The purpose of this paper is two-fold. First, I attempt a taxonomy of the extant accounting literature on disclosure: that is, a categorization of the various models of disclosure in the literature into well-integrated topics. With regard to the taxonomy, I suggest three broad categories of disclosure research in accounting. The first category, which I dub “association-based disclosure”, is work that studies the effect of exogenous disclosure on the cumulative change or disruption in investors’ individual actions, primarily through the behavior of asset equilibrium prices and trading volume. The second category, which I dub “discretionary-based disclosure”, is work that examines how managers and/or firms exercise discretion with regard to the disclosure of information about which they may have knowledge. The third category, which I dub “efficiency-based disclosure”, is work that discusses which disclosure arrangements are preferred in the absence of prior knowledge of the information, that is, preferred unconditionally. Then, in the final section of the paper, I recommend information asymmetry reduction as one potential starting point for a comprehensive theory of disclosure. That is, I recommend information asymmetry reduction as a vehicle to integrate the efficiency of disclosure choice, the incentives to disclose, and the endogeneity of the capital market process as it involves the interactions among individual and diverse investors.

In this discussion I comment on the contribution of two papers toward our understanding of how disclosure affects the cost of capital. Specifically, in the context of these papers, I comment on whether disclosure ameliorates or exacerbates the cost of capital that arises from the existence of information asymmetries in capital markets. This is a notion that should be of fundamental interest in that it provides an economic basis for evaluating the costs and benefits of accounting information.

This study examines a commercial bank manager's incentives to reduce regulatory costs imposed when the bank's capital adequacy ratio falls below its regulatory minimum. It also tests the general political sensitivity hypothesis that a manager seeks to reduce political costs incurred when revenue is unusually large. Tests of adjustments to the loan loss provision, loan charge-offs, and securities gains and losses attempt to control for exogenous economic conditions and previous investing decisions. Results are consistent with hypotheses associating accounting adjustments with capital adequacy ratio guidelines, but fail to support the political sensitivity hypothesis.

Using the model of discretionary disclosure suggested in Verrecchia (1983), I show that an increase in the quality of private information received by a manager results in more disclosure on his part. I also discuss how this result influences a manager's choice over levels of quality.

In the preceding paper Darrough and Stoughton suggest that firm interdependence, modelled as an entry game among firms in a product market, can yield endogenous proprietary costs. This allows the costs associated with the dissemination of information about the firm to depend upon the information's content in some direct way. My comments focus on: first, the structure of the game, which emphasizes the potential for full disclosure; second, the extent to which the entry game exaggerates the usefulness of ‘bad news’; and third, their suggestion that more competition among firms implies more, and not less, disclosure, an observation seemingly contrary to Verrecchia (1983).

This paper considers the role of an agent choosing among reporting alternatives when that choice is unobserved by the principal, and the agent's compensation contract is optimal. The agent is allowed to take post-outcome costly actions which lead to more precise reports of actual profits than would be yielded by less precise, but costless conventional translations of outcomes (e.g., GAAP). The extent to which the principal allows the agent this discretion depends upon the improvement in profits as an indicator of the agent's pre-outcome effort when post-outcome actions take place versus the attendant cost of these actions.

Daily returns are used to investigate the effect of preferred stock rating change announcements on preferred and common stock prices. Announcements that are free of confounding events, ‘clean’ announcements, significantly affect preferred stock prices. However, the effect occurs after the day of announcement, mostly on event day +1. Conversely, there is no evidence ‘clean’ announcements affect common stock prices. Larger preferred stock abnormal returns are associated with announcements that are contaminated by confounding events, but the abnormal returns appear to be the result of the confounding events more than the rating change.

This paper shows how the existence of disclosure-related costs offers an explanation for why a manager exercise discretion in disclsing information even though traders have rational expectations about his motivation to withhold unfavorable reports. In effect, disclosure-related costs introduce noise by extending the range of possible interpretations of withheld information to include news which is actually favorable. Therefore, traders are unable to interpret withheld information as unambiguously ‘bad new’ and thereby discount the value of the firm to the point that the manager is better served to disclose what he knows.