The American Chamber of Commerce Research Association (ACCRA) produces the only source of publicly available regional cost of living data which, this paper suggests, may provide misleading information. An evaluation of the quality of the ACCRA indexesconcludes that they contain substantial errors and biases, predominantly from the estimated prices, although error also is introduced by the choice of index formula. To evaluate the ACCRA index, this paper uses category indexes produced by BLS researchers, Kokoski, Cardiff and Moulton (KCM 1994) to produce new regional cost-of-living indexes which substantially reduce the errors and biases found in the ACCRA indexes. 'The authors thank Mary Daly and other participants at the Federal Reserve System Committee on Regional Analysis, Mary Kokoski, Brent Moulton, W olf Weber, Kim Zieschang and Mark Wynne for valuable insights. Jeff Osborne and John Benedetto provided valuable research assistance. This paper does not represent the official views o f the Federal Reserve Bank o f Dallas or the Federal Reserve System. 2Money, July 1996

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.. American Statistical Association is collaborating with JSTOR to digitize, preserve and extend access to Journal of Business &Economic Statistics. Accurate measures of regional cost of living are vital to businesses and individuals. We compare a commonly used regional cost-of-living index, produced by the American Chamber of Commerce Research Association (ACCRA), to an index we calculate using Consumer Price Index data and research from Kokoski, Cardiff, and Moulton. We find significant differences between the ACCRA and the new indexes that are likely due to theoretical design, data collection, and sampling design. The comparison of these indexes highlights sources of differences in regional cost-of-living measures and suggests caution in the use of ACCRA indexes.

This paper reviews the literature on measurement error in the major US price indexes -the Consumer Price Index (CPI), the Producer Price Index (RPI), and the Gross Domestic Product (GDP) deflators. We take as our point of departure Triplett's, 1975. survey and focus on the studies of measurement error that have appeared since then. We review the problems of substitution bias, quality bias, new goods bias, and outlet substitution bias that are generally considered to be the main sources of error in price indexes. The bulk of the paper is devoted to problems in the CPI and PPI, as the GDP deflators tend to be based mainly on the components of these series. We find that there has been surprisingly little work on the problem of overall measurement error in any of these price indexes, and we conclude that there is very little scientific basis for the commonly accepted notion that measured inflation at 2 to 3 percent a year is consistent with price stability.