Short Run Recommendations3
The BLS should establish a cost of living index (COLI) as its objective in
measuring consumer prices.
The advisory commissions report begins with one overarching recommendation:
"The BLS should establish a cost of living index (COLI) as its objective in measuring
consumer prices." The BLS basically concurs with this; indeed, the BLS long has said
that it operates within a cost-of-living framework in producing the CPI. That framework
has guided, and will continue to guide, operational decisions about the construction of
the index.4 Putting
things slightly differently, if the BLS staff or other technical experts knew how to
produce a true cost-of-living index on a monthly production schedule, that would be what
we would produce. While the BLS broadly agrees with the commission that the objective of
our CPI program ought to be to produce a cost-of-living index, it is important to note
that there are many different cost-of-living indexes that could be taken as the CPIs
target. The cost-of-living index approximated by the CPI answers the following question:
"What is the minimum change in expenditure that would be required in order to leave
the average consumer unit indifferent (or as well off) between a specified reference
periods prices and a comparison periods prices?"5 Consumers well-being depends on many
aspects of life other than market goods and services, e.g., environmental quality and
amenities (such as clean air and low crime), goods provided through taxes (such as
national defense and fire protection), health status, and future consumption goals (which
depend on both current and expected future income, and savings). The BLS defines the scope
of the CPI, however, to include only market goods and services or government-provided
goods for which explicit user charges are assessed. The cost-of-living index approximated
by the CPI is, therefore, a subindex of the all-encompassing cost-of-living concept,
specifically a subindex that is conditional on the excluded factors that affect consumer
well being, such as health status and the quantity and quality of government-provided
goods and services.6
Because the cost-of-living concept does not imply a single all-purpose cost-of-living
index, the BLS will continue to need to make choices about the specific issues of formula,
coverage, and index construction. One criterion in making these choices is the need to
employ methods that are appropriate for a statistical agency. Those methods must be
rigorous and objective rather than speculative, minimizing the role of arbitrary BLS
judgments.7 We recognize, for example, that
changes in disease incidence or public safety levels affect consumer well-being in ways
that are difficult or even impossible to quantify using methods that meet these standards.
We also believe, however, that adhering to the standards is critical to maintaining the
credibility of our official statistics. An important part of the BLS responsibility,
therefore, is to describe for data users the scope and theoretical assumptions of its
price measures, as well as any necessary caveats with respect to their use.
In summary, the BLS has no fundamental disagreement with the commission about what the
objective of the CPI should be, though we would emphasize that it is important to be clear
about what precisely one means by a cost-of-living index. We will continue to exercise our
judgment on the appropriateness of taking any specific step designed to bring the CPI
closer to a true cost-of-living index, bearing in mind that it is crucial that any
official statistic be based on rigorous, objective methods.
The BLS should develop and publish two indexes: one published monthly and one
published and updated annually and revised historically.
The timely, monthly index should continue to be called the CPI and should move
toward a COLI concept by adopting a "superlative" index formula to account for
changing market baskets, abandoning the pretense of sustaining the fixed-weight Laspeyres
The new annual COL index would use a compatible "superlative-index"
formula and reflect subsequent data, updated weights, and the introduction of new goods
(with their history extended backward).
Because these three recommendations address methods for dealing with the upper-level
substitution bias problem, we will discuss them together.
Some index formulas, known as "superlative" indexes, have been shown
theoretically to be closer to the cost-of-living concept than measures like the CPI that
track the cost of a fixed basket.8 The major
superlative indexes are the Fisher and Törnqvist measures. Using these formulas, one can
construct an index that accounts for the changes that consumers make in the quantities of
the goods and services they consume in response to changes in relative prices. By
substituting goods that have become relatively cheaper for those that have become
relatively more expensive, consumers can achieve the same standard of well-being for less
than the cost of purchasing their original market basket. Because the CPI holds the market
basket fixed at base period quantities, it incurs substitution bias by putting too much
weight on the relatively more expensive items from which consumers have shifted away. The
superlative indexes, because they adjust for changes in consumer expenditures, tend to
avoid this type of bias. The superlative indexes do, however, require information on the
current period market basket. Because it takes time to collect and process consumer
expenditure data, a superlative index can be produced only with a time lag.9
The commission recommends that the BLS should move to a "trailing Törnqvist"
formula for the monthly index.10 By this they
mean a geometric mean formula in which the weights are lagged expenditure shares and are
updated each year. The resulting measure would not be a true superlative index, although
it would be intended to provide a feasible real-time approximation to a Fisher Ideal or
Törnqvist index. The commission also recommends that the BLS develop a new annual index
that is calculated using a true superlative formula and is subject to revision.
The BLS continues to investigate several experimental indexes that are designed to
address consumer substitution across item categories. These include superlative formulas
which, due to the need for current expenditure data, create indexes that must be produced
with a lag, as well as new methods that may approximate the superlative formula and allow
the production of indexes in a more timely fashion.
