Publications:

Information in equity markets with ambiguity averse investors (Review of Financial Studies 22(9): 3595-3627)

This paper shows that persistent mispricing is consistent with a market that includes ambiguity-averse investors. In particular, ambiguity-averse investors may prefer to trade based on aggregate signals that reduce ambiguity at the cost of a loss in information. Equilibrium prices may therefore fail to impound publicly available information. While this creates profit opportunities for ambiguity-neutral investors, ambiguity-averse investors perceive that the benefit of ambiguity reduction outweighs the cost of trading against investors who have superior information. The model can explain both underreaction, such as that evident in postearnings announcement drifts and momentum, and overreaction to accounting accruals.

Reporting bias with an audit committee (The Accounting Review, Forthcoming)
(with Venky Nagar and Paolo Petacchi)

This study models a manager who privately reports earnings to an independent audit committee that, after its own due diligence, modifies the report for public release to investors. The audit committee alters the reporting and valuation dynamics by attempting to remove the manager’s reporting bias, but then presents the information it has collected with its own bias. The audit committee’s presence changes the impact of penalties and incentives on reporting, valuation, and due diligence activities. For example, increasing penalties can sometimes degrade the reporting process. Our simultaneous consideration of the manager, audit committee, and investors provides a new framework for reporting and valuation, and sheds light on empirical earnings quality research that has largely studied the management and audit effects separately.

Working papers:

Efficiency properties of impairment accounting in debt contracting
(with John Hughes)

This study shows that conservatism, in the form of accounting for impairments, can serve to improve the efficiency of project choices and subsequent continuation decisions when firm managers acting on behalf of shareholders have discretion over projects after taking on debt. The model considers projects characterized by stochastic abandonment and continuation value processes where abandonment may occur at an interim date. Debt covenants determine whether the firm’s shareholders or its creditors control the continuation decision. The main result is that impairment accounting can serve as a commitment to avoid inefficient project selection. Because the value of debt and, hence, the value of shareholders’ equity reflects the firm’s anticipated project choice, the firm benefits from this commitment. Noteworthy is that the accounting policy plays a crucial role in that merely tightening thresholds for technical default does not achieve similar efficiency.

Leverage, excess leverage and future stock returns
(with John Hughes and Jing Liu)

We examine the relation between leverage and future stock returns while simultaneously considering the dynamic nature of firm's leverage. Using Graham's (2000) kink measure as a proxy for excess leverage, we find supportive evidence that firm's leverage can be characterized by a partial adjustment model. Excess leverage predicts not only future changes in leverage, but also other fundamentals such as investment and profitability. The market does not seem to fully understand the information contained in excess leverage about future fundamentals (especially investments), and under-levered firms earn superior risk adjusted returns through unexpected growth. The anomalous finding by Penman, Richardson and Tuna (2007), that the relation between leverage and future returns is negative, is subsumed by the negative relation between excess leverage and future returns.

On the estimation of the asymmetric timeliness of earnings: Inference and bias corrections
(with Kyle Peterson)

Recent studies have raised concerns about the validity of cross-group comparisons of regression coefficients capturing the asymmetric timeliness of earnings, a form of accounting conservatism.  In the context of an econometric model, we show that, while regression coefficients correctly identify whether or not earnings are conservative, the inclusion of future rents in equity values causes the regression coefficients to mismeasure the degree of conservatism.  This mismeasurement can bias cross-sectional comparisons of the degree of conservatism; however, a test based on a ratio of regression coefficients allows for valid cross-sectional comparisons.  We empirically validate our predictions and show that the ratio measure is fairly stable over time and is positively associated other conservatism measures that capture asymmetric timeliness.

Do dividends indicate honesty? The relation between dividends and the quality of earnings
(with Michelle Hanlon)

This paper investigates whether dividends provide information about earnings quality. Specifically, we examine whether firms that have lower earnings quality, as measured by an accusation of fraud in a Securities and Exchange Commission Accounting and Auditing Enforcement Release, pay dividends less often (and/or increase dividends less often) than similar firms not accused of accounting fraud. Our results are consistent with the alleged fraud firms being less likely to pay dividends prior to the fraud years. This relation is robust to the inclusion of controls for factors thought to be associated with fraud and dividend policy (e.g., growth, leverage, volatility, age of the firm, and others). We obtain similar, although somewhat weaker, results when we examine the dividends paid during the fraud years and the frequency of dividend increases. Thus, overall the evidence is consistent with dividends indicating earnings quality. However, the data also reveal that the alleged fraud firms pay out a total of over $10.5 billion in dividends, or nearly 3% of their pre-fraud market value, while perpetrating the financial accounting fraud. Thus, while dividends do convey information about earnings quality on average, they do not constitute a preventative measure against financial accounting fraud.