CPI Might Spell Less Inflation
Subject Inflation
Topic Employment, Unemployment, and Inflation
Key Words Inflation, Monetary Policy
News Story

The Consumer Price Index (CPI) is used to index social security checks, veterans' benefits and federal pensions, thereby directly affecting about 80 million people in the United States. Since the Internal Revenue Service (IRS) adjusts tax brackets by the CPI, and the Federal Reserve carefully considers inflation measures when setting monetary policy, it is probably not an exaggeration to say that the CPI affects most Americans. Two government commissions and many economists have argued that the CPI is flawed and overstates inflation. As a result of the commission findings, a number of changes have been introduced in the methodology for constructing the CPI, and additional measures will be introduced next month. These changes have the potential of lowering the measure of inflation and saving the government billions of dollars.

In 1996, the Boskin Commission reported to the Senate that the CPI overstates inflation by about 1.1 percent per year. It criticized the index for not recording how consumers were saving money in the face of price increases, or how technology was improving consumer life. The federal government, it concluded, was spending billions of dollars unnecessarily because of this problem.

Next month, the Bureau of Labor Statistics will introduce a number of important changes to increase the accuracy of its principal measure of inflation. These changes are in two areas: revisions to make the index better reflect changing trends in the marketplace; and, methodological changes to better represent consumer behavior. Currently, the CPI bases the relative importance of each of the 211 goods and services that it surveys every month on data collected in 1993 to 1995. The BLS will revise the database to 1999-2000 and, thereafter, make revisions every two years, eliminating the current 5 to 10 year lag. When consumers are faced with changes in relative prices for goods that serve a similar function, they typically substitute the lower-priced good for the relatively more expensive one. This behavior was basically not reflected in BLS procedures. Starting in 2003, the BLS will make adjustments for this, but it will be reported separately.

(Updated January 15, 2002)

1. According to your text, what are the sources of bias in the measurement of inflation by the CPI?

How will using an updated database affect each of these biases? Explain.


Distinguishing quality improvements from inflationary price changes has been recognized as a problem for many years. If, for example, a 2002 Chevy Blazer cost 10 percent more than a 2001 year model, but had a better stereo system and improved safety features, would you consider all of the 10 percent increase as inflation? How would you determine what portion of the increase was inflation?

Source Jolie Solomon, "An Economic Speedometer Gets An Overhaul," The New York Times, December 23, 2001.

Return to the Employment, Unemployment, and Inflation Index

©1998-2002  South-Western.  All Rights Reserved   webmaster  |  DISCLAIMER