Attachment VI - Handbook of Methods Chapter 17

Chapter 17.
The Consumer Price Index (Updated 06/2015)

T

he Consumer Price Index (CPI) is a measure of the
average change over time in the prices of consumer
items—goods and services that people buy for day-today living. The CPI is a complex measure that combines economic theory with sampling and other statistical techniques
and uses data from several surveys to produce a timely and
precise measure of average price change for the consumption
sector of the American economy. Production of the CPI requires the skills of many professionals, including economists,
statisticians, computer scientists, data collectors and others.
The CPI surveys rely on the voluntary cooperation of many
people and establishments throughout the country who, without compulsion or compensation, supply data to the government’s data collection staff.

IN THIS CHAPTER
Part I: Overview of the CPI................................................. 1
CPI concepts and scope................................................... 2
CPI structure and publication.......................................... 3
Calculation of price indexes............................................ 3
CPI publication................................................................ 3
How to interpret the CPI................................................. 5
Uses of the CPI................................................................ 5
Limitations of the index.................................................. 6
Experimental indexes...................................................... 6
History of the CPI, 1919 to 2013..................................... 7
Part II: Construction of the CPI.......................................... 11
Sampling: areas, items, and outlets................................. 11
Area sample..................................................................... 11
Item and outlet samples................................................... 12
Commodities and services other than shelter................. 12
 Shelter ............................................................................. 16
Estimation of price change in the CPI............................ 18
Estimation of price change for commodities
and services other than shelter................................... 18
Item replacement and quality adjustment...................... 19
Estimation of price change for shelter........................... 20
Special pricing and estimation procedures
or medical care........................................................... 22
Special pricing for other items........................................ 25
Special pricing for seasonal items................................... 26
Other price adjustments and procedures......................... 27
Index calculation............................................................. 28
Estimation of upper-level price change .......................... 28
Calculation of seasonally adjusted indexes.................... 34
Calculation of annual and semiannual average
  indexes......................................................................... 35
Average prices................................................................. 35

Part I. Overview of the CPI
Three CPI series. The Bureau of Labor Statistics (BLS;
the Bureau) publishes CPI data every month. The three main
CPI series are
• CPI for All Urban Consumers (CPI-U)
• CPI for Urban Wage Earners and Clerical Workers
(CPI-W)
• Chained CPI for All Urban Consumers (C-CPI-U)
The CPI for All Urban Consumers, or CPI-U, which BLS
began publishing in January 1978, represents the buying
habits of the residents of urban or metropolitan areas in the
United States. The CPI for Urban Wage Earners and Clerical
Workers, or CPI-W, the oldest of the series, covers a subset of
the urban population.1 The prices used for producing these
two series are the same. The CPI-U and CPI-W differ only in

Part III: Precision of CPI estimates..................................... 37
Technical references............................................................ 41
Appendix 1. List of published indexes................................ 43
Appendix 2. List of average prices..................................... 55
Appendix 3. Characteristics of CPI changes...................... 59
Appendix 4. Sample areas and weights.............................. 64
Appendix 5. Expenditure classes, strata, and ELIs............ 68
Appendix 6. Sample allocation methodology for
commodities and services............................................... 82
Appendix 7. POPS categories.............................................. 88
Appendix 8. Non-POPS sample designs............................. 101
Appendix 9. CPI Relative importances............................... 103

1 Specifically, the CPI-U (all-urban) population consists of all urban households in Metropolitan Statistical Areas (MSAs) and in urban places of 2,500
inhabitants or more. Nonfarm consumers living in rural areas within MSAs
are included, but the index excludes rural nonmetropolitan consumers and
the military and the institutional population. The urban wage earner and
clerical worker (CPI-W) population consists of consumer units with clerical
workers, sales workers, protective and other service workers, laborers, or
construction workers. More than one-half of the consumer unit’s income
has to be earned from these occupations, and at least one of the members
must be employed for 37 weeks or more in an eligible occupation.

