In the first chapter, I investigate the relationship between the number of outstanding public debt contracts of a firm, the firm's credit quality, and cost of debt. I find that firms with higher credit quality tend to use fewer public issues, as well as that firms with more issues tend to have a lower cost of debt, controlling for credit quality. My results are consistent with the idea that there are both costs and benefits to increasing the number of public debt contracts. Higher credit quality firms, starting out with a lower cost of debt, find the benefits insufficient to make up for the costs, and thus choose to have fewer debt issues than lower credit quality ones. I further find that information asymmetry is a significant moderating factor of the effect of number of debt issues on the cost of debt, with higher asymmetry decreasing the cost of debt benefit of a greater number of issues. In the second chapter, I investigate the impact of firm-level competitive intelligence on the firm's cost of debt. I find that competitive intelligence is not fully incorporated into debt credit ratings, and further that the effect of increased competitive intelligence varies with firm credit quality. For high credit quality firms, I find that higher CI is associated with higher yield spreads, while the opposite is true for the lower credit quality firms. This suggests that the bondholders of a firm with generally low distress probability view CI expenditure as irrelevant or wasteful, whereas those of a firm for which financial distress is a more significant risk, view it as a valuable activity which reduces default probability. In the third chapter, I examine the occurrence of informed trading in public debt issued by financial institutions. The sample is chosen from the set of firms subject to the FDIC call report regulations, and focuses on companies without publicly traded equity. I find that unexpected earnings are positively associated with price changes in debt instruments as a result of trading within the time period after report filing and before the release of report data to the public. Additionally, I find that the magnitude of the effect is greater for firms without public equity. Evidence further indicates an increase in the incidence of bond trading during this blackout window for firms with a greater magnitude of earnings surprise. These results suggest that there is information leakage taking place during the blackout window, leading to informed trading in public debt instruments of financial institutions. / Business Administration/Finance

Agency theory suggests that conflicts of interest between managers and the suppliers of finance (shareholders and debt-holders) can cause considerable costs on the firm. This study investigates the role of financial contracts and corporate governance in mitigating agency conflicts. Chapter 1 examines the effect of creditor control on CEO compensation. We present evidence that creditor control has significant impact on CEO compensation. CEOs experience a sharp cut of 17% of excessive pay following financial covenant violations. Differences-in-differences test shows that the reduction in abnormal CEO compensations is only associated with violation firms, not with their matched non-violation peers during the same time period. Furthermore, we find that the cut in excessive pay upon violations is greater in firms facing stronger creditor control, i.e., firms borrowed from banks with which they have a stronger prior lending relationship or high reputation banks. Despite the fact that the prior literature has documented greater CEO compensations in firms with weaker shareholder governance, we find that shareholder governance has little significant impact on the reduction of abnormal CEO compensations following debt covenant violations. In addition, we find that managerial pay-risk sensitivity (vega) is significantly reduced after covenant violation, particularly in the presence of greater creditor control power. In contrast, covenant violations are not associated with any significant change in managerial pay-performance sensitivity (delta). Chapter 2 investigates the impact of creditor control on corporate innovation via the lens of corporate events - debt covenant violation, where control right is shifted from equity-holders to creditors. By employing differences-in-differences tests, we document that firms experience a significant cut in corporate innovation following financial covenant breaches, especially in innovation intensive industries. Furthermore, we show that creditor control plays a direct role in curbing corporate innovative activities upon covenant violations. We find that in the presence of stronger bank control, violation firms experience a significantly larger reduction in both the quantity (as measured by number of patents) and quality (as measured by non-self citations received) of innovations. Interestingly, we find that banks' expertise in certain innovative industry can moderate the adverse effect of creditor control on innovations in those industries. These results are consistent with the argument that banks are less tolerant of failures and debt covenants restrict manager flexibilities. Our findings also suggest that banks' experience, knowledge, and expertise in certain innovative industries allow them to have a better assessment about borrowers' innovative projects, and thereby mitigating the agency conflict. Chapter 3 examines the association between managerial time horizon and corporate hedging. We document that CEO's managerial time horizon has a significant effect on firms corporate hedging policy. CEOs are more likely to use derivatives and use significantly more derivatives when they approach retirement, i.e., when they have a short horizon. Propensity score matching method suggests that this finding is not driven by sample selection problem. We find that increases in derivative hedging are results of CEOs' pension entitlement. Considering future pension payments, CEOs have greater incentive to limit firm risk so as to reduce the probability of bankruptcy as they approach retirement. Furthermore, we find that increase in hedging activities is restricted in firms with strong corporate governance (e.g., weak anti-takeover provision, non-dual CEO and high institutional investor holding), suggesting that increased hedging does not benefit shareholders. / Business Administration/Finance

