This paper develops the first systematic attempt to model and empirically estimate the concept of optimal resource renting. Optimal rent is found to be positively affected by increases in the recession buffer and resource endowment, and negatively affected by the opportunity cost of hoarding. The model is then tested empirically on Norway, an oil-rich state, and actual renting is found to be systematically diverging from the optimal rent series. At least a third of the variation in actual renting is always left unexplained by the economic variables of the model, and should be attributed to the institutional and political factors that lie beyond the scope of our analysis.