This paper develops an empirical proxy to capture the markets' expectations of growth and the propensity of investors to extrapolate recent extreme growth too far into the future. The construction of the empirical proxy is predicated on the extant literature that shows that recent sales growth and yield surrogates reflect markets' expectations of growth, which have been found to be too extreme. The paper also re-examines two recent asset pricing anomalies published in the accounting literature—the asset scaled accruals and forecast to price anomalies—in light of the likely connection of accruals to realized growth and the forecast to price ratio acting as a yield surrogate and impounding expectations of growth. I find that the empirical proxy, constructed from recent sales growth and the cash flow to price ratio, produces large abnormal returns to simple hedge strategies. Moreover, the proxy subsumes other expected growth proxies such as the earnings yield, the cash flow to price ratio, weighted average sales growth, and the IBES long-term growth forecast in its ability to explain future returns. Lastly, I find that employing the expected-growth proxy in multivariate returns regressions suggests that overly extreme expectations of growth may be a partial explanation for the abnormal returns generated to hedge portfolios based on the magnitudes of asset-scaled accruals and the forecast to price ratio.

Firms are subject to intra-industry information transfers when one or more of their industry-related peers release an informational event. The information contained in the event not only produces significant price revisions for the announcing firms, but also affects the share prices of their industry rivals. The prior literature on information transfers finds that announcements that carry bad news seem to result in weak information transfers, while those that convey good news seem to be associated with stronger effects. In this context, there is anecdotal evidence suggesting that managers make strategic disclosures to shield their firms from their rivals' bad news. More generally, both theory and empirical evidence suggest that firms disclose information to separate themselves from their rivals. However, no prior studies have investigated the prospect that managers undertake such strategic disclosures in an information transfer context. Using announcements of securities class action litigation and dividend changes, I predict and find that managers of non-announcing firms make strategic disclosures to separate their firms from their industry rivals with bad news and to imitate their rivals with good news. Furthermore, I find that both industry and announcing firms' characteristics are important factors in non-announcing firms' decisions to disclose in this context. I also find that disclosing in the face of bad news considerably reduces negative information transfers while disclosing to imitate rivals' good news enhances positive information transfers. This study offers a novel analysis on the information transfer process by analyzing non-announcing firms' patterns of deliberate interventions, which can be interpreted as the root cause for both the asymmetry and weaknesses in information transfers reported by prior research in this area.

My three-paper dissertation is aimed at applying the concepts of bounded ethicality and ethical fading to accounting fraud. Typical of relatively new fields such as behavioral ethics, theoretical models are scarce (Tenbrunsel & Smith-Crowe, 2008). As such, the purpose of Study 1 is to unify disparate theories and ideas from psychology and behavioral ethics as a means of constructing a theory, the Theory of Unintended Amoral Behavior (TUAB), which includes the concepts of bounded ethicality and ethical fading. In addition, the pressure for management to meet earnings expectations is discussed through the lens of the TUAB as an example of how one may unknowingly misreport. Studies 2 and 3 apply the TUAB to investigate how certain contextual factors interact with egocentric biases to increase the likelihood of ethical fading. Specifically, Study 2 consists of an experiment exploring how inferior pay among managers interacts with egocentric perceptions of fairness and envy to affect the likelihood of one engaging in ethical fading and fraudulent behavior. Study 3 also utilizes an experimental methodology to examine how the pressure to meet earnings forecasts interacts with egocentric perceptions of fairness and negative affect to influence the probability of ethical fading and fraudulent acts. The results for Study 2 indicate that one who is paid at a lower rate is more likely to view this disparity as unfair, which leads to a greater feeling of envy. Although envy had no significant direct effect on ethical fading in the primary analyses, a supplemental analysis revealed that a person's risk preference might moderate this relationship. The primary findings of Study 2 suggest that individuals who experience a higher degree of ethical fading are more likely to commit fraud, and that ethical fading, along with perceived unfairness, seem to be significant psychological processes that explain how differences in pay may lead to fraud. The primary finding of Study 3 is that, like Study 2, fraud is more likely to occur as an individual experiences a higher degree of ethical fading. Furthermore, this study suggests that those who are closest to meeting an earnings target are the most likely to engage in fraudulent behavior. Finally, the results failed to find any support that one's egocentric perceptions of fairness and negative affect contribute towards his or her ethical behavior in a goal achievement setting. The primary contributions of this dissertation is that it unifies various theories and ideas from psychology and behavioral ethics to establish a testable theory (TUAB) that includes the concepts of bounded ethicality and ethical fading, serves as an initial test of TUAB, and provides evidence that unethical behavior is not necessarily the result of one consciously forsaking his or her ethics for some other desired goal (i.e., profit).

