Collusion and Renegotiation in a Principal-Supervisor-Agent- Relationship (REVISION)
0
Citation
0
Reference
20
Related Paper
Abstract:
This paper describes a principal-agent relationship with a supervisor who has information about the agent. The agent and the supervisor have the possibility to collude and misinform the principal. In accordance with the existing literature there exists an optimal contract which excludes collusion in equilibrium. The optimal contract exhibits, however, ex-post inefficient and creates scope for renegotiation. If a renegotiation-stage is incorporated in the game then for some parameter constellations the optimal contract is a contract which necessarily induces collusion. The paper thus shows that the principal's behavior toward ex-post inefficiencies may determine whether collusion occurs in equilibrium.Keywords:
Supervisor
Scope (computer science)
Cite
In this paper we address the question of collusion in mechanisms under asymmetric information by assuming that one of the colluding parties offers a side contract to the other one. We develop a methodology to analyze collusion as an informed principal problem. We show that if collusion occurs after the agents accept or reject the principal’s offer, the dominant-strategy implementation of the optimal contract without collusion is collusion proof. In the second part of the paper, we look at a different timing, assuming that the agents’ decision to accept or reject the principal’s offer is taken after collusion, so the agents share their private information before accepting the principal’s offer. On the other hand, we assume that the collusion offer includes a punishment strategy, to be used whenever the other agent rejects the side contract. We establish the conditions that have to be satisfied for a contract to be collusion proof and we show that the optimal contract without collusion is no longer collusion proof. The optimal collusion proof contract is asymmetric, both in transfers and in quantities.
Private information retrieval
Cite
Citations (7)
We reconsider Tirole's framework of a three-tier principal-agent problem, in which he has shown that an incentive problem is caused by the possibility of monetary side payments between the agent and the middle -level supervisor. We consider the case where monetary transfers are not possible, but a different, implicit kind of collusion arises because the supervisor cares about the utility enjoyed by his subordinate. This approach avoids the crucial problem of how an explicit side contract could be enforced. We show that the difference in the cause of collusion leads to a difference in the way in which collusion is prevented: the agent's incentive scheme is used to give incentives to the supervisor. An interdependence between incentive schemes thus follows from utility interdependence.
Supervisor
Cite
Citations (1)
The standard ex post type of collusion is a supervisor-agent agreement to misrepresent the outcome of supervision. Under ex ante collusion the agent makes a side transfer to the supervisor, who, in return, stops monitoring the agent's productivity. Extending Tirole's [1986] model of hierarchy to include ex ante collusion and supervision costs, we show that the principal can ignore ex ante collusion and the supervisor's incentive constraint if supervision technology is likely to generate information at a low cost. To prevent ex ante collusion the principal increases the difference between the wages paid when the supervisor's report is empty and when it contains productivity evidence.
Cite
Citations (6)
We analyze an adverse selection environment with third party supervision. We assume that the and the can collude while interacting with the principal. As long as the supervisor is symmetrically informed with the agent, the former's existence does not improve the principal's rent extraction. This is due to the efficiency between the supervisor and the agent. However, asymmetric information between these two parties can cause a failure, which undermines the coalitional efficiency. In that case, we show that the principal can increase his payoff, by manipulating the agent's opportunity cost for colluding with the supervisor. Delegating the authority to contract with the agent to the supervisor is not successful in enhancing the principal's payoff, since the principal loses the instrument to manipulate the opportunity cost of collusion under this organizational form. The increase in the principal's rent extraction does not necessarily imply an overall welfare improvement. Social welfare may decline with the introduction of the supervisor.
Supervisor
Adverse selection
Stochastic game
Mechanism Design
Rent Seeking
Cite
Citations (7)
Cite
Citations (4)
The three hierarchies organization model is established by introducing the supervisor, and the potential collusion behavior in the structure is analyzed. Then the theoretical proof about the principal′s hiring the supervisor and concerning the collusion is given by using the cost benefit approach.
Supervisor
Cite
Citations (0)
Supervisor
Budget constraint
Cite
Citations (1)
Supervisor
SIGNAL (programming language)
Cite
Citations (5)
This paper studies a mechanism-design problem involving a principal-supervisor-agent in which collusion between supervisor and agent can only occur after they have decided to participate in the mechanism. We show how collusion can be eliminated at no cost via the use of a mechanism in which the principal endogenously determines the scope of supervision. A simple example of such a mechanism is one in which the agent bypasses the supervisor and directly contracts with the principal in some states of the world. The result that collusion can be eliminated at no cost in this environment highlights the important assumptions required for collusion to be a salient issue in the existing literature. The result is robust to alternative information structures, collusive behaviours and specification of agent's types. Applications include work contracts with different degrees of supervision, self-reporting of crimes, tax amnesties, immigration amnesties and mechanisms based on recommendation letters.
Cite
Citations (6)
We use experiments to analyze what type of communication is most effective in achieving cooperation in a simple collusion game. Consistent with the theories of collusion and cheap talk, an initial burst of collusion rapidly collapses if subjects can only use a limited message space that does not allow communication of contingent strategies. When unlimited pre-game communication is allowed, a similar initial decline in collusion reverses over time. Content analysis is used to identify multiple channels by which communication improves collusion in this setting. Explicit threats to punish cheating are the most important factor to successfully establish collusion, consistent with the theory. Surprisingly, collusion is even more likely when we allow for renegotiation, contrary to standard theories of renegotiation. With renegotiation cheaters are often admonished in strong terms. Allowing renegotiation therefore appears to increase collusion by allowing for an inexpensive and highly effective form of punishment.
Scope (computer science)
Cite
Citations (9)