A dynamic programming framework for the analysis of firm decisions

Abstract

In this chapter, I present a dynamic programming model designed for the empirical analysis of firm behavior under various non-standard constraints. The main purpose of the model is not to serve as a tool for the theoretical analysis of firms’ investment decisions, but to give empirical studies a solid foundation in recent theoretical advances in this field. Hence, the model is quite stylized in some respects (such as its focus on discrete decision variables). Its general structure, though, allows it to be readily adopted in empirical studies of firm behavior, using advanced structural estimation procedures that are firmly based on the underlying optimization model. The analysis starts from the observation that a firm’s market entry and market exit are endogenous events in a general model of firm behavior. In the model presented here, I do not analyze the determinants of entry and exit in an explicit market-interaction framework, but emphasize the endogeneity of the firm’s market exit decision in a model of investment behavior. In such a framework, market interactions are reflected implicitly in the firm’s exit rule; they enter via the effects of equilibrium output prices on the firm’s relevant state variables.