| In this dissertation, the consequences of incomplete information in principal-agent models with moral hazard were explored. Moral hazard requires imposing risk; however, when the returns to effort are uncertain, inefficiencies arise. Three types of incomplete information are examined: a principal who has private information about her technology; a principal who has perfect, imperfect, or null information about her technology when she may have secret, private, or symmetric information; and, an agent who undertakes effort and then must select a project in which to invest. It is shown that, under fairly general conditions, neither the principal nor the agent would voluntarily acquire information if the acquisition would be common knowledge, unless contracting occurs prior to information acquisition.;In the first essay, the contract she offers signals this information to the agent. Perfect Bayesian Equilibria of the game possess the following properties that don't arise with complete information: a principal with a more informative technology earns less profit that an principal with a less informative technology; the actions implemented can be distorted; and, the agent can be better off when the principal has private information.;In the second essay, it is shown that: with ex post contracting, when the principal has private information, the value of information for the principal may be negative; although the agent prefers that the principal has private information when there is ex post contracting, both ex post contracting when information is symmetric and ex ante contracting when the principal has private information are superior to ex post contracting when the principal has private information using the Potential Pareto Criterion; and, in contrast to Maskin and Tirole (1992), the principal may actually earn more profit when she has private information than when information is symmetric.;The third essay extends the analysis of Lambert (1986), who showed that the optimal contract may have to trade off underinvestment or overinvestment in order to encourage the agent to exert effort. It is shown that a two-period contract reduces both the underinvestment and overinvestment problem in the first period, but may reduce or exacerbate the problem in the second period. |
