I investigate: (i) Agency problems between debt and equity holders, and their impact on capital structure and investment policy; (ii) Agency problems between firm managers and capital providers. The first chapter, "Investment and Financing under Reverse Asset Substitution", shows that banks place investment and borrowing restrictions on firms that are in lending relationships even when firms face no risk of bankruptcy, in order to continue extracting surplus from the firms over multiple periods. This agency problem is especially severe for firms that suffer from larger information asymmetries with the credit market. I use the term Reverse Asset Substitution (RAS) to express this partial transfer of control that benefits banks at the expense of equity holders of the firm. Using six different approaches, including triple difference in difference, Instrumental Variables, Propensity Score Matching and Endogenous Self-Selection, I provide empirical evidence consistent with the existence of RAS. I find that firms enjoying perfect competition in credit supply invest 2% more in PP&E than firms facing a monopoly in credit supply by banks. RAS reduces firmgrowth (11% lower PP&E) and leverage (24% lower). In the chapter titled "Heterogeneity in Corporate Governance: Theory and Evidence", I propose that the amount of management autonomy in a firm is chosen as a best response to exogenous firmcharacteristics, such as output variance. Shareholders in firms with higher exogenous variance attempt to reduce the information disadvantage they face by reducing autonomy of management. In addition, I find that over time, this information gap has decreased in US capital markets, and since Sarbanes-Oxley the information asymmetry does not play a role in the choice of corporate governance mechanisms. In the last chapter, "Effort, Risk and Walkaway under High Water Mark Style Contracts", Sugata Ray begin_of_the_skype_highlighting end_of_the_skype_highlighting and I model a hedge fund style compensation contract in which management fees, incentive fees and a high water mark (HWM) provision drive a fund manager’s effort and risk choices aswell as walkaway decisions by both the fund manager and the investor.Welfare results for the calibrated model show that a higher management fee and lower incentive fee (e.g. a 2.5/10 contract) lead to Pareto improvement.