Foot-Dragging on Price-Index Repair
Byline: Alfred Tella, SPECIAL TO THE WASHINGTON TIMES
One reason the government spends taxpayers' money to improve economic statistics is to minimize data inaccuracies that can balloon the federal budget deficit. But once the data are improved, political paralysis can set in and the budget savings are not realized.
Example: the Consumer Price Index (CPI), estimated and reported monthly by the Bureau of Labor Statistics (BLS).
By way of background, in 1995 the Senate Finance Committee appointed a temporary panel of experts to recommend ways to improve the accuracy of the CPI, criticized by many economists and data users as upward-biased in measuring the cost of living. The BLS already had a program to improve the price data, and the panel's recommendations gave it further impetus.
Since then, the BLS has made a number of improvements to the data, which have reduced the overstatement of price increases. One of the most significant improvements resulted in creation of an alternative price index, called the Chained Consumer Price Index.
The chained index (C-CPI) corrects for what economists call substitution bias in the traditional CPI. The CPI tracks a fixed market basket of goods and services that does not vary in the short run, thereby failing to take into account month-to-month changes in consumer choices when prices change.
When the price of steak rises relative to the price of chicken, for example, consumers eat more chicken and less steak. A sharp drop in computer prices may lead more consumers to buy a computer rather than a television set or something else. For the C-CPI series, the BLS uses improved methodologies to capture continuing shifts in purchasing patterns.
Although the chained index is subject to slight periodic revision, up-to-date data have been reported monthly for all urban consumers since August 2002 and have been judged reliable and useful. The BLS said the chained index "was designed to be a closer approximation to a 'cost-of-living' index" than its traditional series. Reflecting its confidence in the data, starting in February 2004 the BLS moved up its reporting of chained prices to the opening section of its monthly press release.
In its latest monthly release, the BLS reported the CPI for all urban consumers in December 2004 was 3.3 percent higher than a year earlier, whereas the chained index for all urban consumers was up 2.9 percent. In the past four years, the annual increase in the chained index averaged 0.4 percentage point less than the traditional CPI, a meaningful difference. During periods of above-average price increases, the gap tends to widen as consumers shift from buying items that are becoming more expensive.
The BLS also publishes a monthly price index for urban wage earners and clerical workers (CPI-W), which is used to make annual cost-of-living adjustments to Social Security and other programs. …