The literature of sociobiology and culture recognize that, statistically, females often make different choices than males across a wide range of issues. Scholars of business, economics, and finance find that females react differently than males to diverse financial and business situations. Moreover, extant research indicates that females on boards of directors exert a positive impact on monitoring, value, and performance. This dissertation extends the gender literature by empirically testing the hypothesis that female board representation limits the use of debt in firms’ capital structures because of females’ greater risk aversion, lower overconfidence, and less competitive nature compared with males. The empirical results indicate that influential female representation, such as a female chair of the board, has a causal negative and significant impact on the leverage of the company. / Business Administration/Finance

The ability to predict corporate bankruptcy is critically important to investors, creditors, borrowing organizations and governments alike. Bankruptcy occurs when an organization is unable to afford its financial obligations or pay its creditors. While research has illustrated the role of financial ratios on predicting bankruptcy, social factors are largely not considered an effective element. In this paper, I investigate the social and human capital determinants of bankruptcy and explore them as new avenues for enhancing predictive power. Specifically, this study develops new models for predicting bankruptcy based on non-financial factors. The two social variables that I examine are (1) networking ability as a proxy of social capital and (2) the power of managers based on their education as a proxy of human capital. I also added to the Altman’s and Zmijewski’s models with two categorizes of financial variables, the first of which includes five that are financially based on the Altman model, and second three that are financially based on the Zmijewski model by industry and year fixed effect. The results demonstrated a significant and negative relationship between social and human capital and bankrupt companies, the most financial ratios of Altman and Zmijewski are also significant. The results are confirmed using Logistic regression, Cox Proportional Hazard Model and Neural Network. / Business Administration/Finance

With the continued increase in executive compensation and the resulting increase in pay disparity between executives and rank-and-file employees, CEO compensation relative to the average worker pay underscores the popular apprehension related to equity vs. efficiency. This dissertation empirically examines the determinants and consequences of the CEO-worker pay ratio, and the association between acquisitions and CEO compensation. In the first chapter, I show that country-level factors such as national culture, matter in determining the CEO-worker pay ratio across countries. Using global firm-level data from 44 countries for 2002-2015, I show that the CEO-worker pay ratio is associated with national characteristics such as culture and societal equity orientation. Specifically, I find that the CEO-worker pay ratio is positively associated with power distance and masculinity of the national culture, and it is negatively associated with uncertainty avoidance and long-term orientation. This pay ratio also reflects societal equity orientation, measured by income and wealth distribution proxies. This chapter contributes to the existing literature on executive compensation by documenting national culture as an important determinant of CEO-worker pay ratios globally. In the second chapter, I show that the relationship between firm value and the CEO-worker pay ratio exhibits an inverse-U shaped relation, which is consistent with elements of tournament theory, efficient contracting theory, rent extraction theory, and equity fairness theory. This is also consistent with there being an optimal pay ratio, or inflection point, beyond which increases in the pay ratio decrease firm value. I then show that the relationship between the pay ratio and firm performance differs systematically with regard to firm characteristics, i.e., in firms with a greater need for collaboration and information sharing, the optimal ratio is lower. In the final analysis, I show that pay disparity within the executive suite has little effect on firm value, rather it is the pay disparity between the named executive officers and the rank and file that drive my results. This chapter contributes to the existing literature by showing that the relationship between firm value and CEO-worker pay ratio is nonlinear. In the third chapter, I examine the effect of acquisitions, especially international acquisitions on CEO compensation, using firm-level panel data for 1995-2016, covering both international and domestic acquisitions by US firms. I find that acquisitions lead to higher CEO compensation, which can be explained by size premium, complexity premium, and opportunism. I also find that international acquisitions lead to higher CEO compensation than domestic acquisitions, which is consistent with the matching theory, as international acquisitions are larger and more complex to manage. This chapter provides direct empirical evidence on the effect of acquisitions on CEO compensation with a large database based on US firms. This chapter also adds to the literature on the comparison of international acquisition and domestic acquisition in terms of their impact on CEO compensation, which has been lacking in existing work. Overall, this dissertation advances our understanding on the determinants and consequences of the CEO-worker pay ratio, and adds insights to the literature on the implications of international acquisitions and managerial compensation. / Business Administration/Finance

Recent literature provide widespread and robust evidence on the impact of corporate governance. Ownership structure and management characteristics are among the center of the debate. Empirical studies report conflicting evidence regarding the information environment of public family-controlled firms. We use staggered exogenous shocks to the information environment to test whether family control influences corporate disclosure. After an exogenous decrease in the information environment, we find that family firms provide greater, more informative, and more rapidly produced disclosures than their nonfamily peer firms. Family control increases the likelihood of voluntary disclosure by 190% relative to nonfamily firms after a negative information shock. These disclosure increases occur across founder-, descendant-, and externally- led family firms, suggesting families possess strong incentives to protect the firm’s information environment. Beyond ownership structure, I examine the relation of CEO overconfidence on compensation incentive. My findings suggest that the cost-reduction hypothesis applies when firms offer higher incentive to overconfident CEOs to exploit their positively biased views of firm performance; risk-reduction hypothesis dominates when CEOs are extremely overconfident, where firms offer reduced compensation convexity to lower CEO’s excessive risk-taking incentive. Extremely overconfident CEOs receive less convex compensation than moderately overconfident CEOs and this relation amplifies with history of value-destroying acquisition and better corporate governance. / Business Administration/Finance

