Frank Gigler

Self-Enforcing Voluntary Disclosures

This paper extends previous accounting research which asks why firms do not voluntarily disclose all value-relevant information, as implied by, for example, Grossman [1981] and Milgrom [1981]. While these two models assume disclosures are costless and verifiable, subsequent papers by Verrecchia [1983], Dye [1985 b], Darrough and Stoughton [1990], Wagenhofer [1990], Feltham and Xie [1992], and Newman and Sansing [1993] suggest firms withhold private information to avoid proprietary costs.1 In contrast to these models, I show that proprietary costs can actually increase voluntary disclosures by generating credibility for such disclosures. I model the decision to disclose private information when independent verification is impossible, too costly, or otherwise unavailable. The results rationalize unverified voluntary disclosures, and market reactions to them, by showing that the costs of disclosing proprietary information can make such disclosures credible, even without verification. In the model, a firm with private information about the demand for its product makes a direct public disclosure to both a competitor and the capital market. The firm would like to convince the capital market that the demand for its product is high, thereby increasing the value of the

The Effect of Earnings Forecasts on Earnings Management

We develop a theory of the association between earnings management and voluntary management forecasts in an agency setting. Earnings management is modeled as a “window dressing” action that can increase the firm’s reported accounting earnings but has no impact on the firm’s real cash flows. Earnings forecasts are modeled as the manager’s communication of the firm’s future cash flows. We show that it is easier to prevent the manager from managing earnings if he is asked to forecast earnings. We also show that earnings management is more likely to follow high earnings forecasts than low earnings forecasts. Finally, our analysis shows that shareholders may not find it optimal to prohibit earnings management. Earlier results rationalize earnings management by violating some assumption underlying the Revelation Principle. By contrast, in our model the principal can make full commitments and communication is unrestricted. Nonetheless, earnings management can be beneficial as it reduces the cost of eliciting truthful forecasts.

Assessing the Information Content of Mark-to-Market Accounting with Mixed Attributes: The Case of Cash Flow Hedges

We examine how outsiders rationally interpret a reported loss on derivatives when the application of mark-to-market accounting to cash flow hedges creates a mixed attribute problem. We find that because of the mixed attribute problem, the information content of mark-to-market accounting is related to the information content of historical cost accounting in a very specific way. This relationship allows us to identify the circumstances under which mark-to-market accounting facilitates and when it detracts from the objective of providing an early warning of potential financial distress. We show that the reporting of an impending derivative loss by a distressed firm can actually lead outsiders to infer that the firm is in a better financial position than what they would have inferred under the silence associated with historical cost accounting. Without the mixed attribute problem, mark-to-market accounting would always yield more accurate assessments of the firms financial position.

Informational costs and benefits of creating separately identifiable operating segments

We provide an informational theory for how the ownership claims to a firm might be structured. When the market price of equity provides valuable contracting information there is a benefit to creating separate ownership claims to each of a firms divisions. However, creating this information also generally has adverse incentive effects because it enriches the agents strategy space. We show in a complete contracting setting that under a large class of agencies the firm is strictly better off bundling the ownership claims to divisions that are sufficiently similar and creating separate ownership claims only to divisions that are sufficiently different.

Discussion of Centralization, Delegation, and Shared Responsibility in the Assignment of Capital Investment Decision Rights

Baiman and Rajan use the joint forces of relationship-specific investment and incomplete contracting to model a principals choice between a centralized organization, where he decides the level of investment in a capital asset, and a decentralized organization, where those decision rights are contractually assigned to the agent. When contract incompleteness prevents the principal from committing to a level of capital investment, his ex post incentive to underinvest in the capital asset causes an inefficient mix of capital and labor. The principal can eliminate the inefficiency resulting from his own opportunistic behavior by assigning the capital investment decision to the agent, but the agent will overinvest in the capital asset in order to reduce his personal cost of effort when he is given the decision rights. As a result, the principals centralizationdelegation choice is essentially a choice between the inefficiency resulting from the principals opportunistic underinvestment of capital and the inefficiency resulting from the agents incentive to minimize his effort by overinvesting in capital. Some type of contract incompleteness is necessary for the organizational form to matter. Baiman and Rajan assume that the agents compensation cannot depend on the realization of a state variable, 0, which determines the efficient mix of capital and labor. However, the agent