Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/189148 
Authors: 
Year of Publication: 
1991
Series/Report no.: 
Queen's Economics Department Working Paper No. 824
Publisher: 
Queen's University, Department of Economics, Kingston (Ontario)
Abstract: 
Truthful revelation mechanisms with auditing have the undesirable property that audits are not actually performed in equilibrium because all inference problems have been solved. A model is proposed in which the inference problem is preserved by separating the regulatory and auditing functions and transfers are costly. The auditor designs a Bayesian audit procedure and the regulator credibly commits to using this procedure in the regulatory mechanism. The auditor is conservative, that is, he does not like to make mistakes for his client. Second-best allocation is achieved over a well-defined auditing region of the regulator's prior beliefs about firm type. The auditing region is increasing in the precision of the auditing technology and is decreasing in the strength of the regulator's prior beliefs.
Document Type: 
Working Paper

Files in This Item:
File
Size





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.