Survey Methodologies and the Expenditure Basket Foundation

The U.S. Bureau of Labor Statistics (BLS) CPI-U tracks average price changes for goods and services purchased by urban consumers. BLS uses Consumer Expenditure Surveys to define expenditure-weighted categories. The resulting aggregate market basket organizes hundreds of item categories into expenditure-weighted components encompassing shelter, energy commodities, food and beverages, transportation services, and medical care, establishing a macroeconomic yardstick for purchasing power stability.

Substitution Bias and U.S. CPI Aggregation

A fixed-weight Laspeyres index holds the base-period consumption pattern constant and may not fully capture consumers’ responses to relative-price changes. The U.S. CPI-U is not a single, pure Laspeyres calculation: BLS uses geometric means for some lower-level item groups and updates expenditure weights periodically; it also publishes a separate chained CPI-U (C-CPI-U). How substitution is reflected depends on the index and item-level formula.

Hedonic adjustment uses estimated quality differences

BLS's quality-adjustment explanation uses regression estimates relating product characteristics to prices when comparing different models. This is not applied to every item and does not perfectly isolate “pure inflation.” Twice the performance is not automatically assigned twice the value.

In BLS's methodological shirt illustration, a $20 short-sleeve blend shirt is replaced with a $30 long-sleeve all-cotton shirt. The assumed log-price coefficients are 0.15 and 0.25. The comparable old price becomes 20×exp(0.15+0.25) = $29.8365. The unadjusted increase is 50%; the adjusted relative change is 30/29.8365−1 ≈ 0.5480%. Simply adding the coefficients and applying a 40% price adjustment is incorrect. This unpacks BLS's illustrative model arithmetic; these are neither current shirt prices nor a 0.5480% contribution to the overall CPI.

Identify the index before comparing it with your budget

BLS's calculation handbook distinguishes basic item indexes from upper-level aggregation. The use of geometric means does not make the entire CPI one simple geometric average. When comparing CPI-U and C-CPI-U, match periods and check whether the chained index is preliminary or final.

To explain a household's higher spending, record quantities, brands/specifications and purchase dates as well as prices. A quality-adjusted index and the cash needed to buy a replacement answer different questions. A higher replacement bill does not contradict an adjusted index. The methods discussed here concern the U.S. CPI and should not be transferred unchanged to Korean or Japanese indexes.