SPS, Part IIa, Pol 3
Pieter van Houten
§ Introduce the main characteristics of coordination problems, and its implications for the analysis of group action.
§ Illustrate applications in political science of one type of coordination problem (‘tipping phenomena’).
§ Discuss limitations of rational choice theories, which result from the presence of multiple equilibria and are exemplified by the analysis of coordination problems.
This lecture focuses on aspects and potential problems associated with group action even if there are no incentives to ‘free ride’ on the activities of others.
The main insight derived from ‘coordination problems’ is that in the absence of incentives to free ride, it is still not certain that an outcome preferred by a group will occur. Further requirements need to be met.
Simple example used in this lecture: leaving a college dinner. Most people may want to leave, but no one wants to be the first. Some coordination is required to obtain the desired outcome.
Political example: coordination of groups in society may be required to maintain the ‘rule of law’ in a country or overthrow an existing regime (Weingast 1997).
‘Classical’ example: side of the road to drive on. It does not matter much which side is selected, as long as all people select the same side. In the case of two people, this situation can be represented as is done in the table below [see next page].
There are two equilibria in this case. Note that it is not guaranteed that coordination will be successful. Some social organisation or political action may be required. Some laws and regulations can be seen to serve as such ‘coordinating devices.’ The situation is somewhat (but not fundamentally) different if one of the two equilibria is preferred by both actors.
|
|
PERSON 2 |
|
|
|
L |
R |
|
PERSON |
1, 1 |
0, 0 |
|
1 |
0, 0 |
1, 1 |
Another interesting case occurs when actors disagree about which equilibrium is most preferred:
|
|
PERSON 2 |
|
|
|
A |
B |
|
PERSON |
2, 1 |
0, 0 |
|
1 |
0, 0 |
1, 2 |
Are still two equilibria, but it may be harder to achieve coordination. Situation is known as ‘Battle of the Sexes.’ Examples: choice of an official language, reaching an international agreement.
These illustrations suggest a potential weakness of rational choice theories, namely, the presence of multiple equilibria. On the one hand, this means that rational choice arguments are compatible with various outcomes. On the other hand, it makes such theories potentially inconclusive, and thus less useful for predicting outcomes.
Rational choice theory has no solutions to this problem. The best suggestion is the notion of a focal point (Schelling 1960), which usually refers to characteristics of the situation not captured by the model itself. Example: meeting someone without communication. As an interesting aside, the idea of focal points provides one possible explanation for path dependence.
Dinner example again: most likely, a cascade of leaving people will occur. Such a dynamic is quite common (Schelling 1978).
General feature: individual decisions depend on the (expected) number of other people doing certain things. Number of people required to change one’s behavior: threshold or tipping point. These can differ for different actors.
These situations have a few characteristics of a ‘collective action problem’ (lecture 4), but are mostly a coordination problem. One of the key issues: How to get enough people to initiate a cascade?
Applications:
§ Mass protest and rebellion (Kuran 1991; Lohmann 1994).
Kuran focuses on the East European revolution in 1989, and tries to account for the fact that this took most analysts by surprise. He argues that there was widespread dissatisfaction, but that it was short of triggering a cascade of protests and was not publicly expressed. Certain external developments triggered a cascade in the late-1980s, overthrowing the existing regimes. With hindsight, this seemed inevitable.
Implication: it is hard to infer individual preferences from collective outcomes. See also Schelling (1978) on segregation of neighborhoods. Detailed research is necessary to be able to infer individual preferences.
Limitation of tipping models: can explain dynamics of group action, but not what triggers it.
§ Language choice, identity formation, ethnic violence (Laitin 1994, 1995, 1998; Hardin 1995).
Laitin looks at possible ways to induce a switch from English to a local language in Ghana, ethnic violence against members of one’s own minority group to raise the costs of not changing one’s behavior (learn a language, support independence, etc.), and identity formation among Russian minorities in former Soviet republics. In all these cases, he identifies tipping phenomena, and uses this to explain particular outcomes.
Coordination problems and tipping phenomena are rational choice ‘tools’ with various applications in the study of politics. Importantly, they also give us some sense of the limitations of such analyses.
Schelling (1978), chapters 3 (esp. 83-110) and 4, gives several examples of coordination and tipping phenomena, and provides a good discussion of the use of models in the social sciences. Chapter 7 provides a more formal (and more confusing) analysis of these phenomena. Schelling (1960) [see reading list for lecture 7], chapter 3, introduces the notion of a ‘focal point’ to explain particular outcomes in situations with multiple equilibria. For the argument that the maintenance of the ‘rule of law’ requires coordination among groups in society, see Weingast (1997). Kuran (1991) uses a tipping model to explain the East European revolution of 1989, while Lohmann (1994) applies a more complicated model to the protests in East Germany, which were part of this revolution. Laitin (1994) presents the choice of an official language as a coordination or tipping game, while the application to violence by ethnic minorities can be found in Laitin (1995). Laitin (1998) uses a tipping game framework to discuss identify formation among Russian minorities in several former Soviet republics. Hardin (1995), chapter 6, applies the logic of tipping phenomena to various instances of violent ethnic conflict.