1

the consumer spending weights used to combine, or average
together, basic indexes.2
The Chained CPI for All Urban Consumers (or C-CPI-U),
also represents the urban population as a whole. BLS began
publishing this series in August 2002 with data beginning
in January 2000. The prices used in the C-CPI-U are the
same as those used to produce the CPI-U and CPI-W, but the
C-CPI-U uses a different formula and different weights to
combine basic indexes. The formula used in the C-CPI-U accounts for consumers’ ability to achieve the same standard of
living from alternative sets of consumer goods and services.
This formula requires consumer spending data that are not
immediately available. Consequently, the C-CPI-U, unlike
the other two series, is published first in preliminary form
and is subject to scheduled revisions.

items in various categories of consumer spending—such as
food, clothing, shelter, and medical services—that people
buy for day-to-day living. The monthly movement in the CPI
derives from weighted averages of the price changes of the
items in its sample. A sample item’s price change is the ratio
of its price at the current time to its price in a previous time.
A sample item’s weight in this average is the share of total
consumer spending that it represents. The algebraic formulas
used for this averaging are called index number formulas.3
A unifying framework for dealing with practical questions that arise in the construction of the CPI is provided by
the concept of the cost-of-living index (COLI)4. As it pertains
to the CPI, the COLI for the current month is based on the
answer to the following question: “What is the cost, at this
month’s market prices, of achieving the standard of living
actually attained in the base period?” This cost is a hypothetical expenditure—the lowest expenditure level necessary
at this month’s prices to achieve the base-period’s living standard. The ratio of this hypothetical cost to the actual cost of
the base-period consumption basket in the base period is the
COLI. Unfortunately, because the cost of achieving a living
standard cannot be observed directly, in operational terms a
COLI can only be approximated. Although the CPI cannot be
said to equal a cost-of-living index, the concept of the COLI
provides the CPI’s measurement objective and is the standard
by which we define any bias in the CPI. BLS long has said
that it operates within a cost-of-living framework in producing the CPI.5 That framework has guided, and will continue
to guide, operational decisions about the construction of the
index.
Because the COLI is not directly observable, the CPI employs index number formulas that offer approximations to
the measurement objective. The CPI-U and the CPI-W use
a Laspeyres formula to average the price changes across
categories of items. It is sometimes said that the Laspeyres
formula provides an “upper bound” on the COLI index. The
C-CPI-U uses a Törnqvist formula to average price changes
across item categories. This formula belongs to a class of formulas called superlative because, under certain assumptions,
they can provide close approximations to a COLI. Since 1999,
the CPI program has used the geometric mean formula to average price change within most item categories. Under certain
assumptions that are likely to be true within most categories,
an index based on the geometric mean formula will be closer
to a COLI than will a Laspeyres index.

CPI populations. A consumer price index measures the
price-change experience of a particular group called its target population. The CPI uses two target populations for its
main series:
• All Urban Consumers (the “CPI-U” population)
• Urban Wage Earners and Clerical Workers (the “CPIW” population)
Both the CPI-U and the C-CPI-U target the CPI-U population. The CPI-U population, which covers about 88 percent
of the U.S. population, covers households in all areas of the
United States except people living in rural nonmetropolitan areas, in farm households, on military installations, in
religious communities, and in institutions such as prisons and
mental hospitals.
The CPI-W population, the target of the CPI-W, is a subset of the CPI-U population. The CPI-W population consists
of all CPI-U population households for whom 50 percent or
more of household income comes from wages and clerical
workers’ earnings. The CPI-W’s share of the total U.S. population has diminished over the years; the CPI-W population is
now about 28 percent of the total U.S. population. The CPI-W
population excludes households of professional and salaried
workers, part-time workers, the self-employed, and the unemployed, along with households with no one in the labor
force, such as those of retirees.

CPI concepts and scope
The CPI provides an estimate of the price change between
any two periods. The CPI follows the prices of a sample of

3 For a review of index number formulas, their properties, and their relationship to economic theory, see W. E. Diewert, “Index numbers,” in J.
Eatwell, M. Malgate, and P. Newman eds., The new Palgrave: a dictionary
of economics, vol. 2 (London: The MacMillan Press, 1987), pp. 767–780.
4 For more information on the cost-of-living index concept, see the technical references at the end of this chapter.
5 On the use of a cost-of-living index as a conceptual framework for practical decision making in putting together a price index, see Robert Gillingham, “A conceptual framework for the revised Consumer Price Index.”
Proceedings of the American Statistical Association, Business and Economic Statistics Section (Alexandria: VA, American Statistical Association, 1974), pp. 46–52.