This dissertation comprises three studies, one qualitative and two experimental, that center on auditor's use of data analytics. Data analytics hold the potential for auditors to reallocate time spent on labor intensive tasks to judgment intensive tasks (Brown-Liburd et al. 2015), ultimately improving audit quality (Raphael 2017). Yet the availability of these tools does not guarantee that auditors will incorporate the data analytics into their judgments (Davis et al. 1989; Venkatesh et al. 2003). The first study investigates implications of using data analytics to structure the audit process for nonprofessionalized auditors. As the public accounting profession continues down a path of de-professionalization (Dirsmith et al. 2015), data analytics may increasingly be used as a control mechanism for guiding nonprofessionalized auditors' work tasks. Results of this study highlight negative ramifications of using nonprofessionalized auditors in a critical audit setting. The second study examines how different types of data analytics impact auditors' judgments. This study demonstrates the joint impact that the type of data analytical model and type of data analyzed have on auditors' judgments. This study contributes to the literature and practice by demonstrating that data analytics do not uniformly impact auditors' judgments. The third study examines how auditors' reliance on data analytics is impacted by the presentation source and level of risk identified. This study provide insights into the effectiveness of public accounting firms' development of data scientist groups to incorporate the data analytic skillset into audit teams. Collectively, these studies contribute to the literature by providing evidence on auditors' use of data analytics. Currently, the literature is limited to demonstrating that auditors are not effective at identifying patterns in data analytics visualizations when viewed before traditional audit evidence (Rose et al. 2017). The three studies in this dissertation highlight that not all data analytics influence judgments equally.

This dissertation is comprised of three studies investigating the effects of informal management control systems (MCS) and different types of incentive compensation schemes on employees' performance. Prior research describes informal MCS as implicit sets of structures that management adopts to encourage employees to act in a way that aligns with overall organizational goals (Berry et al. 2009). Management usually puts informal MCS in place to inspire self-regulation behaviors among employees; hence, management may not reward or penalize employee behavior that is consistent or inconsistent with this informal MCS (Berry et al. 2009; Christ et al. 2008). Informal controls are implied by social pressures, such as employees' feedback, and management communication, such as a value statement or the organizational culture, where no explicit enforcement measures exist (Berry et al. 2009; Kachelmeier, Thornock and Williamson 2015). The first study examines whether the presence of a value statement (an informal MCS) can be used to motivate employees to perform important, but uncompensated subsequent tasks. The second study extends the first study by examining whether the interactive method of delivery of a value statement (informal MCS) can be used in conjunction with an incentive scheme to improve employees' performance. Lastly, the third study investigates the impact of an important aspect of organizational context, specifically organizational culture (informal MCS), and different types of incentive compensation schemes on strategy surrogation. The first study investigates whether the presence of a value statement (an informal MCS) can be used to motivate employees to perform important, but uncompensated tasks. Additionally, this study seeks to examine the type of incentive scheme that will result in the highest subsequent uncompensated task performance in the presence of an organizational value statement. Considering that incentive contracts cannot completely govern all the employees' responsibilities (Christ, Emett, Summers and Wood 2012), this study investigates how employees will perform their important but uncompensated tasks. The study shows that under fixed pay compensation, the presence of a value statement improves the performance of employees compared to the absence of a value statement. Conversely, under a piece rate incentive compensation, the presence of a value statement negatively influences the performance of employees in the important but uncompensated task. The study also shows that the intrinsic motivation of employees operating under piece rate compensation is more likely to be crowded out by their incentive pay relative to employees operating under a fixed wage. The second study examines whether the interactive method of delivery of a value statement (informal MCS) through electronic integration can be used in conjunction with an incentive scheme to improve employees' performance. Prior research shows that effectiveness of incentive systems is influenced by the presence or absence of a nonbinding value statement in the organization. A value statement is a declaration that communicates an organization's priorities and core beliefs to its customers and employees. Drawing upon the mere-exposure effect, the results of the study show that the employees who experience the interactive delivery of a value statement do not perform significantly better than employees who experience the passive delivery of a value statement. However, employees who receive a piece-rate incentive perform significantly better than employees who receive a fixed pay incentive. As predicted, the method of delivery of an organizational value statement moderates the effectiveness of a fixed pay incentive scheme. The third study draws upon the theory of inattentional blindness to investigate whether different types of organizational culture, control dominant or flexibility dominant, impacts strategy surrogation. Strategy surrogation occurs when managers focus on the measures in the SPMS on which they are compensated and completely or partially lose focus on the overall strategic objectives of the organization (Choi et al. 2012, 2013). Organizational culture is defined as a set of dominant values, beliefs, and assumptions that governs how people behave in organizations (Henri 2006). The results of the study show that there is no significant difference between employees operating under a control-dominant culture and employees operating under a flexibility-dominant culture. Similarly, the type of organizational culture does not moderate the relationship between incentive systems and strategy surrogation. However, employees operating under a pay-for-performance compensation scheme significantly surrogate more than employees operating under a fixed pay compensation scheme. Collectively these studies contribute to management accounting research by examining how different types of informal MCS such as organizational value statement and organizational culture interact with incentive compensation scheme. Specifically, these three studies highlight how and when we can use informal MCS to improve employees' performance as well as their decision making in the organization. Study one contributes to research and practice by highlighting situations where a pay-for-performance incentive scheme may result in unintended consequences. Study two contributes to the management control literature by demonstrating how utilizing technology can enhance the delivery of an organization's value statement and ultimately improve employees' performance. Study three contributes to the incentives and organizational culture literature as well as strategy surrogation research by examining institutional factors that may inhibit or exacerbate surrogation.

