Abstract
Strategic behavior between an auditor and a client can significantly increase both audit cost and achieved audit risk. This paper tests the widely cited Fellingham and Newman (1985) analytic model of strategic auditor-client interaction in a series of controlled experiments. The Fellingham and Newman model is tested in both single-period and multiperiod settings, with parameter values chosen to yield either a pure or a mixed Nash equilibrium strategy that is Pareto dominated by a strategy of mutual auditor-client cooperation. The Fellingham and Newman model is best supported by auditor behavior in both the single- and mutiperiod experiments of the mixed strategy Nash equilibrium setting. Generally, the Fellingham and Newman model fails to predict a significant proportion of observed auditor and client behavior. Much of the observed non-Nash behavior involves the socially beneficial strategy of mutual cooperation. Client and auditor behavior is also found to be sensitive to both the time horizon and the asymmetric nature of the auditor and client payoffs. Limitations of the Fellingham and Newman model and implications for future modeling of strategic auditor-client behavior in multiperiod settings are discussed.