With the discovery of scanner data by statistical agencies and researchers comes a wealth of new information upon which price index calculations can be based. Old problems, such as the appearance and disappearance of goods over time, are likely to be an important feature of such data. However, given that scanner data includes the prices and quantities of the population of transactions we have more information than is traditionally available to deal with the new and disappearing goods problem. We adopt a recently developed approach using the Constant Elasticity of Substitution cost function to provide a detailed empirical analysis of the effects of new and disappearing goods for an Australian scanner data set of supermarket products. Our results indicate that the failure to account for new and disappearing goods in the cost-of-living index leads to a significant upward bias.