A is rational and non-suicidal he stands nothing to gain from setting off the bomb, so his threat cannot be considered credible. On the other hand, a person in the situation of B might give A his money, fearing that A is not rational, or might even be suicidal.

We can check this through backward induction.

• We analyze the role of morality, in the form of social values and norms. Injunctive social norms describe what people in a society are supposed to do in a situation. Descriptive social norms describe what people in a society typically do in the same situation. We argue that shared values constitute
• a major part of a society’s culture. The other major part is shared understandings. Together, values and understandings shape injunctive norms.
• Internalized injunctive norms can be very powerful, but they are fragile foundations of cooperation. Often, people are only compelled to comply with the
• injunctive norm as long as others do so (see the utility functions). The descriptive norm may therefore
• be not to cooperate, even if the injunctive norm prescribes cooperation.

Abstract

Empirically, disincentive effects (hämmande effekter) on work of generous welfare state arrangements tend to appear with a substantial time lag. One explanation is that norms concerning work and benefit dependency delay such effects. We model altruistic parents' economic incentives for instilling such work norms in their children. Anticipated economic support from parents may reduce work effort, and parental altruism makes threats to withdraw such support noncredible. Instilling norms mitigates this problem. However, generous social insurance arrangements tend to weaken parents' incentives to instill such norms in their children. We find empirical support for this prediction.

Sammanfattning: Barnen lär sig normer av föräldrarna, normen kan vara arbeta hårt eller var bidragsberoende. I en stat där föräldrarna är bidragsberoende kommer de sedan att föra detta vidare till barnen via en "lagged effect". Det leder till att hot om indragna bidrag osv blir mindre credible i takt med att beroendet ökar.

AbstractSome research presumes that, when rational parties bargain, nothing is left on the table, so that social outcomes are efficient and leave countries on the frontiers of their aggregate production functions. A study of differences in per capita incomes across countries shows that this cannot be the case. Countries' endowments of natural and human resources do not explain any significant part of the variation in incomes and the mobility of capital assures that it is impartially available to all countries. National differences in the quality of policies and institutions across countries mainly account for differences in per capita incomes.

• Olson (1965) challenged the presumption that the possibility of benefit for a group would be sufficient to generate collective action to achieve that benefit; instead, he argued:
• Unless the number of individuals is quite small, unless there is coercion or some other
• special device to make individuals act in their common interest, rational self-interested
• individuals will not act to achieve their common or group interests. (Olson 1965, p.2)
• In his analysis, free riding is a problem in all but the smallest groups. In small groups: each member, or at least some one of them, has an incentive to see that the collective good is provided, even if he has to bear the full burden of providing himself. (Olson 1965, p.50)

They argue that the zero-contribution thesis fails to account for factors which encourage collective action and self-regulating capabilities of user groups  (these factors would, I guess, be the Ostrom design principles for long-term collective action). (Runge 1986). Geographers, political scientists and anthropologists argue that many societies have devised, maintained or adapted collective arrangements to manage common pool resources.

• Clearly defined boundaries of the user group and resource
• Congruence between allocation and access rules and local conditions
• Users’ ability to modify the operational rules through collective-choice arrangements (Collective-choice arrangements, that allow most resource appropriators to participate in the decision-making process)
• Monitoring of management system (att det finns en tillit i övervakningen av systemet, kanske där alla deltar som övervakare i enligt ett roterande schema.)
• Graduated sanctions
• Availability of conflict resolution mechanisms (ett effektivt sätt att förändra/döma osv. typ ett rättssystem)
• Management rights of resource users not challenged by external agents (Self-determination of the community recognized by higher-level authorities)
• Nested enterprises (Larger CPRs are organised in the form of multiple layers of nested enterprises, with small local CPRs at the base level.)

• Där V är något positivt värde, C är kostnad associerad med att Hawk-Hawk kommer att innebära ett krig och en kostnad. Vi ser att Dove-Dove och Dove/Hawk el. Hawk/Dove är optimala.

Solution: One tactic in the game is for one party to signal their intentions convincingly before the game begins.

"One real-world example is a protester who handcuffs himself to an object, so that no threat can be made which would compel him to move (since he cannot move)."

So we need to make credible commitments in order to reach either a Dove/Hawk or Dove/Dove solution. This could be the case in nuclear wars etc.

This opens up the possibility to make credible threats or commitments because we then can view ahead and think of possible future outputs.

So in the prisoner's dilemma game, we can see that if we cooperate continously, we can both be better off. The possibility of a repeated game is then solving the problem with untrust/cheating etc.

Cooperate forever until someone cheats, then defect forever. 

If you can communicate this threat in a credible way, then the other person will have strong incentives to cooperate as well, forever.

This is better than unconditional cooperation, because unconditional cooperation does leave incentives for the other player to cheat forever.

The grim trigger strategy is credible if the game is known to be iterated and the likelihood of it suddenly ending is low enough. (If it is a one-off game this strategy naturally has no credibility.)

Om vi har hög sannolikhet att ett spel upprepas kommer vi inte behöva lika hårda straff eftersom det är mer att förlora på att bryta mot en norm/lag/spel.

Om det är väldigt osannolikt att ett spel upprepas måste straffet vara hårdare, eftersom det annars finns mycket höga incitament att avvika från en annars lukrativ överenskommelse.

• The provision of many public and private goods hinges on the actions of the weakest link
• in the chain, that is the lowest contributor (Hirshleifer 1983 and Camerer 2003). The
• evidence suggests that in such cases the likely outcome is coordination failure. Various
• possible solutions have been considered in the literature and our objective in this paper
• was to consider a novel solution, leadership. By leadership we mean that one person, the
• leader, announces their contribution first thereby setting an example for the rest of the
• group. Our main hypothesis was that by choosing a high number the leader could
• improve efficiency.
• As predicted, we find that leadership increases efficiency.