Harvard Business School

This paper investigates how recent institutional developments impact the potential channels, and thus available discretion, by which managers can manipulate reported fair values. First, I use extensive field research to document the mechanisms used by banks to procure and report fair values—particularly incorporating the impact of the 2011 FINRA’s Trade Reporting and Compliance Engine (TRACE), and concurrent increase in independent third-party vendors. Key insights include that (i) banks predominantly apply third-party vendors’ feeds to generate financial statements (with nearly 100% of vendors’ feeds passing automatically to reported financial statements, with only rare adjustments); and (ii) external auditors predominantly relying on different vendors’ prices to verify and challenge banks’ inputs. Second, I employ three proprietary datasets of daily financial-instrument level pricing—capturing both TRACE and third-party vendors—to document the following insights. I find that vendors’ evaluated prices dominate historical costs in all performance metrics, confirming they provide a more accurate, objective, and reliable proxy for fair value than historical cost. I also find that vendors’ fair values are value-relevant and account for 90% of the trade-to-trade price variance, creating an upper bound on managerial discretion (of only 15% of the original level). Finally, I find that bank managers respond to these newly imposed constraints by alternatively engaging in more spoofing-transaction based fair value manipulations: suggesting this is a likely (even primary) channel by which manipulation can be attained. Overall, the evidence suggests that fair values, particularly after the above institutional developments, appear less subjective, less costly to implement, and more convenient for auditors to verify and challenge, than the literature has previously reported.

Firms spend significant resources on compliance training, but it is often criticized as being cosmetic. Using proprietary records on compliance training and allegations of misconduct from a large multinational firm, I investigate whether compliance training decreases corporate misconduct. I find that in-person training impacts employee behavior but do not find evidence for video training, which indicates that compliance training can be effective when employees are attentive. However, this effect lasts for only two months, suggesting that compliance training only temporarily raises awareness. I also find that the effectiveness of training is curtailed by employees’ economic incentives to misbehave, such as high performance pressure and weak public enforcement. Overall, this study advances our understanding of how and when compliance training can impact employee behavior and describes its limitations.

This dissertation comprises of two essays on the accounting rules and regulations. The first essay explores the accounting rules for bank loan loss recognition. Motivated by the FASB's new proposal that introduces an expected loss methodology for recognizing losses, it examines two questions that are related to the current GAAP and the new accounting rule. It develops an empirical model of loan loss prediction from the machine learning literature, and shows that it outperforms current GAAP. It then demonstrates the value of expanding the inputs to the model, as proposed by the new rule. Finally, it examines the drivers of the performance difference between the developed model and the current GAAP. The second essay studies the regulation around shareholder activism and the proposals that they submit for firms. In particular, it considers the proposals that managers seek to exclude from the proxy statement. Using a hand-collected data set of SEC ``no-action'' letters, it documents that the shareholder proposal mechanism has a broader set of components than that considered by prior research, and provides a number of empirical regularities. It further documents that shareholder proposals are part of a larger mosaic of shareholder intervention in companies that often go together.

