Press Release
For further information:
Ed Fiedler (919) 933-5540
Gail Fosler (212) 339-0300
For Release 3:00 p.m. ET, Thursday, December 16, 1999 Release #0001A
CONFERENCE BOARD STUDY GROUP RECOMMENDS IMPROVEMENTS TO THE CONSUMER PRICE INDEX
December 16 -- The Conference Board's Study Group on the Consumer Price Index today recommended substantial improvements in methods used to compile the index to keep pace with the ever changing U.S. economy. In particular, the Group emphasized a need for the timely introduction of new products and more comprehensive adjustment of changes in the quality of products and services. These two improvements represent "the house-to-house combat of price measurement," the report noted, "and are probably the most difficult, and surely the most costly of the recommendations."
TWO YEAR REVIEW
The Study Group's report is the culmination of a two year review of measurement issues relating to the CPI. The Group is chaired by Nobel Prize Winner James Tobin of Yale University and Paul W. McCracken of the University of Michigan, a former chairman of the President's Council of Economic Advisors.
"The accuracy of the consumer price index remains an open and important issue," says Professor Tobin. "Because it is the most commonly used benchmark for measuring inflation, the CPI is a key measure for the Federal Reserve and other policy authorities. About half a trillion dollars in Federal spending are linked to the consumer price index. Even small errors in measuring prices, when cumulated over the years, may cause substantial unintended allocations of funds.
"The issues raised in this report," Tobin added, "will become still more important in a world being reshaped by e-commerce."
The Conference Board Study Group on the Consumer Price Index
Co-Chairs:
Paul W. McCracken, University Professor Emeritus, University of Michigan (734) 764-1581
James Tobin, Sterling Professor of Economics - Emeritus, Yale University (203) 432-3720
Other Members
Charles R. Hulten, Professor of Economics, University of Maryland (301) 405-3549
Marvin Kosters, Director, Economic Policy Studies, American Enterprise Institute (202) 862-5846
Robert D. Reischauer, Senior Fellow, The Brookings Institution (202) 797-6056
Ex-Officio Members:
Edgar R. Fiedler, Senior Fellow and Economic Counsellor, The Conference Board (919) 933-5540
Gail D. Fosler, Senior Vice President and Chief Economist, The Conference Board (212) 339-0300
NEW PRODUCTS AND QUALITY BIAS
When entirely new products appear in the economy-items such as cellular telephones, skateboards, hip replacements, CD-ROM encyclopedias, and Viagra-the key requirement is to introduce these items into the CPI promptly. Otherwise the CPI will miss the price declines that often occur in the earliest stage of the product cycle.
Quality bias creeps into the CPI whenever a change in the quality of goods and services included in the CPI goes undetected or is miscalculated. "The issue arises most often," said Co-chairman, Paul McCracken, "during periods like the present when major innovations are almost rearranging the economic landscape." Finding and correcting quality bias is a formidable, trained-researcher-intensive task. There is no alternative to thorough, detailed analyses that slog through the data category by category, item by item-a massive research effort.
THE CPI IN A DYNAMIC ECONOMY - A 'WORK IN PROGRESS'
The Study Group praised the Bureau of Labor Statistics (BLS) for the high quality of its work and for the important changes it has made to the index in recent years. "The CPI is a work in progress, always being improved, never finished," says Ed Fiedler, Economic Counsellor of The Conference Board and Executive Director of the Study Group. "In recent decades, we've come to realize more fully than before how often consumers substitute one good for another when prices change, how many new products and services are introduced each year, how many changes take place in the quality of existing items, and how frequently consumers shift their spending from one type of retail outlet to another - most recently to the internet. This recognition of the highly dynamic nature of the economy is the prime stimulus for many of the improvements already made in the CPI, and for the further changes advocated by the Study Group."
"The changes undertaken by the BLS," notes Gail Fosler, The Conference Board's Chief Economist and ex-officio member of the Study Group, "have shifted the CPI away from its previous more static orientation, reducing the upward bias in the index substantially. But there is still much work - perhaps the most difficult work - to be done.
"This issue has greater urgency now, given the explosion of e-commerce," she added. A recently released Conference Board survey showed that, although the dollar amounts were small, 25 percent of all consumers had purchased something from an internet service. "With the U.S. economy becoming increasingly dynamic, it is a major challenge for the statistical system to keep up."
OTHER RECOMMENDATIONS
Other important recommendations offered by The Conference Board's Study Group in this report include use of a superlative index formula in upper level aggregation (a formula that limits substitution bias), a large increase in the sample size of the annual Consumer Expenditure Survey, and more frequent rotation of the Point of Purchase Survey.
"Once implemented," observes Co-chairman Paul McCracken, "the Group's recommendations will improve the CPI substantially. And when the CPI is improved, so are other important statistics, such as those for output and productivity. Decision-makers throughout the private and public sectors will benefit from a more reliable CPI."
RECOMMENDATIONS IN BRIEF
1. In addition to the official CPI, the BLS should publish an historically benchmarked index.
2. To limit substitution bias in the official CPI, upper level aggregation should employ an approximation to a superlative index formula, probably the Lloyd-Moulton-Shapiro-Wilcox approach. For the historically benchmarked index, a standard superlative index formula should be used.
3. The sample size of the annual Consumer Expenditure Survey should be increased - perhaps initially to at least 30,000 households.
4. Full engagement in the "house-to-house combat of price measurement" should no longer be put off. This will require a very substantial research effort, the first part of which should focus on quality change and pricing outcomes rather than inputs.
5. When outlet rotation shows price changes on the same items from the old to the new sales outlets, the BLS should estimate what portion represents a difference in the quality and convenience of the transaction versus what portion represents a "true" change in price.
6. The second part of the major research initiative should focus on the timely introduction of new products into the CPI, and include more frequent rotation of the Point of Purchase Survey.
7. Coverage of the CPI should be expanded to include the rural population.
8. The CPI should undergo a thorough review by independent outside experts roughly every five years.
9. Full implementation of these recommendations will require adequate funding, mostly from the Administration and Congress, with help also from the broader research and philanthropic communities.
Source: Measuring Prices in a Dynamic Economy: Re-Examining the CPI
The Conference Board
# # #
RSS