Modem society requires reliable and safe operation of its infrastructure. Policymakers believe that, in many industries, competitive markets and regulatory incentives will result in system performance superior to that under command-and-control regulation. Analytical techniques to evaluate the reliability and safety of complex engineering systems, however, do not explicitly account for responses to market and regulatory incentives. In addition, determining which combination of market and regulatory incentives to use is difficult because policy analysts' understanding of complex systems often depends on uncertain data and limited models that reflect incomplete knowledge. This thesis confronts the problem of evaluating the reliability of a complex engineering system that responds to the behavior of decentralized economic agents. Using the example of restructured and partially deregulated electric power systems, it argues that existing engineering-based reliability tools are insufficient to evaluate the reliability of restructured power systems. This research finds that electricity spot markets are not perfectly reliable, that is, they do not always result in sufficient supply to meet demand. General conclusions regarding the reliability of restructured power systems that some economic analysts suggest should be the basis of reliability policies are either verified or demonstrated to be true only when applied to extremely simple and unrealistic models. New generation unit and transmission component availability models are proposed that incorporate dependent failure modes and capture the behavior of economic agents, neither of which is considered with current adequacy techniques.