A series of high-profile corporate fraud scandals in the early 2000s have drawn the attention from the public, regulators, and academia. These cases of the high-profile corporate fraud imply that the existing institutions are lack of incentives and monitoring. Therefore, this study aims to investigate the effectiveness of different governance mechanisms in limiting the fraud propensity. Chapter 1 investigates whether monitoring by non-CEO executives can effectively reduce the likelihood of CEOs committing corporate fraud. Controlling for other traditional governance mechanisms, we find that firms with stronger non-CEO executives monitoring have a lower probability of committing fraud. Monitoring by non-CEO executives appears to be a substitute for traditional governance channels, as it is more effective when traditional governance mechanisms are weak. Moreover, we argue that monitoring by non-CEO executives fails to prevent corporate fraud if both CEO and subordinate executives involve in the fraud event. Finally, the strength of such monitoring is larger in more heterogeneous industries, where the human capital of non-CEO executives is less replaceable. Chapter 2 examines the association between employee relation and the firm's incentive of committing fraud. We find that firms treating their employees fairly (as measured by employee relation ratings) have less incentive in committing fraud. Better employee relation facilitates interest alignment between shareholders and the management. Moreover, we find that the CEO duality weaken the negative association between employee relation and the likelihood of fraud commitment. Furthermore, we find that the negative association is more pronounced in R&D-intensive industries, where human capital is more valuable to firm performance. The results are robust to alternative models and measures. Chapter 3 examines the association between corporate political connection and corporate fraud, and its detection, in China for 2003-2009. Using the enforcement action data from the Chinese Securities Regulatory Commission (CSRC), we find that corporate political connection is an important determinant of corporate fraud, while the type of ultimate owner is also relevant. Politically connected firms are 27% less likely to be detected by the CSRC conditional on their fraud commitment. Low detection rate in turn implies that politically connected firms have 23% more probability to commit fraud than non-connected firms. Government controlled firms are 21% less likely to be investigated by the CSRC. However, due to the irrelative tie between firm performance and management team's compensation and promotion, government controlled firms are 12% less likely to commit fraud. Furthermore, we find that our results are mostly driven by the local political connection rather than the central political connection. Finally, our results provide information that can inform policy debates among the regulation policy makers. / Business Administration/Finance

Corporate governance of financial and non-financial firms is critical in modern corporations with diffuse stock ownership, which deals with the agency conflicts between managers and shareholders. Corporate governance has a profound impact on various corporate policy, and firm value in the end. This study examines the importance of corporate governance and its influences on various corporate policy and firm value and risks for both financial and non-financial firms. Chapter 1 investigates the association between the firm's liquidity level and liquidity mix on the one hand and CEO entrenchment on the other. CEO entrenchment may distort the firms' liquidity policy because managers and shareholders may have conflicting preferences between cash and lines of credit. Using lines of credit data from 1996 to 2008, we find five main results. First, entrenched CEOs hold more liquidity as measured by the sum of cash and lines of credit. Second, entrenched managers have a preference for cash over lines of credit because while cash gives them flexibility, lines of credit are accompanied with bank restrictions and monitoring. Third, entrenched CEOs also use more lines of credit because of the extra liquidity it provides, despite the associated bank monitoring. Fourth, entrenched CEOs in smaller and opaque firms tend to hold more liquidity. Five, entrenched CEO's preference for cash versus lines of credit is stronger for large and transparent firms, compared to small and opaque firms. These findings imply that firms should better align the interests of the entrenched managers with those of the shareholders in order to limit the excessive liquidity holding of firms when CEOs are entrenched and to thereby increase firms' profitability. Chapter 2 examines the relationship between bank holding company (BHC) performance, risk and "busy" board of directors, an overlooked dimension of corporate governance in the banking literature. Busy directors are defined as directors with three or more directorships. The sample covers the 2001-2010 period. We employ a simultaneous equation framework and estimate the models employing the three stage least square (3SLS) technique in order to account for endogeneity. Several interesting results are obtained. First, BHC performance, as measured by return on assets (ROA), Tobin's Q and earnings before interest and taxes (EBIT) over total assets is positively associated with busy directors. Second, BHC total risk (standard deviation of stock returns), market risk (market beta), idiosyncratic risk (standard errors of the CAPM model) credit risk (percentage of non-performing assets over total assets) and default risk (HigherZ-Score) are inversely related to it. Third, busy directors are not more likely to become problem directors, in the sense of failing the meeting-attendance-criterion (75% attendance). Fourth, the benefits of having busy directors in terms of performance improvement strengthened but the benefits of risk reduction declined during the recent financial crisis These findings partially alleviate concerns that when directors become too busy with multiple directorships, they shirk their responsibilities. Major implications for investors, regulators, and firm managers are drawn. Chapter 3 investigates the effect of CEO entrenchment on the loan syndication structure. Over the past decade, syndicated loans have played an increasingly important role in corporate financing. Unlike a traditional bank loan with only a single creditor, a syndicated loan involves a group of lenders: a lead arranger and a number of participant lenders. The syndication process, therefore, generates an additional dimension of agency problem between the lead arranger and the participant lenders, besides the traditional agency cost of debt between the borrowing firm and the lender (Diamond, 1984; Holmstrom and Tirole, 1997). Several results are obtained about syndicated loans made to firms with more entrenched CEOs. First, in these loans the number of participant lenders and their share in the loan are smaller; the lead arranger retains a larger loan share. Second, these loans are more closely held resulting in a higher Herfindahl index of loan concentration. Third, foreign lenders are less involved in these loans. Specifically, the number of foreign lenders and the percentage of loans held by foreign lenders are both smaller. Our findings shed light on the two types of agency problems associated with the syndicated loans, and have great implication for the firms' shareholders, creditors and regulators. / Business Administration/Finance

