Beatrice Michaeli joined the UCLA Anderson faculty as assistant professor of accounting in 2014 after completing a Ph.D. and an M.Phil. in accounting at Columbia Business School. Her research addresses analytical models in the area of financial and managerial accounting, with a focus on financial reporting, performance measurement, incentives and contracting.
Bulgarian-born and raised in a family of judges and lawyers, Michaeli initially pursued a law degree in Bulgaria, followed by an advanced law degree in Israel, where she learned Hebrew in six months and prepared to embark on an internship in prosecution of white collar crimes. Possessing a broad variety of interests, she became involved in research on the interaction between anti-trust law and industrial organization. She later pursued an advanced degree in business administration, with finance and accounting as major interests. A serendipitous teaching invitation made her realize she enjoyed it, and she accepted the offer to join the faculty of Interdisciplinary Center Herzliya, Israel’s first private university, teaching undergraduate accounting and corporate finance classes. She felt her natural path from there was to pursue a doctoral degree in order to focus on academic research and teaching.
Her dissertation constructs a Bayesian persuasion model to study the interplay between producing and disseminating information. The results call into question the belief that forcing managers to provide unrestricted access to information to all potential users is always beneficial. She says, “If managers could selectively disseminate information they might gather more precise information to begin with and thereby make the users better-off on an aggregate level.” Michaeli’s current research continues to use Bayesian persuasion models to study firms’ information gathering and dissemination choices. For example, one of her recently published papers examines firms’ optimal public reporting when subsequent private information might arrive. A current project considers the investment distortions caused by suboptimal information gathering. Michaeli is also applying a principal-agent framework to explore the link between relationship-specific investments of business units involved in joint projects and compensation risk. The results shed light on the investment distortions and their mitigation through optimally adjusted compensation schemes. She also studies the optimal investment authority allocation and the ability of divisional managers engaged in joint projects to collude against the principal and earn “arbitrage” profits.
Whereas Michaeli’s research hinges on abstract analytical models, in the classroom she draws from concrete case studies to teach corporate incentives and decision-making. “My teaching is very hands-on,” she says.“ I encourage students to observe what’s going on and think of a solution. Sometimes there’s not one super solution; students need to consider under what circumstances one or another specific solution is better. At the very least, I expect them to learn to ask the right questions, to be critical of the default practices that are often applied in corporations, and to be aware of how to make better decisions.”
featured in “Do fair disclosure rules lead to more or less information?" by Michael Totty, UCLA Anderson Review, May 16, 2018
This article develops a Bayesian persuasion model examining a manager’s incentives to gather information when the manager can disseminate this information selectively to interested parties (“users”) and when the objectives of the manager and the users are not perfectly aligned. The model predicts that, if the manager can choose the subset of users to receive the information, then the manager may gather more precise information. The article identifies conditions under which a regime that allows managers to grant access to information selectively maximizes aggregate information. Strikingly, this happens when the objectives of managers and users are sufficiently misaligned. This finding is robust to variations of the model such as information acquisition cost, unobservable precision, sequential noisy actions taken by the users and delayed choice of the subset of users in “the know.” These results call into doubt the common belief that forcing managers to provide unrestricted access to information to all potential users is always beneficial.
We demonstrate a novel link between relationship-specific investments and risk in a setting where division managers operate under moral hazard and collaborate on joint projects. Specific investments increase efficiency at the margin. This expands the scale of operations and thereby adds to the compensation risk borne by the managers. Accounting for this investment/risk link overturns key findings from prior incomplete contracting studies. We find that, if the investing manager has full bargaining power vis-a-vis the other manager, he will underinvest relative to the benchmark of contractible investments; with equal bargaining power, however, he may overinvest. The reason is that the investing manager internalizes only his own share of the investment induced risk premium (we label this a "risk transfer"), whereas the principal internalizes both managers' incremental risk premia. We show that high pay-performance sensitivity (PPS) reduces the managers' incentives to invest in relationship-specific assets. The optimal PPS thus trades off investment and effort incentives.
We study how the potential for discretionary disclosure affects the way a firm designs its reporting system. In our model, the firm's primary but nonexclusive concern is to induce beliefs that exceed a threshold. Such thresholds arise in numerous contexts, including investing decisions, liquidation/continuation choices, covenants, audits, impairments, listing requirements, index inclusion, credit ratings, analyst recommendations, and stress tests. The optimal reporting system is characterized by informative good reports when the threshold is high and, potentially, uninformative reports when the threshold is low. Under an optimal impairment-type reporting system, the likelihood of reported impairments and the information content of non-impairment reports both increase in the probability of the firm observing private information. We provide a novel motivation for the quiet period around an IPO and empirical predictions relating the probability of discretionary disclosure to the properties of financial reports. In extensions, we consider disclosure mandates, report manipulation, endogenous thresholds, and alternative payoff functions.