This dissertation consists of three studies that explore the potential for relational reasoning to advance research on the facilitation of expertise in complex audit domains. Study One seeks to explicate the potential that theory and methods from relational reasoning and associated research have to advance the audit expertise research stream. The implications for future research on facilitating auditing expertise are discussed in synchrony with future research questions, including whether or not such strategies will be effective in domains with more than minor relational complexity. Studies Two and Three experimentally examine the use of metacognitive skills intended to enhance relational knowledge, which is considered to be a fundamental component of domain expertise. Study Two investigates the effects of alternate forms of prompting for analogical comparison and Study Three explores the impact of combining analogical comparison with direct instruction on discerning the relational structure of a domain. The results of Study Two do not support the expected positive effects of the analogical comparison interventions. Implementation of effective interventions to prompt the comparison requires further research. Additionally, the results of Study Three do not support the hypotheses, by conventional standards. However, there is some evidence of positive effects associated with the analogical comparison intervention. This dissertation contributes to the literature on audit expertise by describing how relational reasoning can play a role in advancing research in this stream and by providing some preliminary information regarding the effectiveness of specific implementations aimed at enhancing relational knowledge.

This dissertation consists of three studies investigating the relationship between business analytics and decision making in accounting. In an effort to improve performance, organizations increasingly emphasize fact-based decision making supported by business analytics, which translate complex data into manageable information through statistical analysis. While the recent focus on business analytics is transforming how managers make decisions, analytics alone do not generate increased performance; the synergy between business analytics and user judgments is a vital component of realizing value. To this end, Study I experimentally investigates how various characteristics of business analytics affect individuals' reliance and perceptions of the analytic. Through the lens of cognitive fit a 2×2 between-subjects experiment is conducted examining business analytics effects of input and process attributes on users' reliance. Cognitive fit theory posits that effective problem solving depends on the match between the technology and the decision process. The second study investigates the impact of management interventions (i.e. actions influencing adoption) toward improving reliance on business analytics. From an organizational perspective, an important concern for management is promoting greater employee acceptance and utilization of business analytics. Building on the Technology Acceptance Model, Study II experimentally examines the effect of management support and consensus of multiple analytics on increasing reliance and on participants' evaluation of the business analytic. Study III further explores the relationship between characteristics of business analytics and the decision maker by developing a theoretical model regarding the effects of perceived decision similarities between the user and the business analytic on users' perceived usefulness. Overall, the results reported in this dissertation suggest that 1) characteristics of business analytics influence users' judgments and decision making, 2) management can take actions to influence the relationship between users and business analytics, and 3) users are likely to evaluate their cognitive similarity to these business analytics, and these perceptions influence perceived usefulness of the business analytics.

