This paper considers the monetary policymaker’s joint problem of model estimation and the design of a policy rule in the face of uncertainty regarding the process of structural change in the economy. Unobserved structural change is modeled through time variation in the natural rates of interest and unemployment. I show that certainty-equivalent optimal policies perform poorly when there is uncertainty about the natural rate processes. I then examine the properties of combined estimation methods and policy rules that are robust to this type of model uncertainty. I find that weighted sample means are robust estimators of natural rates for the purpose of setting policy. The optimal policy under uncertainty incorporates a significant degree of policy inertia and a muted response to the perceived unemployment gap; by comparison, the certainty-equivalent optimized policy in this model exhibits little policy inertia and a more aggressive response to the unemployment gap.