[This paper examines the voluntary disclosure of nonproprietary information using the model of uncertain information endowment developed by Dye (1985) and Farrell (1986), and extended by Jung and Kwon (1988). The paper focuses on a broad family of functions relating the probability of information acquisition to ex post information quality. The paper shows that for each function there is some region that displays a negative relation between ex ante information quality and the frequency of disclosure. In addition, a sub-family of functions is identified for which ex ante information quality and the frequency of disclosure are negatively related everywhere. These results indicate that the economic intuition that higher informational asymmetry is accompanied by more voluntary disclosure is not generally true.]

Abstract This paper examines the voluntary disclosure of nonproprietary information using the model of uncertain information endowment developed by Dye (1985) and Farrell (1986), and extended by Jung and Kwon (1988). The paper focuses on a broad family of functions relating the probability of information acquisition to ex post information quality. The paper shows that for each function there is some region that displays a negative relation between ex ante information quality and the frequency of disclosure. In addition, a sub-family of functions is identified for which ex ante information quality and the frequency of disclosure are negatively related everywhere. These results indicate that the economic intuition that higher informational asymmetry is accompanied by more voluntary disclosure is not generally true.

Abstract We use a linear contracting framework to study how the relation between performance measures used in an agent's incentive contract and the agent's private predecision information affects the value of delegating decision rights to the agent. The analysis relies on the idea that available performance measures are often imperfect representations of the economic consequences of managerial actions and decisions, and this, along with gaming possibilities provided to the agent by access to private predecision information, may overwhelm any benefits associated with delegation. Our analytical framework allows us to derive intuitive conditions under which delegation does and does not have value, and to provide new insights into the linkage between imperfections in performance measurement and agency costs.