This thesis presents three essays on the determinants and costs of corporate security offerings. The essays contribute to an ongoing debate in the literature on what determines firms’ security choice by examining the following issues: “Does corporate governance influence convertible debt issuance?”; “The signaling content of security offerings proceeds”; and “The costs of raising capital: New evidence.”In the first essay, we explore the influence of corporate governance on firms’ choice between equity, convertible debt and straight debt. For a sample of Western European corporate security offerings between 1999 and 2010, we find that firms with weaker firm- and country-specific corporate governance are more likely to issue convertible debt. They thus use convertible debt as a substitute for corporate governance, which is confirmed by a more favorable stock price reaction to convertible debt announcements by firms with weaker corporate governance. Moreover, these results suggest that corporate governance is a significant determinant of firms’ security choice. The second essay examines the determinants and signaling content of security offering proceeds, controlling for the endogeneity of issue size. For a sample of US equity, convertible debt and straight debt offerings between 1999 and 2011, the findings show that stockholders can partly predict issue size by analyzing firms’ funding needs and financing costs. We find that stockholders use predicted issue sizes of equity and convertible offerings as signals of growth opportunities, whilst larger than predicted issue sizes signal issuer overvaluation. For straight debt issues, we find that unpredicted issue sizes have a positive impact on announcement returns, which is consistent with them serving as a signal of growth opportunities. Further analysis of firms’ actual uses of predicted and unpredicted offering proceeds confirms these interpretations. The results shed light on previous inconsistent findings on the impact of issue size on security offering announcement returns. The final essay examines the magnitude and determinants of direct issuance costs, controlling for firms self-selecting into different security classes, namely equity, convertible bonds, and straight bonds, and flotation methods, namely non-shelf, shelf and 144a. For a recent sample of US corporate security offerings between 1999 and 2011, findings show that the magnitude of direct issuance costs has decreased over the last decade. These costs are higher for equity than straight bond offerings and of intermediate magnitude for convertible bond offerings. Within each security class, costs are larger for non-shelf than 144a offerings, which again have larger direct issuance costs than shelf offerings. Finally, underwriter spreads are directly related to underwriter effort on due diligence, pricing and selling, and direct issuance costs are truncated by firms’ self-selection into particular security types.

The objects this thesis want to study can be separated into two parts. The first part is investigating why firms choose to issue convertible bond. We use the public financial information and macroeconomics factors to establish a security choice model. In this security choice model, we can understand the motivation of issuance and investors¡¦ expectation of security type (equity-like or debt-like) which will be issued. The second part of this thesis is about convertible bond issuance announcement effect. We want to know if the public information and the pre-issuance security type expectation would affect the announcement effect. Following is the conclusions: (1) We find that less information asymmetry, less agency cost, more operating risk will lead to higher probability of equity-like security issuance. (2) Most convertible bonds issued in Taiwan are more debt-like. (3) Equity-like convertible bonds issuances have negative announcement effect. The issuances different from expectation will lead to more negative price effect, especially those debt-like firm but issue equity-like security. (4) The variables that can explain security choice may not explain the announcement price effect.