Description
Works backwards from a desired outcome to design the rules, information and incentives of a system so that self-interested users produce that outcome.
Mechanism design inverts game theory: instead of predicting behaviour from fixed rules, it asks which rules will make strategic participants produce a desired outcome. Key properties are incentive compatibility (truthful behaviour is each participant's best strategy), individual rationality (participation is worthwhile) and efficiency or fairness of the outcome. In product terms, the 'mechanism' includes matching rules, pricing, reputation systems, what information is revealed when, and how rewards are allocated. A rigorous application specifies the objective formally, enumerates the information each party holds privately, proposes rules, proves or simulates their incentive properties, and then tests them with real users, since bounded rationality means people do not always follow the theoretically optimal strategy. Market designers such as Roth applied it successfully to kidney exchange and school choice.
Attributes
- Discipline
- Game Theory & Decision Science
- Loss type
- Unmigrated
- Data
- Analytical / expert
- Effort
- Months
Steps
- State the platform's objective formally (e.g. maximise successful matches subject to fairness constraints).
- Identify participants, their private information and their likely strategic incentives.
- Propose candidate rules: matching algorithm, pricing, information disclosure and reputation rules.
- Analyse or simulate incentive compatibility and individual rationality under each candidate.
- Pilot the preferred mechanism with real users and check for deviations from predicted behaviour.
- Refine the interface so the incentive-compatible strategy is also easy to understand and act on.
Sources
- Preference Belief and Similarity
- Who Gets What and Why (Alvin Roth)
Also known as
- Reverse game theory
- Incentive design
- Market design
Categories
- Strategy & Collaboration
- Game Theory & Decision Science