Using state charitable tax subsidy rates to proxy for tax incentives, I examine the effect of donors’ charitable tax incentives on the relation between donor contributions and nonprofit organizations’ (NPOs) program efficiency, following Weisbrod and Dominguez (1986) who model a negative relation between charitable tax incentives and NPO program efficiency. I focus my study on NPOs with large donors (i.e., donors that make a single contribution of $5,000 or more) because large donors are more likely to be the beneficiaries of these tax incentives given that the incentives lower their after-tax cost of giving. In addition, NPOs typically appoint large donors to the board, which places these donors in a position to influence NPO efficiency. I use Form 990 Schedule B data to identify NPOs that have one or more large donors, and I find that as state tax rates increase, an increase in donor contributions is associated with higher NPO program efficiency. The results suggest that contrary to the expectations of Weisbrod and Dominguez (1986), tax incentives do not have a negative effect on NPO program efficiency in NPOs with large donors. Rather, my results are consistent with large donors having higher efficiency expectations and being more likely to ensure that the NPO operates efficiently. Given that large donors on the NPO’s board can influence NPO program efficiency and based on the expected negative effect of taxes on the relation between donor contributions and NPO program efficiency predicted by Weisbrod and Dominguez (1986), I further examine whether the effect of donor tax incentives on the relation between donor contributions and NPO program efficiency in NPOs with large donors differ by board size, board independence, and state regulations. While the coefficients indicate that tax incentives have a more positive effect on the relation between donor contributions and NPO program efficiency for NPOs with large board sizes compared to NPOs with small board sizes, Wald statistics indicate that the results do not differ statistically between NPOs split by board sizes. However, the results examining the effect of board independence indicate that NPO program efficiency is higher in NPOs with independent boards compared to those with non-independent boards, suggesting that an independent board has a positive differential impact on how tax incentives affect the relation between donor contributions and NPO program efficiency. Conversely, the results indicate that there is no statistical difference in the effect of tax incentives on the relation between donor contributions and NPO program efficiency between samples split by state regulations. Taken together, the results suggest that while NPO board size and state governance do not impact how tax incentives affect the relation between donor contributions and NPO program efficiency, an independent board has a positive impact on how tax incentives affect the relation between donor contributions and NPO program efficiency. In additional tests, I use changes in state tax rates to examine the effect of tax incentives on the relation between donor contributions and NPO program efficiency. The results show that state tax rate changes (decreases and increases) do not affect the relation between donor contributions and NPO program efficiency. I also conduct tests using federal tax changes to proxy for tax incentives. I first exploit the exogenous shock from the 2017 Tax Cuts and Jobs Act (TCJA) to examine changes in program efficiency, and the results are statistically insignificant, indicating that the TCJA had no effect on the relation between donor contributions and NPO program efficiency. Thereafter, I use the percentage of itemizers in each state as a proxy for tax incentives, and the results also suggest that the percentage of itemizers does not affect the relation between donor contributions and NPO program efficiency. / Business Administration/Accounting

The following dissertation consists of two studies investigating the relationships between industry controversy and accountability. In Study 1, I develop a theoretical framework for identifying industry controversy and I discuss the applications to accountability in social and environmental accounting (SEA) research. The framework consists of criteria to define industry controversy (a difference of opinion at a societal level about a routine feature) as well as two primary theories (organizational legitimacy and organizational stigma) and two secondary theories (utility attribution and stigma transfer) that can explain organizational outcomes. Study 1 concludes with a discussion of areas of accounting research where this framework can further our understanding of accountability within SEA. Future research can examine the formation of conflicting judgments within society, the management of conflict through accountability mechanisms by organizations, and the responses to accountability by various stakeholder groups. In Study 2, I further investigate one of the propositions related to industry controversy discussed in Study 1. When controversies occur, organizations can be left with a stigma that causes stakeholders to distance themselves, yet the organizational stigma research identifies two types of stigma that organizations experience (event stigma and core stigma). Further, independent rating agencies provide score that purport to quantify the severity of a controversy. I explore investor responses to event stigma and core stigma and how controversy scores influence their judgment and decision making. Using non-professional investors, I run a 2x2 between-subjects experiment. Results suggest that investments decrease more when a high controversy score is given to an organization with core stigma (versus event stigma), but that invests perceived organizations with event stigma to be more responsible for the controversy. The results of this study provide a meaningful contribution to the theory of organizational stigma and our understanding of the effects of quantifying complex non-financial indicators.

This thesis presents an empirical evaluation of the validity of the Capital Asset Pricing Model (CAPM) in South Africa. More specifically, the behaviour of share prices on the Johannesburg Stock Exchange during the eight years from 1973 to 1980 is evaluated. The study is the first direct test of the CAPM in South Africa. The methodology employed is a cross-sectional regression technique which has been used successfully in testing overseas security markets. An extension to the usual methodology is made by comparing the results obtained using a I published market-index with those obtained using an internally generated index. The historical development and the derivation of the CAPM is discussed in the thesis, as is the relationship between the CAPM and the Efficient Markets Hypothesis. The results indicate a strong possibility that the CAPM is a valid model in a South African context. Refinements to the research methodology strengthen this conclusion. A potential problem in the interpretation of the results of tests of this sort is also discussed, as is a recent extension to the theory. The overall conclusion is that the CAPM is a valid model, however further research is required to establish this with greater certainty.