Book contents
- Frontmatter
- Contents
- Preface
- Notation
- 1 Introduction
- 2 The psychological predictability problem
- 3 Rational choice responses
- 4 Behaviourally informed responses
- 5 Behaviourally determined responders
- 6 Outlook: implications for interaction with higher complexity
- 7 Predictability at the crossroads of competing institutionalisms
- Equations
- References
- Index
4 - Behaviourally informed responses
Published online by Cambridge University Press: 22 September 2009
- Frontmatter
- Contents
- Preface
- Notation
- 1 Introduction
- 2 The psychological predictability problem
- 3 Rational choice responses
- 4 Behaviourally informed responses
- 5 Behaviourally determined responders
- 6 Outlook: implications for interaction with higher complexity
- 7 Predictability at the crossroads of competing institutionalisms
- Equations
- References
- Index
Summary
Introduction
There is no way to escape the trade-off between parsimony and fit. Parsimonious models are sharp lenses. They make it possible to see details that would remain concealed to the unaided eye. But their sharpness comes at a cost – namely a narrow field of vision. Institutional design is about taking action. It profits from the details uncovered by rational choice modelling. But before intervening, the institutional designer must have a second look at his object from a greater distance. This is the external limitation of (rational choice) models, resulting from their limited fit to reality.
Internal limitations are added to this. Some of them are inherent in all rational choice models. The most important is the decisive role of a uniform normative currency, like price. The rational choice model is made for marginal analysis. It is most powerful when describing how the alteration of one parameter plays itself out in other parameters. This capacity of the model was exploited in chapter 3 when the expected utilities of the several options available to the parties were compared. Such calculations become difficult, if not impossible, if there is not one normative currency with which to compare the several parameters. A case in point is the following: the model is hard to apply if Ego cares for Alter's co-operation for reasons other than the gains from trade. He might, for instance, have a strong dislike for being the sucker, since he sees that as a violation of his self-esteem.
- Type
- Chapter
- Information
- Generating PredictabilityInstitutional Analysis and Design, pp. 192 - 242Publisher: Cambridge University PressPrint publication year: 2005