Essays in economics: 1. Pre-committed government spending and partisan politics. 2. Investment in energy efficiency: Do the characteristics of firms matter? 3. Information processing and organizational structure

1. Spending commitments requiring future outlays are important for understanding partisan politics because they prevent a conservative government from scaling back spending programs. In a one-government-good model, a "stubborn liberal" policy maker can use precommitted spending to prevent a later conservative government from imposing spending cuts. In a model where parties differ about spending priorities, re-election uncertainty creates a bias towards higher government spending and higher taxes.;3. Standard theories of the firm stress profit maximization as the foundation for derivation of predictable behavior. Yet evidence continues to accumulate that firms do not act as required by the neoclassical framework. Instead of being represented by ever more elaborate maximization models, the firm can be modeled simply as a network of information-processing agents. The actions of the firm are then a function only of the network structure and the information-processing capabilities of the agents. This approach can be used to explain a number of features of organizational behavior. It also suggests that derivation of the optimal organizational structure may be computationally complex, with a number of implications for economic theory and policy development.;2. The literature on energy efficiency provides examples of profitable technologies that are not universally adopted. Theory indicates that firms should undertake all investments with a positive net present value, and that the discount rate for computing the present value of a project should be the return available on other projects in the same risk class, not on characteristics of the firm. This model is tested by examining whether firms' characteristics influence their decision to join the Environmental Protection Agency's Green Lights program. A discrete choice regression is estimated over a sample of participating and non-participating firms. Missing values in the data matrix are replaced with multiple imputations using the EM algorithm. The results show that: (1) substantial improvements in the power of hypothesis tests can be achieved through imputation of missing data, and (2) characteristics of firms do affect their decision to join Green Lights.