Information Disclosure and Opinion Divergence

Abstract

Does communicating more information always lead to greater agreement, in the sense that agents are more likely to agree on a given course of action, conditional on their information sets? According to information economics, the answer is yes. Moreover, the famous Representation Theorem asserts that in equilibrium agents will voluntarily disclose all they know, and this theorem has profoundly influenced the literature on information disclosure. In this paper, I show that both results critically depend on the assumption that all information is objective. I develop a model in which managers of firms are voluntarily communicating objective and subjective information, and prior beliefs about the strategy to maximize project value are rational but heterogeneous, to establish that more information does not necessarily lead to more agreement. Three main sets of results are derived. First, all firms voluntarily signal (objective) private information they have about their “type”, but not all firms disclose additional (subjective) information about strategy that is prone to multiple interpretations and hence potential disagreement. This lack of disclosure occurs even though there are no exogenous communication costs and “proprietary signaling costs” are precluded because information is being communicated only to the financial market and not to the firm’s product‐market competitors. Second, more valuable firms voluntarily disclose less information in equilibrium. Third, information disclosure has real consequences because it interacts with capital budgeting, investor relations and corporate governance. Firms that disclose more have a stronger preference for faster‐payback projects. While all firms that disclose additional strategy information also invest in investor relations, only a subset of firms that do not disclose this subjective information choose to invest in investor relations. Finally, corporate governance interacts with information disclosure – as the quality of corporate governance improves, voluntary information disclosure declines. Thus, this paper provides an economic rationale for higher‐quality firms to withhold more strategic information, and this has pervasive consequences.