2 Until 1982, BLS maintained separate (but overlapping) samples of outlets and specific items for the CPI-U and CPI-W populations. Given little
variance in the movements between the CPI-U and CPI-W, BLS dropped the
separate samples for the CPI-W population. The CPI-U converted to rental
equivalence effective with the indexes for January 1983; the CPI-W moved
to rental equivalence 2 years later. Since January 1985, the movements of all
CPI-W basic indexes have been identical to those of the CPI-U.

2

Government-provided and government-subsidized items.
The CPI treats as price changes any changes to fees that the
government charges for items, such as admission to a national park. The CPI also counts the price of subsidized items
that are available to the general public. For example, governments may subsidize local transit operation. If the subsidy is
cut and the fare is raised, the CPI will reflect this price increase. On the other hand, the CPI does not reflect changes to
means-tested (dependent on the recipient’s income) subsidies,
such as the Supplemental Nutrition Assistance Program or
Section 8 housing allowances. Changes in such subsidies are
treated as changes to the recipient’s income and, therefore,
out of scope.

Scope

The cost of maintaining a standard of living is affected by
phenomena that go beyond the traditional domain of a consumer price index—changes in the cost of consumer goods
and services. The broadest form of a COLI, which is called
an unconditional COLI, would reflect changes in non-price
factors such as crime rates, weather conditions, and health
status. The objective of the CPI, by contrast, is to provide
an approximation to a conditional COLI that includes only
the prices of market goods and services or governmentprovided goods for which explicit user charges are assessed.
Free goods, characteristics of the environment (such as air
and water quality), the value of leisure time, and items that
governments provide at no cost are not in scope, although
they undeniably can have an impact on the cost of living as
broadly defined.

CPI structure and publication

Excluded goods and services. The CPI covers the consumption sector of the U.S. economy. Consequently, it excludes
investment items, such as stocks, bonds, real estate, and business expenses. Life insurance also is excluded for this reason, although health, household, and vehicle insurance are in
scope. Employer provided in-kind benefits are viewed as part
of income. Purchases of houses, antiques, and collectibles are
viewed as investment expenditures and therefore excluded.
Gambling losses, fines, cash gifts to individuals or charities, and child support and alimony payments also are out
of scope. Changes in interest costs or interest rates are now
excluded from the CPI scope, although some were in the CPI
for many years. And, for practical reasons, the CPI excludes
illegal goods and services and the value of home-produced
items other than owners’ equivalent rent.

Calculation of price indexes
In the CPI, the urban portion of the United States is divided into 38 geographic areas called index areas, and the set
of all goods and services purchased by consumers is divided
into 211 categories called item strata. This results in 8,018 (38
× 211) item–area combinations.
The CPI is calculated in two stages. The first stage is the
calculation of basic indexes, which show the average price
change of the items within each of the 8,018 CPI item–area
combinations. For example, the electricity index for the Boston CPI area is a basic index. The weights for the first stage
come from the sampling frame for the category in the area.
At the second stage, aggregate indexes are produced by averaging across subsets of the 8,018 CPI item–area combinations. The aggregate indexes are the higher level indexes; for
example, the all-items index for Boston is an average of all of
the area’s 211 basic indexes. Similarly, the aggregate index
for electricity is an average of the basic indexes for electricity
in each of the 38 index areas. The U.S. city average All-items
CPI is an average of all basic indexes. The weights for the
second stage are derived from reported expenditures from
the Consumer Expenditure Survey (CE).

Taxes. Both the CPI and the conditional COLI measure
changes in expenditures—including the effect of changes in
sales taxes and similar taxes that are part of the final price of
consumer products—needed to achieve the base-period standard of living. Neither the CPI nor the COLI, however, measures the change in before-tax income required to maintain
the base-period living standard. For this reason, neither the
COLI nor the CPI is affected by changes in income and other
direct taxes. For certain purposes, one might want to define
price indexes that include, rather than exclude, income taxes.6
The CPI does include the effects of changes in sales taxes and
other indirect taxes. As previously noted, however, these are
included as part of the price of consumer products. No attempt is made to reflect changes in the quantity or quality of
government services paid for through taxes.

