Summary: Modeling consumer behavior is a relevant and growing research area in revenue management. Single-resource (single-leg) capacity control problems comprising consumer choice modeling constitute the backbone of more complicated models. In existing models, the distribution of demand is assumed to be independent of external factors. However, in reality demand may depend on the current external environment which represents the prevailing economic, financial or other factors that affect customer behavior. We formulate a stochastic dynamic program that comprises a discrete choice model of consumer behavior in a randomly fluctuating demand environment with a Markovian structure. We derive some structural results on the optimal policy for capacity control. The model and the results generalize earlier work of K. Talluri and G. van Ryzin [Manage. Sci. 50, No. 1, 15–33 (2004; Zbl 1168.91427)]. In particular, the concept of an efficient set of products plays an important role but such sets may depend on the particular external environment. We also present some computational results which illustrate the structural properties and explore the benefits of explicitly modeling the external environment.