[Firm-specific information can affect expected returns if it affects investor uncertainty about risk-factor loadings. We show that a stock's expected return is decreasing in factor-loading uncertainty, controlling for the average level of its factor loading. When loadings are persistent, learning by investors can induce time-series variation in price-dividend ratios, expected returns, and idiosyncratic volatility, even when the aggregate risk-premium is constant and fundamental shocks are homoscedastic. Consistent with our predictions, we estimate that average annual returns of a firm with the median level of factor-loading uncertainty are 400 to 525 basis points lower than a comparable firm without factor-loading uncertainty.]

We examine a duopolistic setting in which firms can preannounce their future competitive decisions before they actually implement them. We show that there is a unique equilibrium in which both firms preannounce and overstate their future actions when uncertainty of demand is low. We find that firms choose higher real actions than the ones they would choose in the absence of preannouncements. Moreover, the real actions of both firms are single-peaked functions of their credibility. In a Cournot market, firms face a prisoner's dilemma in which preannouncing lowers firms' profits. If firms could commit not to preannounce, they would remain silent. In a Bertrand market, firms overstate their future actions to foster collusion.

We examine the effects of mandating compensation disclosure on executive incentive contracts, earnings management, firm value, and social welfare. We develop a moral hazard model with multiple principal-agent pairs facing an external monitor who allocates resources across firms to verify earnings management. With such scrutiny allocation, contracts exhibit externalities that create a coordination problem among principals. Contract disclosure enables principals to design the contract anticipating the monitor’s reaction. However, it may also exacerbate the coordination problem among principals because they do not consider externalities on other principals caused by the effects of their contract choices on the monitor’s scrutiny allocation. If internal controls are relatively weak, contract disclosure may make contracts more strongly contingent on reported earnings, increase earnings manipulation, and nevertheless increase social welfare. Contract disclosure improves firm value only if the scrutiny resources available to the monitor are not strongly constrained.

Often, firms reveal oversights and bad decisions publicly through their financial reporting (for instance, restating earnings, impairing goodwill, etc.). These “confessions,” which usually lead to immediate reputation losses, may be attributed to attempts to be perceived as transparent or to attempts to avoid likely litigation costs. In this paper, however, we argue that reputational concerns about perceived ability alone can provide firms with strong enough incentives to confess their mistakes, even in the absence of other non-reputational disciplinary mechanisms. Analyzing the repeated interaction between a firm and an external evaluator who may detect the firm's mistakes, we show that, in equilibrium, a confession places the firm under higher future scrutiny, which is more costly for lower-quality firms. Consequently, in equilibrium, higher-quality firms confess mistakes more often.

We examine whether stress tests distort banks' risk‐taking decisions. We study a model in which a regulator may choose to rescue banks in the event of concurrent bank failures. Our analysis reveals a novel coordination role of stress tests. Disclosure of stress‐test results informs banks of the failure likelihood of other banks, which can reduce welfare by facilitating banks' coordination in risk‐taking. However, conducting stress tests also enables the regulator to more effectively intervene banks, coordinating them preemptively into taking lower risks. We find that, if the regulator has a strong incentive to bail out, stress tests improve welfare, whereas if the regulator's incentive to bail out is weak, stress tests impair welfare.

Firm-specific information can affect expected returns if it affects investor uncertainty about risk-factor loadings. We show that a stock's expected return is decreasing in factor-loading uncertainty, controlling for the average level of its factor loading. When loadings are persistent, learning by investors can induce time-series variation in price-dividend ratios, expected returns, and idiosyncratic volatility, even when the aggregate risk-premium is constant and fundamental shocks are homoscedastic. Consistent with our predictions, we estimate that average annual returns of a firm with the median level of factor-loading uncertainty are 400 to 525 basis points lower than a comparable firm without factor-loading uncertainty.

Uniformity is an essential feature of financial reporting, yet its desirability has long been debated. We study a model in which firms decide whether to adopt either their locally preferred accounting methods or a common method, followed by an investor allocating capital across firms. Firms’ choices of a common method are strategic complements in attaining more comparable reports. As a result, multiple equilibria may exist. Specifically, an equilibrium in which firms use their local methods always exists. However, an equilibrium in which firms adopt a common method exists if uniformity improves comparability significantly and firm-specific productivity shocks are large relative to the common productivity shock. Firms may fail to coordinate on adopting the Pareto-dominant accounting method, which may not even emerge as an equilibrium if investments exhibit substitutability. These coordination problems provide accounting regulation an opportunity to facilitate efficient capital allocation, thus providing a microfoundation for accounting measurement regulation.

This paper examines banks' choice between fair-value and historical-cost accounting when reported accounting information is used in capital requirement regulation. We center our analysis on a key difference between fair-value and historical-cost accounting: the frequency with which asset value changes are reported. We show that the elasticity of banks' loan returns to aggregate lending is a critical determinant of the interaction between capital adequacy requirements and accounting choices. If lending returns are inelastic, then higher capital requirements reduce fair-value usage. By contrast, higher capital requirements encourage fair value if capital requirements are low and lending returns are sufficiently elastic. In equilibrium, banks may elect different accounting choices, and we find that mandating uniform adoption of historical cost (fair value) is desirable when capital requirements are loose (tight). Our study offers many other implications about fundamental links between accounting and prudential choices.

We study the interaction between interbank competition and accounting information quality and their effects on banks' risk-taking behavior. We identify an endogenous false-alarm cost that banks incur when forced to sell assets to meet capital requirements. We find that when the interbank competition is less intense, an improvement in the quality of accounting information encourages banks to take more risk. Keeping the banks' investments in loans constant, the provision of high-quality accounting information reduces the false-alarm cost of assets sales and improves the discriminating efficiency of the capital requirement policy. When considering the banks' endogenous investment decisions, however, this improvement in discriminating efficiency causes excessive risk-taking, because banks respond by competing more aggressively in the deposit market, and the increase in deposit costs motivates banks to take more risk. Our paper shows that improving information quality increases risk-taking with mild competition, but has no effect under fierce competition.