Optimal mechanism design and money burning

Abstract

Mechanism design is now a standard tool in computer science for aligning the incentives of self-interested agents with the objectives of a system designer. There is, however, a fundamental disconnect between the traditional application domains of mechanism design (such as auctions) and those arising in computer science (such as networks): while monetary transfers (i.e., payments) are essential for most of the known positive results in mechanism design, they are undesirable or even technologically infeasible in many computer systems. Classical impossibility results imply that the reach of mechanisms without transfers is severely limited.Computer systems typically do have the ability to reduce service quality-routing systems can drop or delay traffic, scheduling protocols can delay the release of jobs, and computational payment schemes can require computational payments from users (e.g., in spam-fighting systems). Service degradation is tantamount to requiring that users burn money, and such "payments" can be used to influence the preferences of the agents at a cost of degrading the social surplus.We develop a framework for the design and analysis of money-burning mechanisms to maximize the residual surplusthe total value of the chosen outcome minus the payments required. Our primary contributions are the following.

What Money Can't Buy: Allocations with Priority Lists, Lotteries and Queues

“…non monotone hazard rates). Hartline and Roughgarden (2008) contains a result which was obtained independently and is very similar to my Proposition 1. In contrast to my analysis, the authors discuss applications to computer science and do not deal with the practical implementation of the optimal mechanism.…”

“…In contrast to the papers above, I provide, within the same environment, a closed form solution to the problem of designing an optimal mechanism. Closely related environments, where agent compete for a set of homogenous goods by engaging in costly effort or money burning, are studied in a number of papers in the context of different applications: McAfee and McMillan (1992), Chakravarty and Kaplan (2006), Yoon (2009), and Hartline and Roughgarden (2008). My results are significantly more gen-3 If an institution has a direct interest in screening individuals on the basis of specific characteristics a distribution based on the willingness to pay may not be optimal (e.g.…”

Corruption, Intimidation, and Whistle-blowing: a Theory of Inference from Unverifiable Reports

“…12 Our perspective here echoes that developed in work by Hurwicz and Shapiro (1978), Segal (2003), Hartline and Roughgarden (2008), Madarász and Prat (2010), Chassang (2013), Frankel (2014), Carroll (2013.…”

Making Corruption Harder: Asymmetric Information, Collusion, and Crime

“…This includes maxmin optimal design (Hurwicz and Shapiro, 1978, Hartline and Roughgarden, 2008, Chassang, 2013, Frankel, 2014, Madarász and Prat, 2014, Prat, 2014, Carroll, 2013, as well as data-driven design (Segal, 2003, Chassang and Padró i Miquel, 2016, Brooks, 2014). …”