Nicholas Dopuch, Ronald R. King, The Impact of MAS on Auditors' Independence: An Experimental Markets Study, Journal of Accounting Research, Vol. 29, Studies on Accounting Institutions in Markets and Organizations (1991), pp. 60-98

In this paper we report the results of experimental markets designed to test a disclosure model based on the models of Dye [1985] and Jung and Kwon [1988]. These authors show that when receivers of information do not know whether senders possess private information, the senders will not fully disclose their private information. Their models were motivated by the discrepancy between the theoretical predictions for voluntary disclosures and empirical results. In the theoretical area, Grossman [1981] and Milgrom [1981] predicted that private information will be fully disclosed when disclosures are credible and receivers know the holder has private information; agents disclose to identify themselves as not possessing the worst possible information. Watts [1977] argued that voluntary disclosure will occur because of market pressures and the threat of regulatory intervention. Beaver [1977; 1988] described how auditing and legal liability may induce full disclosure. Nevertheless, empirical research indicates full disclosure is not always observed (e.g., Penman [1980] and Seligman [1986]).