In these essays, I explore the role of accounting information and governance in emerging institutions. In the first essay, “Credibility of Disclosures in Weak Enforcement Institutions: Evidence from Shanghai-Hong Kong Connect,” I study whether voluntary disclosure can be credible when the enforcement institutions to deter managers from engaging in cheap-talk are weak. Using the case of China, I examine the effect of a market liberalization pilot program’s announcement, which increased foreign investors’ future ability to invest in select Shanghai stocks, on affected firms’ disclosure policies. I find that affected firms did not change public disclosure (press releases and management forecasts), but significantly increased private disclosure (corporate access and private dial-ins) in anticipation of the program’s implementation. Private disclosure increases were concentrated among firms in need of capital and these firms experienced an increase in foreign institutional holdings after the implementation. Further, their stock prices suffered less during a subsequent market crash and they retained more foreign institutional investors. Overall, the results suggest that voluntary disclosure supports investor confidence even in weak environments, albeit through private (instead of public) channels. In the second essay, “Corporate Sustainability: First Evidence on Materiality,” co-authored with Mozaffar Khan and George Serafeim, we develop a novel dataset by hand-mapping sustainability investments classified as material for each industry into firm-specific sustainability ratings using newly available materiality classifications of sustainability topics. This allows us to present new evidence on the value implications of sustainability investments. Using both calendar-time portfolio stock return regressions and firm-level panel regressions, we find that firms with good ratings on material sustainability issues significantly outperform firms with poor ratings on these issues. In contrast, firms with good ratings on immaterial sustainability issues do not significantly outperform firms with poor ratings on the same issues. These results are confirmed when we analyze future changes in accounting performance. The results have implications for asset managers who have committed to the integration of sustainability factors in their capital allocation decisions. In the third essay, “Shareholder Activism on Sustainability Issues,” co-authored with Jody Grewal and George Serafeim, we examine shareholder activism on sustainability issues which has become increasingly prevalent over the years, with the number of proposals filed doubling from 1999 to 2013. We use recent innovations in accounting standard setting to classify 2,665 shareholder proposals that address environmental and social issues as financially material or immaterial, and we analyze how proposals on material versus immaterial issues are related to firms’ subsequent environmental or social performance and market valuation. We find that 42 percent of the shareholder proposals in our sample are filed on financially material issues. We document that filing shareholder proposals are related to subsequent improvements in the performance of the company on the focal environmental or social issue, even though such proposals nearly never received majority support. Improvements occur across both material and immaterial issues. Proposals on immaterial issues are associated with subsequent declines in firm valuation while proposals on material issues are associated with subsequent increases in firm value. We show that managers increase performance on immaterial issues in companies with agency problems, low awareness of the materiality of sustainability issues, or poor performance on material issues.

In this dissertation, I utilize proprietary field data to examine two different management control mechanisms used by organizations seeking to optimize fit in their personnel decisions. I also describe my experiences conducting field-based research in management accounting. In the first essay, “Beyond Performance: When Potential Matters to Employee Career Outcomes”, I examine how managers’ assessments of employee potential (i.e. promotion prospects) are related to employee career outcomes. I first document that (as intended) potential ratings play an important role in managers’ promotion decisions, above and beyond performance ratings, and that these ratings also play an incremental role in termination decisions. Next, I examine how potential ratings are related to employees’ voluntary separations. Amongst newly-hired employees, I find that the likelihood of an employee voluntarily leaving the organization is decreasing in rated potential, and that an upward revision in potential is associated with a reduced likelihood of leaving. Conversely, for longer-tenured employees, voluntary departures are unrelated to potential ratings, except that the likelihood of departure is greater following a downward revision in potential. Finally, I investigate how the aggregate potential of an organization’s employee base evolves over time. I find that with the passage of time the organization experienced an increase in the proportion of employees assessed as “high potential”, attributing this to both employee selection effects and motivational effects. In the second essay, “Who Should Select New Employees, the Head Office or the Unit Manager? Consequences of Centralizing Hiring at a Retail Chain”, co-authored with Tatiana Sandino, we examine the allocation of hiring rights in the employee selection process. Specifically, we examine whether centralized hiring (in our study, by the head office of a U.S. retail chain) or decentralized hiring (by store managers) leads to higher quality employee-company matches. In addition to examining the main effect of centralized hiring on match quality, we develop and test hypotheses pertaining to various store characteristics that may moderate the effect of centralized hiring – specifically, instances where headquarters may have a hiring advantage relative to store managers, and instances where store managers may possess an informational advantage relative to headquarters. While we find no evidence of a main effect of centralized hiring, we do find evidence consistent with our moderating hypotheses – specifically that centralized hiring leads to higher quality matches when store managers are overly busy, while centralized hiring leads to relatively lower quality matches when the store manager is at an informational advantage due to serving a divergent market or repeat customers. In the third essay, “Field Studies in Management Accounting”, I describe my experiences conducting field-based research in management accounting (focusing in particular on the studies featured in the abovementioned essays). In so doing, I provide an overview of each of the main phases involved in a typical field study, discuss lessons learnt, and share tips for other researchers considering conducting a field-based empirical research study.