Integrated ownership is often seen as a way to foster specific investments. However, even in integrated firms, managers invest to maximize their compensation which is chiefly driven by divisional income. Thus it is not clear that integration has any effect on investments in a world of decentralized decision-making. Building on recent findings that efficiency-enhancing investments raise not only the expected value of a project but also its variance, we show that under plausible conditions integration calls for low-powered incentive contracts: the managers invest more as they are less exposed to the investment-related (endogenous) risk, and the principal of an integrated firm has more to gain from greater investment. On the other hand, integration may result in higher-powered incentives if the project is inherently very risky or if the project-specific input is personally costly to the managers (rather than a monetary investment). The qualitative takeaway remains, however, that the contract adjustments under integration mitigate any input distortions present under non-integration. We also allow for firmwide performance evaluation under integration and show that it may lead to larger input distortions, but those are outweighed by improved risk sharing.
We consider the optimal allocation of decision rights in an incomplete contracting setting where business unit managers choose inputs that enhance the efficiency of "joint projects" (projects that benefit their own and other divisions). With scalable project inputs, decision rights should be bundled in the hands of one division manager. Which of the managers to designate the investment center manager-the one facing the more volatile or the more stable environment-depends on whether the project input is a monetary investment or personally-costly effort. With discrete project specific inputs, on the other hand, it is always optimal to split decision rights symmetrically between the managers provided they face comparable levels of operating volatility. This runs counter to the conventional wisdom that bundling stimulates the provision of complementary inputs. The model also generates empirical predictions for the association of organizational structure and managers' relative incentive strength: bundling of decision rights results in PPS divergence across divisions; splitting them results in PPS convergence.
The aim of general purpose financial reporting is to provide information that is useful to investors, lenders, and other creditors. With this goal, regulators have tended to mandate increased disclosure. We show that increased mandatory disclosure can weaken a firm's incentive to acquire and voluntarily disclose information. Specifically, we provide conditions under which a regulator, seeking to maximize the total amount of information provided to investors via both mandatory and voluntary disclosures, would impose an imperfectly informative mandatory reporting regime even when a perfectly informative regime entails no direct costs. The results contribute to our understanding of potential interactions between mandatory reporting and voluntary disclosure, and demonstrate a novel benefit of accounting standards that mandate imperfectly informative reports.
We study how a firm owner motivates a manager to create value by optimally designing an information system and a compensation contract based on a manipulable performance measure. In equilibrium, the firm either implements a perfect or an uninformative system. The information system and the pay-performance sensitivity (PPS) of the compensation contract can be substitutes in a sense that the firm optimally combines a perfect information system with a low PPS or an uninformative system with a high PPS. Because the information design is endogenous, firms facing relatively high manipulation threat may offer financial incentives that are higher-powered than the ones offered by their peers facing lower manipulation threat. If the manager is in charge of implementing the information system, he chooses a perfect one unless the firm uses the information for internal control. The firm may prefer to commit to an internal control level before observing any information.
(8) How Does Media Coverage Affect Corporate Disclosures?
With Elyashiv Wiedman
This paper considers the effect that external sources of information, such as media, have on firm disclosure. A manager, who may be endowed with accurate information about the value of the firm, decides whether to voluntarily disclose this information to the market. The manager maximizes firm price, which reflects the expected firm value, conditional on the disclosure and a media report. The media representative (journalist) always observes and truthfully communicates, in a news report, a noisy signal of the firm value. We focus on the dynamic interaction between the manager’s disclosure and the news report. If the manager discloses before or simultaneously with the news report, he or she withholds some values that would have been disclosed in the absence of media coverage. We further find that, if the manager waits until after the media report, he may respond with a disclosure of some information that is less favorable than the media report but may also withhold some more favorable information. As one would expect, if the manager does not respond to the news report, the price drops. However, we find that the price may drop even if the manager responds. This happens when the manager discloses values that, while decreasing the price, still keep it above the price that would have prevailed had the manager remained silent
(9) Antagonists, experts, or both? Board composition when inattentive boards are persuaded
With Martin Gregor
We study a CEO's choice of a technology that disseminates information about the prospects of a project to a board of directors. In our model, the CEO proposes the project and the board decides whether and how much additional information to collect at an entropy-based personal cost before approving or rejecting the project. The higher the expertise of the board, the lower the cost of acquiring additional information. The CEO is biased towards undertaking some projects that destroy firm value whereas the board prefers to reject some projects that increase firm value, i.e., the board is antagonistic. The higher the board's expertise and antagonism, the more informative the signaling technology introduced by the CEO and the fewer the approval errors. However, searching for directors with higher expertise and antagonism is costly to the shareholders. Notably, we find that the board does not gather additional private information in equilibrium. We show that the board's expertise and antagonism are substitutes in reducing the approval error and predict that, in industries with higher information acquisition costs (e.g., new industries), the shareholders appoint more antagonistic boards to substitute for their (lack of) expertise.
Traveling, swimming, running and yoga