Measurement of Quality Changes in Consumer Durables

This exhibits some of the tentativeness of a report on research in progress.1 Nevertheless it develops notions of a theoretical character which, together with an empirical survey, may have broad interest. It reaches three central conclusions: 1. If one distinguishes between consumer goods as sources of services and the service streams proper, intertemporal comparisons of costs of living become more feasible than has heretofore been suggested. 2. Forms of industrial organization have much bearing on the appropriateness and effectiveness of procedures that might be undertaken to measure quality changes of sources of services. Specifically, we are concerned with the perfection of second-hand markets, concentration ratios, and the degree to which there are important indivisibilities in the production process. Supply considerations cannot be neglected in cost-of-living studies. Ideally-organized second-hand markets make possible precisely-calculated ‘improvement factors’, measures of quality change. 3. Empirical investigation of the retail market for household refrigeration in the United States indicates that the failure of the US Bureau of Labor Statistics to account for changes in quality (including size) may have imparted a downward bias to indexes computed up to 1956.