Computer Science > Computer Science and Game Theory
[Submitted on 15 Feb 2023 (this version), latest version 16 Sep 2024 (v5)]
Title:Ambiguous Contracts
View PDFAbstract:In this paper we initiate the study of ambiguous contracts, capturing many real-life scenarios where agents engage in contractual relations that leave some degree of uncertainty. Our starting point is the celebrated hidden-action model and the classic notion of a contract, where the principal commits to an outcome-contingent payment scheme for incentivizing an agent to take a costly action. An ambiguous contract generalizes this notion by allowing the principal to commit to a set of two or more contracts, without specifying which of these will be employed. A natural behavioral assumption in such cases is that the agent engages in a max-min strategy, maximizing her expected utility in the worst case over the set of possible contracts. We show that the principal can in general gain utility by employing an ambiguous contract, at the expense of the agent's utility. We provide structural properties of the optimal ambiguous contract, showing that an optimal ambiguous contract is composed of simple contracts. We then use these properties to devise poly-time algorithms for computing the optimal ambiguous contract. We also provide a characterization of non-manipulable classes of contracts - those where a principal cannot gain by employing an ambiguous contract. We show that linear contracts - unlike other common contracts - are non-manipulable, which might help explain their popularity. Finally, we provide bounds on the ambiguity gap - the gap between the utility the principal can achieve by employing ambiguous contracts and the utility the principal can achieve with a single contract
Submission history
From: Daniel Peretz [view email][v1] Wed, 15 Feb 2023 12:23:54 UTC (772 KB)
[v2] Wed, 31 May 2023 07:34:04 UTC (781 KB)
[v3] Tue, 30 Jan 2024 11:58:31 UTC (71 KB)
[v4] Wed, 7 Aug 2024 11:00:54 UTC (42 KB)
[v5] Mon, 16 Sep 2024 17:48:08 UTC (42 KB)
References & Citations
export BibTeX citation
Loading...
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Connected Papers (What is Connected Papers?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
alphaXiv (What is alphaXiv?)
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Hugging Face (What is Huggingface?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)
Demos
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
CORE Recommender (What is CORE?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.