The timeliness of the CPI might be maintained by using some form of an approximation to
a superlative index. The commissions proposed "trailing Tornqvist"
formula, however, has been shown to produce price changes that systematically understate
the increases in the cost of living, as measured by the superlative formulas.11 More recently, other approximation strategies
have been proposed, including a method based on the "constant elasticity of
substitution" (CES) formula.12 But
such an approximation would not track the superlative indexes preciselyduring some
years an index based on an approximation would rise more than the superlative index,
during other years it would rise less. This feature raises the issue of whether such an
index subsequently would need to be revised once the data were available to calculate the
superlative index. Another issue that needs to be addressed in considering use of
approximations is the issue of estimating the subaggregate indexes, i.e., the indexes for
intermediate levels of aggregation, such as for "food" or
"transportation." Some of these indexes may consist of item categories that are
relatively close substitutesfresh fruits, for example, consists of apples, bananas,
oranges, and other fruitswhereas others may consist of item categories that probably
are not close substitutesmedical professional services, for example, includes
physicians, dentists, and eye care. Because the CES function is based on a single
elasticity parameter which is assumed to be the same for all items, while consumers
willingness to substitute is likely to vary across categories of items, further research
is needed to determine whether a simple approximation such as the CES would produce
sensible approximations for all of these subaggregates. Also, the use of an index based on
statistical approximation might be difficult to interpret and explain to users of the
data. We believe we would gain little, and possibly do much damage to the credibility of
our statistical system, if we were to move hastily to adopt untested techniques for
producing the official CPI.
The BLS has requested funds to improve the accuracy and timeliness of the CPI, and an
important part of this request will support the production of a superlative index,
produced to a greater degree of accuracy than is now possible. The BLS plans to begin
publishing this measure in 2002. Selection of a specific form for the index requires that
several design issues be resolved. Most fundamental will be the choice between production
of an annual index with a lag and production of a monthly or quarterly index that is
subject to revision as new annual expenditure data became available. The BLS also will
have to address more technical issues such as functional form, estimation of expenditure
weights, and whether substitution across areas will be treated in the same way as
substitution across item categories. In the interim, the superlative measures we currently
produce can be used to estimate the magnitude of the upper level substitution bias in the
CPI, and indeed are the best measures currently available for this purpose.
Another key aspect of these commission recommendations is the annual updating of
expenditure weights. Historically, the CPI expenditure weights have been updated at
roughly ten-year intervals, corresponding to the availability of decennial Census
population data and (prior to the 1980s) of Consumer Expenditure Survey (CEX) data. Most
recently, in January 1998 the CPI was updated using the 1990 Census and 1993-95 spending
patterns. The BLS has indicated, however, that future updates will take place more
frequently. A decision concerning the appropriate interval between weight updates will be
made soon, based on additional study both of the importance of employing current spending
weights and of the "chain drift" distortions that potentially could result from
too-frequent weight changes. In this regard, the BLS has requested funds for an enhanced
CEX sample size and for computer system enhancements that will permit the introduction of
more current spending data whenever the weights are updated.
The BLS also is evaluating the production of a historical research series that would be
revised as new methodological improvements are developed and implemented in the official
index. Such an index could be of value to research economists and others concerned with
past movements in prices, income, productivity, and other economic variables.
In summary, the BLS does not accept the specifics of these recommendations but has
initiatives underway to address upper-level substitution bias and the currency of the CPI
market basket. Further research is needed to determine whether the experimental techniques
under investigation for producing the official monthly CPI would yield a more accurate
measure. We believe that hasty implementation of such techniques without more complete
testing would damage the credibility of our statistical system. The BLS has requested
funds to support publication of a superlative index as a complement to the CPI beginning
in 2002. Additionally, the BLS has requested funds to expand the Consumer Expenditure
Survey sample size and develop enhanced computer systems to support the timely
introduction of more current spending data when the index weights are next updated.
Finally, the BLS will announce a decision to make more frequent updates to expenditure
weights in the near future.
The BLS should change its procedure for combining price quotations by moving to
geometric means at the elementary aggregates level.
To address lower-level substitution bias, the commission suggested the adoption of a
geometric mean formula for aggregating price quotations, a formula that has been under
investigation by the BLS over the past several years. On April 17, 1998 the BLS announced
that the geometric mean formula will be used in the CPI in place of the current Laspeyres
formula in all item strata except for selected shelter services, selected utilities and
government charges and selected medical services.13
This change will be implemented with the CPI release for January 1999. Our best
estimate is that the planned use of the geometric mean formula will lower the growth rate
of the CPI by approximately 0.2 percentage point per year. A discussion of the decision
and the rationale for the adoption of the geometric mean formula is presented in the paper
entitled "Planned Change in the Consumer Price Index Formula" which is appended
to this document.
Additional recommendation regarding expanded use of hedonic regressions for making
The advisory commission report did not contain an explicit recommendation concerning
the use of hedonic regressions, but the commission members later emphasized their support
for increased use of this approach to quality adjustment.14
This is, in fact, a critical element of BLS current practice and future plans. For
several years, hedonic quality adjustments have been used in the CPI apparel and rent
indexes. Starting in January 1998, the hedonic model developed and used in the Producer
Price Index for adjusting computer prices has been employed in the Personal Computers and
Peripheral Equipment category of the CPI. New hedonic research is underway on video and
audio equipment data. Most importantly, funds have been requested to support ongoing
collection of product price and characteristic data, focused on consumer durable items. It
is expected that this information will enable the BLS to expand significantly the use of
hedonic quality adjustment methods in future years.
In summary, the BLS agrees with the commission that the use of hedonic regression
techniques in quality adjustment should be expanded. The BLS has used hedonic quality
adjustment for selected CPI components for several years, has recently extended its use to
additional components, and has requested funds for continued expansion.
Table of Contents
Download PDF version (77K)
Last Modified Date: October 16, 2001