Jeremy Bertomeu

From Casual to Causal Inference in Accounting Research: The Need for Theoretical Foundations

On December 5th and 6th 2014, the Stanford Graduate School of Business hosted the Causality in the Social Sciences Conference. The conference brought together several distinguished speakers from philosophy, economics, finance, accounting and marketing with the bold mission of debating scientific methods that support causal statements. We highlight key themes from the conference as relevant for accounting researchers. First, we emphasize the role of formal economic theory in informing empirical research that seeks to draw causal inferences, and offer a skeptical perspective on attempts to draw causal inferences in the absence of well-defined constructs and assumptions. Next, we highlight some of the conceptual limitations of quasi-natural experimental methods that were discussed at the conference, and discuss the role of structural estimation. Finally, we illustrate many of the points from the conference by estimating a novel, theoretically-grounded measure of disclosure costs. ___________________________________________ We thank an anonymous referee, Chris Armstrong, Qi Chen, Paul Fischer, Joseph Gerakos, Ian Gow, Wayne Guay, David Larcker, Christian Leuz, Ivan Marinovic, Jeremy Michels, Valeri Nikolaev, Ro Verrecchia, and Ivo Welch for helpful conversations and comments. We are grateful to the conference organizers for giving us the opportunity to write this piece.

Toward a Positive Theory of Disclosure Regulation: In Search of Institutional Foundations

This article develops a theory of standard-setting in which accounting standards emerge endogenously from an institutional bargaining process. It provides a unified framework with investment and voluntary disclosure to examine the links between regulatory institutions and accounting choice. Disclosure rules tend to be more comprehensive when controlled by a self-regulated professional organization than when they are under the direct oversight of elected politicians. These institutions may not implement standards desirable to diversified investors and, when voluntary disclosures are possible, allowing choice between competing standards increases market value over a single uniform standard. Several new testable hypotheses are also offered to explain differences in accounting regulations.

Incentive Contracts, Market Risk, and Cost of Capital

Should incentive contracts expose the agent to market-wide shocks? Counterintuitively, I show that market risk cannot be filtered out from the compensation and managed independently by the agent. Under plausible risk preferences, the principal should offer a contract in which performance pay increases following a favorable market shock. In the aggregate, however, the effect of market risk on individual contracts diversifies away and the agency problem does not directly affect the cost of capital. The analysis suggests caution in interpreting changes in cost of capital in terms of the stewardship role of accounting information.

How Often Do Managers Withhold Information

We estimate a dynamic model of voluntary disclosure, using annual management forecasts of earnings, that features a manager with price motives and an uncertain but persistent information endowment. Our estimates imply that: (i) managers face disclosure frictions 35% of the time; (ii) conditional on being informed, managers withhold information 17% of the time; and (iii) conditional on being silent, managers possess information 24% of the time. Managers’ strategic withholding motives increase investors’ uncertainty about earnings by 3%. We find that managers’ price motives reduce strategic withholding by one-third, in response to investors’ increased skepticism in the event of non-disclosure.

Economic Consequences of Equity Compensation Disclosure

The primary role of equity compensation is to provide incentives to an effort-averse agent. Here, the author shows that the chosen level of equity incentives, when publicly disclosed, will also convey information about future earnings, causing two-way linkages between incentive compensation, and financial reporting. If (a) market prices respond more (less) to information, (b) the manager is more (less) risk averse, or (c) earnings are more (less) noisy, then the firm’s owners choose more pronounced (muted) incentives, in turn leading to greater (lower) future earnings. The model explains observed spurious correlations between firm performance and executive compensation, and it provides several new predictions linking managerial, earnings, and market determinants to optimal equity holdings.

A Simple Theory-Based Estimation of Earnings Management

We derive a measure of earnings management cost and the associated equilibrium level of earnings management from the cross-sectional properties of earnings and prices. This approach enables us to separate economic shocks from reporting discretion by modeling the economic trade-off faced by management. The trade-off can be easily estimated from a closed-form likelihood function. Overall, the estimates suggest that earnings management is modest, consistent with the conjecture in Ball (2013) that earnings management is not as rampant as what prior research would suggest. Consistent with prior studies, the measure suggests more earnings management during seasoned equity offerings, for smaller and growing firms, as well as in industries with more irregularities.We derive a structural estimator of earnings manipulation based on signaling theory, incorporating cross-sectional properties of earnings and prices. Identification requires partial knowledge of the distribution of true earnings and non-linearities in the relationship between earnings and prices. As an application, we estimate manipulation costs and reporting biases before vs. after the Sarbanes-Oxley Act of 2002. We also characterize implied changes in reporting bias across industries, subsamples of firm size and growth opportunity and over time, and propose an identification strategy that relies on exogenous shocks to misreporting costs. Our empirical model extends to other settings provided reporting benefits are observable to the researcher, such as environments with costly communication.

Voting over disclosure standards

Abstract This article examines the nature of disclosure standards, under the assumption that (i) standards preferred by more firms are collectively chosen and (ii) privately informed firms prefer standards that increase market perceptions about the value of their assets. A standard is stable if it is preferred by a large enough super-majority of firms over any other standards. Absent any restriction on possible standards, only unanimity would make a standard stable. By contrast, when requiring standards that classify news from best to worst, there is at most a single stable standard, and it must be full disclosure. For a large class of distributions over valuations, the required super-majority is about two-thirds, close to the majority required in many standard-setting boards. Value distributions with heavy tails, such as news that contains extreme risks, require higher super-majorities to be stable. These insights are robust to settings in which the information is used in decision-making.

On the Possibility of Socially Accepted Accounting Standards

This article examines the nature of disclosure standards, under the assumption that (i) standards preferred by more firms are collectively chosen and (ii) privately informed firms prefer standards that increase market perceptions about the value of their assets. A standard is stable if it is preferred by a large enough supermajority of firms over any other standards. Absent any restriction on possible standards, only unanimity would make a standard stable. By contrast, choosing standards that classify news from best to worst, there is at most a single stable standard, and it must be full disclosure. For a large class of distributions over valuations, the required supermajority is about two thirds, close to the majority required in many standard-setting boards. Distributions with heavy tails, such as news that contains extreme risks, require higher super majorities to be stable. These insights are robust to certain settings in which the information is used in decision-making.This article examines the nature of accounting standards under super majority voting by parties informed about the value of their asset. A standard is stable if it is preferred by enough voters over any other standard. Absent any restriction on the space of possible standards, we show that there is no stable standard regardless of the super majority. Restricting the space of standards that classify news monotonically, there exists a super majority such that the unique stable standard is full disclosure. For a large class of common distributions, the required super majority is about 63%.

Disclosure Policy and Industry Fluctuations

This paper examines voluntary disclosures in a repeated oligopoly and their association with price-setting behavior and industry profits along industrial fluctuations. The analysis focuses on the collectively optimal equilibrium among oligopoly firms. We show that, in industries that are highly concentrated or feature low cost of capital, nondisclosure is prevalent and results in stable product prices and high profit margins. Otherwise, firms may selectively disclose to soften competition in the product market. Under partial disclosure, firms withhold information during sharp industry expansions or declines. Consequently, the disclosure policy dampens the dissemination of shocks to the industry. This paper was accepted by Mary Barth, accounting.