The use of energy cost as an indicator of natural resource quality is examined. Energy cost is the direct and indirect energy used to produce a unit of a resource commodity. Different approaches to the role of energy in the economy are discussed and it is demonstrated theoretically that energy cost can give false indications of changes in resource quality. This is because energy cost is only an accurate indicator of resource quality if either energy is the single primary factor of production and the economy can be represented by an input-output model, or if there is substitutability between the factors of production, then marginal products must be proportional to the factors' embodied energy. These conditions are tested empirically using data from the US agricultural sector. The results show that there is substitutability and marginal products are not proportional to embodied energy. However, the strong neoclassical assumption of price-taking profit-maximization can be rejected at a 5% level of significance, though not at the 3.5% level. The generality of the empirical results may be restricted due to data limitations but the theoretical framework can be applied to other data sets, other resources, and even other indicators.