Optimal Delegation in a Multidimensional World
Andreas Kleiner ()
CRC TR 224 Discussion Paper Series from University of Bonn and University of Mannheim, Germany
Abstract:
We study a model of delegation in which a principal takes a multidimensional action and an agent has private information about a multidimensional state. We provide necessary and sufficient conditions for an arbitrary mechanism to maximize the principal’s expected payoff. A key step shows that a function is the indirect utility of an incentive-compatible mechanism if and only if it is convex and lies below the agent’s full-discretion payoff. We apply our optimality conditions to convex delegation sets and characterize when a single budget optimally links decisions. Linking can preserve discretion even when separate dele gation would eliminate it. Optimal linking can also require a nonconvex delegation set, even with uniformly distributed types. If the principal cares little about aligned decisions, every optimal mechanism randomizes: risk in an aligned decision provides incentives in the biased ones at little cost to her.
Keywords: delegation; mechanism design without transfers (search for similar items in EconPapers)
JEL-codes: C61 D82 D86 (search for similar items in EconPapers)
Pages: 53
Date: 2026-09
New Economics Papers: this item is included in nep-des, nep-mic and nep-upt
References: View references in EconPapers View complete reference list from CitEc
Citations:
Downloads: (external link)
https://www.crctr224.de/research/discussion-papers/archive/dp784 (application/pdf)
Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.
Export reference: BibTeX
RIS (EndNote, ProCite, RefMan)
HTML/Text
Persistent link: https://EconPapers.repec.org/RePEc:bon:boncrc:crctr224_2025_784
Access Statistics for this paper
More papers in CRC TR 224 Discussion Paper Series from University of Bonn and University of Mannheim, Germany Kaiserstr. 1, 53113 Bonn , Germany.
Bibliographic data for series maintained by CRC Office ().