In the first chapter, I investigate how CEO’s risk incentive (vega) affects firm innovation. To establish causality, I exploit compensation changes instigated by the FAS 123R accounting regulation in 2005 that mandated stock option expensing at fair values. My identification tests indicate a positive and causal effect of CEOs’ vega on innovation activities. Furthermore, dampened managerial risk-taking incentive after the implementation of FAS 123R leads to a significant reduction in innovation related to firm’s core business and explorative inventions. It implies that managers diversify their innovation portfolios and decrease explorative inventions to curtail business risk when their risk-taking incentive is reduced. In the second chapter, I document that IPO underwriters implicitly collude on their price targets to support the stock post-IPO. While it is well known that underwriters are biased and have higher average price target (first moment), my evidence of implicit collusion is based on the dispersion in price target (second moment), with lower dispersion implying stronger implicit collusion. I find that, at initiation following expiry of quiet period, the dispersion in price target among underwriters of a firm is only 65% of that for non-underwriters. In 24.5% of the cases, at least two underwriters forecast the exact same price target. Such implicit collusion is also prevalent around lockup expiry. My results are robust to alternative, more direct, proxies for implicit collusion such as the proportion of underwriters that come out with exact same forecasts of price target. Refuting the alternative explanation that lower dispersion in price target among underwriters is due to common information that underwriters possess because of their involvement in the IPO, I find no such pattern in dispersion of Sales or EPS. In the last chapter, I study the security lending market. Stock lending markets are unique due to connections with stock markets: stock buyers become potential stock lenders. However, I show that equity loan supply is effectively fixed over time scales relevant to short sellers because short-term investors (less than three month holding period) do not lend shares. Transitions to stock specials are characterized by demand spikes, and slow-moving supply contributes to boom-and-bust cycles among stock specials. Consistent with my findings, I show stronger results among higher turnover stocks as well as around news events and earnings announcements. / Business Administration/Finance

The trading range of a security is the difference between the high and low recorded price over some time interval. Prior theoretical research has shown that the trading range of a security contains more information about the security's variance than does the return. These results were derived under the restrictive assumption that prices follow a geometric diffusion process. A growing body of research suggests the geometric diffusion process does not sufficiently describe security returns. A number of alternative return-generating processes have been proposed in recent years, and in many cases these alternative processes have been empirically shown to describe security returns better than the geometric diffusion process. This dissertation focuses on the statistical properties of the trading range when security returns are not generated by a geometric diffusion process. Five alternative processes are considered: A binomial random walk, an autoregressive random walk, a mixture of normal distributions, a mixture of binomial distributions, and a stable law model. For each of these processes, a closed-form solution to the distribution of the trading range is derived. Using historical data for Treasury Bond and SP500 futures contracts, goodness-of-fit techniques are employed in order to determine which models best explain observed trading ranges. Results show that the geometric diffusion process does not describe returns well, and that a mixture distribution is a more suitable candidate. The variance of a security's return-generating process is an important unobservable variable. Several researchers have devised ways to use the trading range to estimate the variance of a security when the security's returns follow a constant-variance diffusion process. In this dissertation, those results are extended to allow variance estimation under less restrictive assumptions about the underlying process. Results show that the binomial random walk model provides the most accurate forecasts of variance. Results also provide support for prior research that indicates a security's variance is not stationary through time.