Recent research on investment decisions under uncertainty has highlighted the importance of option values, or of waiting to resolve some uncertainty and gather more information. This work is mostly restricted to "classical" economic environments, with perfect risk markets and without economies of scale or externalities. This project extends the analysis to "non- classical" settings, using corresponding extensions of techniques of stochastic optimal control. The more general theory will then be applied to some important policy issues. Examples include the following. (1) Criteria for investment under uncertainty with scale economies and externalities will be developed, built into a model of industry evolution, and linked to the new theory of growth. (2) Since labor income risk is hard or impossible to diversify, trade adjustment, and more generally job losses due to plant closure, alter risk- bearing. The issue, and appropriate policies, will be modelled. (3) Real rigidities of slow adjustment will be combined with nominal inertia into macroeconomic models of employment fluctuations. (4) When evaluating new technologies to reduce environmental damage, the value of waiting to gather more information must be traded off against the risk of irreversible damage while we wait. This will be formulated as a stochastic control problem with filtering.