Growing investor interest in sustainability performance and recent regulation have led to increased firm disclosure of environmental (i.e., carbon emissions, water consumption, waste generation, etc.), social (i.e., employee and workplace practices, product safety, etc.), and governance (i.e., political lobbying, anticorruption, etc.) information. This dissertation examines market and non-market forces that shape corporate sustainability disclosure and investment. The first essay studies voluntary sustainability disclosure across financial and sustainability reports and sheds light on investors’ interpretation and use of this information. The second essay examines whether mandated disclosure of greenhouse gas emissions leads to emissions reductions among firms already disclosing prior to regulation. The third essay provides the first empirical evidence on investor perceptions of mandated sustainability disclosure.

This dissertation addresses how private and public performance reports affect performance and the information content of performance measures. First, I show how disclosing consumer ratings to the general public affects performance and biases raters. Using data from a health care system, I find that publicly disclosing patient ratings of physicians leads to: 1) performance improvement by the ratings and by objective measures of quality, and 2) a bias among raters, who positively weight a physician’s published average rating in deriving subsequent ratings for the physician. To understand the moderating effects of public attention, I use variation in web traffic to a physician’s disclosed rating. I find evidence consistent with public attention reinforcing raters' bias toward concurring with a physician’s published average rating, thus impeding rating improvement. Within a national distribution of ratings, the disclosure leads to an improvement in ratings by 17 percentile points and a bias in a given physician’s ratings toward his or her published average rating by 24 percentile points. These findings demonstrate that consumer-rating disclosure is a means of performance management, and that resulting bias is a reason to interpret subsequent trends in ratings as understated signals of trends in service. The second section of the dissertation shows an understudied and low-cost way of customizing private performance reports to best drive reported performance, and warns that the private reporting causes reported performance to diverge from unreported performance. A field experiment reveals the performance benefit of customizing a private performance report to include the peer-performance reference point that will most motivate improvement. The below-average performers improve most when shown the median as a reference point. The 50th-75th percentile performers improve most when shown the top-quartile as a reference point. The top-quartile performers improve most when shown the top-quartile as a reference point, but only when reported performance is outcome-based as opposed to process-based. Neither the median nor top-quartile reference point has a more positive performance effect overall. With regard to the performance measure’s content, privately reporting a measure causes the measure to become less correlated with unreported performance. These findings have the following implications. First, the optimal reference point for peer performance comparison depends on 1) an individual’s initial performance relative to each reference point, and 2) whether the performance measure regards an outcome or process. Second, a performance measure, once reported, becomes a less informative signal of unreported performance.

The management accounting literature investigates the role of control systems in maximizing organizational performance by facilitating managerial decision-making and shaping employee behaviors. Despite the use of diverse management control systems in practice, there is a lack of systematic empirical evidence that examines their effectiveness and (un)intended consequences. In my dissertation, I study the use of different management control systems and their effect on organizational performance by using proprietary data from different organizations. Thereby, my work contributes to the literature on the design of management control systems that can inform corporate leaders and managers on how to deliberately shape organizational outcomes.

In U.S. M&A transactions, target directors are effectively required to seek and consider a “fairness opinion” and supporting valuations before accepting a takeover offer. Despite their legal status, critics have argued that these valuations are biased, and that even unbiased valuations would not be useful to public companies, which can use their pre-deal stock price as an appraisal of their value. I develop and implement tests for the usefulness and the bias of fairness opinion valuations, and find tight evidence for both. They impound information about fundamental mispricing in targets, and prospective deal synergies, which could make them useful to directors in exercising their duties. They also exhibit predictable bias: providers toggle their discount-rate assumptions ex post to rationalize negotiated deal prices. However, this bias has been constrained by judicial scrutiny in recent years. These findings suggest that third-party appraisal could have a useful role in M&A governance, and suggest avenues for reform.