By Jim Cline and Kate Kremer
In the last blog article we addressed the recent .5% gap between the “U” and “W” indices. We discussed whether over time there was a significant difference between the CPI-W and the CPI-U, explaining that there was not:
In fact, if you trace the historic numbers all the way back to 1960, even in periods of inflation much higher than current, you will find that there have been few occasions in which the these two measures CPI deviated more than .2%. Looking at the Seattle index [Seattle Annual CPI-W & U Difference , Seattle Annual CPI-W & U Table], for example, only in 19 of the past 55 years has the gap exceeded .2%. And in only 3 of those 55 years has the gap been as wide as or wider than reported in this June’s index. (The variability of the All-Cities numbers [All Cities Annual CPI-W & U Difference, All Cities Annual CPI-W & U Table] are similar.)
We are often asked about what the difference is in what these CPI reports measure. We usually offer the brief explanation that the W index is the “Wage earners” index and is usually the one more labor contracts are tied to.
The City of Seattle Finance Department website offers a fairly good brief explanation of these measures, as well as a discussion about the differences between the Seattle and All-Cities CPI. As they accurately explain:
In addition to the national CPI, the BLS publishes CPI statistics for 26 of the nation’s metropolitan areas. However, it does not compute the CPI for states. For the Seattle area, the BLS computes both the CPI-U and the CPI-W for the Seattle-Tacoma-Bremerton, Washington, metropolitan area, which includes Island, King, Kitsap, Pierce, Snohomish, and Thurston Counties. In 2000, this area was home to 3.6 million people, which is 60 percent of the state’s total population.
A more complete explanation of the differences between U and W are provided by BLS. In an article labelled “Why does BLS provide both the CPI-W and CPI-U?” the BLS economists Stephen Reed and Kenneth Stewart explain the history and methodology involved in these two measures.
They explain that the “U” index is actually the broader index, measuring all consumers. The W index more narrowly focuses on wage earning occupations:
The Consumer Price Index for Urban Wage Earners and Clerical Workers is based on the expenditures of households in which more than one-half of the household’s income comes from clerical or wage occupations, and at least one of the household’s earners has been employed for at least 37 weeks during the previous 12 months. The CPI-W population represents about 28 percent of the total U.S. population.
They explain that an effort occurred in the 1970s to drop the W measure entirely but that was abandoned after resistance from labor unions that preferred the use of that measure in their labor contracts.
Reed and Stewart explain that there are no separate surveys for the two reports. BLS uses the same sample and then adjusts the “weights” in formula to reflect the difference basket of goods captured between the indices.
So it should not be a surprise that there deviation is slight. This year’s difference more likely represents a temporary situation, not a longer term trend. If unions may have been using the W index, there’s no compelling reason that we see to switch over on a long term basis to the U index. As always in negotiations, we expect that both parties may point to various measures of inflation to support their position on wages. But we don’t see why that compels a long term tactical shift in how you approach CPI. The bigger issues, as we’ve noted, are whether you stick with the Seattle index over the All Cities index as well as the question of whether you forego the use of the CPI in your contract entirely.
**Please visit our Premium Website for more information regarding current and past CPI reports.
