Cost-of-living, price indexes.

At What Price? Conceptualizing and Measuring Cost-of-Living and Price Indexes. By the National Research Council. Washington, DC, National Academy Press, 2002, 332 pp., $49.95/ hardcover.

At What Price? Conceptualizing and Measuring Cost-of-Living and Price Indexes is the product of an 11-member panel convened by the Committee of National Statistics (CNS) and sponsored by the Bureau of Labor Statistics (BLS). BLS requested the panel to analyze the development of a cost-of-living index (COLI) and evaluate the proper use of consumer price indexes for Federal programs such as Social Security, food stamps, and Federal Government wages.

Discussion of the COLI dates back to 1961 when the Stigler Committee of The National Bureau of Economic Research summarized the differences between a consumer price index (CPI) based on a cost-of-goods index (COGI) and an index that measures the cost of living. The Stigler Committee recommended that BLS conduct long-term research to improve the CPI by transforming it into a better approximation of a COLI. In 1995, the Senate Finance Committee appointed the Boskin Commission to evaluate biases in the CPI. The Senate reasoned that upward bias caused overcompensation to Social Security recipients. The Boskin Commission determined that the CPI overstates inflation by 1.1 percentage points per year and recommended that BLS change the CPI methodology from a COGI, or fixed market basket framework, to a COLI.

The CNS panel promotes the COLI, similar to the previous commissioned reports, but it diverges from the Boskin Commission conclusions by highlighting the relevance of the COGI. Katharine Abraham, former Commissioner of the Bureau of Labor Statistics, supported the production of both indexes. "An index that is good for one purpose will not always be good for another ... Each purpose leads to a somewhat different conceptual framework." The panel concurs with the former commissioner by recognizing the importance of the COGI as an indicator of the level of prices, and the COLI as a measure of the change in the cost of living.

Specifically, the panel defines COLI as a measurement of "the percentage change in expenditures a household would have to make in order to hold constant some specified standard of living." According to economic theory, when prices change, consumers generally shift their purchases toward goods with relatively lower prices. For example, if the price of beef increases relative to chicken, consumers will tend to purchase more chicken relative to beef. Therefore, an advantage of the COLI compared to the COGI is how it accounts for substitution between items, while maintaining an equivalent standard of living between two time periods. The COGI is markedly different from the COLI because it does not account for substitution that may occur between items.

At What Price? contains 18 recommendations from the panel. These range from the development of a conditional COLI to conducting research on issues like quality change and data collection. The panel supports a conditional COLI where private goods and services are accounted for, but environmental factors are held constant. Accounting for nonmarket prices presents numerous conceptual problems such as measuring price in changes to the environment, quality of life, and public goods. The panel uses temperature as an example. When it is extremely hot or cold, people tend to spend more money on heat or air conditioning. If the price for heat and air conditioning remains constant throughout the shift in temperatures, then the price index should not move regardless of the change in consumption.

The panel recognizes that BLS currently evaluates what percent of a price change is caused by quality and what percent is caused by `real' price change. BLS is currently investigating a hedonically adjusted price change, where statistical regressions are applied to monetary values based on changes in product characteristics. The panel is cautious about completely integrating hedonics into the entire CPI market basket and recommends further research.

Another area of research recommended by the panel is the exploration of new methods of data collection. Currently, the Consumer Expenditure Survey accounts for the level of expenditures across items. Given the high degree of aggregation, BLS is unable to measure living costs for specific commodities or demographic groups. One means of disaggregating, or collecting household level expenditures, is with handheld computers and scanners. In addition to new survey techniques, the panel recommends researching the feasibility of integrating expenditure weights from the personal consumption expenditure (PGE) survey prepared by the Bureau of Economic Analysis.

BLS initiated the production of the Chained Consumer Price Index (C-CPIU) in August of 2002; this new index is based on a COLI framework. Successful implementation, however, may depend on understanding the methodology behind the COLI and COGI, and specific uses for each index. At What Price? serves as a good resource for business analysts and economists to social science researchers and policymakers.

Joshua Klick
Division of Consumer Prices
and Price Indexes,
Bureau of Labor Statistics