Learning and Smooth Stopping

We propose a simple model of optimal stopping where the economic environment changes as a result of learning. A primary application of our framework is an optimal job search problem when the worker's labour market opportunities are initially uncertain. We distinguish between two interpretations of the model. In the first, a worker learns about common market conditions, such as the number of potential employers, that affect all searchers. In the second, the worker learns about her idiosyncratic productivity distribution across firms. For the first model, we show that learning leads to higher wage demands by the workers. In the second model, we give sufficient conditions so that learning leads to higher wage demands for optimistic workers and lower demands for pessimistic workers due to learning.