:::

Defaults and Benchmarks in Mechanism Design


  • 研討會日期 : 2018-12-21
  • 時間 : 10:30
  • 主講人 : Professor Olivier Bochet
  • 主持人 : Professor Ritesh Jain
  • 地點 : Conference Room B110
  • 演講者簡介 : Professor Bochet received his Ph.D. in Economics from Brown University in 2004. He is currently an Associate Professor of Economics at New York University Abu Dhabi. His research interests are Mechanism Design, Market Design, and Experimental Economics.
  • 演講摘要 : We consider the robust implementation problem with behavioral agents who compare their envisioned choices to some default or benchmark strategies. We derive new interim solution concepts whose relative strengths are determined by the collection of possible type spaces in consideration. We call these interim default reversion proof (IDRP) equilibria. We prove a strong version of the revelation principle: any social choice function (SCF) that is implementable in IDRP equilibrium is also fully implementable by a direct revelation mechanism in which agents use truthtelling as a default. Specializing to direct mechanisms, we define robust resilient implementation and show that it is equivalent to implementation in a new notion of interim rationalizability that iteratively eliminates strategies that are dominated by truthtelling. This echoes an important equivalence result in Bergemann and Morris (2009, 2011). Under strict ex-post incentive compatibility, an SCF is robust resilient implementable if and only if it satisfies a new condition, the strong contraction property. In a leading example of Bergemann and Morris (2009), we show that the degree of interdependence in preferences necessary for robust resilient implementation is half the upper bound they identify.