CPI publication
Indexes. Each month’s index value displays the average
change in the prices of consumer goods and services since
a base period, which currently is 1982–84 for most indexes.
For example, the CPI-U for July 2013 was 233.596. One interpretation of this is that a representative set of consumer
items that cost $100 in 1982–84 would have cost $233.60 in
July 2013.
Percent change. Rather than emphasizing the level of the
index in comparison to the base period, the monthly CPI
release stresses the CPI’s percent change from the previous
month and from the previous year. The most commonly reported monthly percent changes are the one-month seasonally adjusted percent change, and the 12-month not seasonally

6
One could develop an index along these lines. Such an index (sometimes
called a tax-and-price index) would provide an answer to a different question (along the lines of “At current prices, what is the least before-tax income needed to buy…”) from the one that is relevant to the CPI. It would
be appropriate for different uses. For a research measure of a consumption index inclusive of income taxes and Social Security contributions, see
Robert Gillingham and John Greenlees, “The impact of direct taxes on the
cost of living.” Journal of Political Economy, August 1987, pp. 775–796.

3

CPI item indexes. BLS classifies the CPI market basket of
consumer goods and services into a hierarchy of categories.
The top levels of the item category hierarchy consist of

adjusted percent change. For example, the July 2013 CPI was
233.596 and the July 2012 CPI was 229.104, so the CPI increased 2.0 percent (not seasonally adjusted) from July 2012
to July 2013.

•
•
•
•

CPI area indexes and CPI item indexes. BLS publishes a
large number of additional CPI index series. (See appendix
1.) For the CPI-U population areas—the broadest geographic coverage—detailed item indexes for most categories of
consumer spending are published every month. Also every
month, BLS publishes all-items indexes, along with a limited
set of detailed indexes, for the three largest metropolitan areas and for the major geographic areas. In addition, detailed
food, energy, and shelter indexes are published monthly for
all CPI publication areas. Bimonthly or semiannually, allitems indexes for selected metropolitan areas are published
along with the limited set of detailed indexes.
The primary reason for publishing CPI area-item detail
indexes is to aid in analysis of movements in the national allitems CPI. Decisions on which detailed indexes to publish
depend, in part, on the reliability of their estimates7. CPI area
indexes and CPI item detail indexes use only a portion of the
CPI sample; this makes them subject to substantially greater
sampling error than the national CPI. For this reason, BLS
strongly urges users to consider the U.S. city average allitems CPI for use in escalator clauses.

The eight major groups
Other groups
Expenditure classes
Item strata

For the U.S. CPI, BLS publishes all levels down to item
strata. BLS publishes less item detail for the CPI area indexes.
Special aggregations. BLS also calculates and publishes
indexes for special aggregations, such as energy items, that
cut across the preceding classification scheme. Some users
consider the series All items less food and energy to measure
the ‘core’ rate of inflation. Food and energy are two of the
most volatile components of the CPI. For this reason, many
analysts regard the measure of core inflation as more useful
for their purposes.
The C-CPI-U. The Chained CPI-U uses a superlative index
formula which reflects consumers’ behavior in response to
changes in relative prices. Unfortunately, this requires current expenditure data, and expenditure data become available
only after a significant lag. Consequently, C-CPI-U index
values, unlike the values of the CPI-U and CPI-W, are not
final when first published. Before 2015, BLS issued two annual preliminary estimates before issuing final C-CPI-U data.9 Starting in 2015, BLS intends to issue four preliminary
estimates of the C-CPI-U. The “initial” values will come out
every month concurrent with the CPI-U and CPI-W. In each
of the following four quarters, “interim” values will replace
the initial values. One year later, the interim values will be
replaced with the final C-CPI-U. For example, in February
2016, the BLS is scheduled to release the January 2016 CPIU, the CPI-W, and the initial C-CPI-U. For the next three
quarters (i.e., April, July, and October of 2016), BLS will
publish updated interim C-CPI-U indexes. With the fourth
revision in January 2017, the January 2016 C-CPI-U will be
issued as final.

CPI area indexes. BLS calculates and publishes separate
area indexes for
• Four geographic regions (sometimes called census
regions): Northeast, Midwest, South, and West
• Three population-size classes: large metropolitan areas, small metropolitan areas,8 and nonmetropolitan
urban places
• Selected region-size classes—regions cross-classified
by population size (for example, large metropolitan
areas in the Northeast)
• Selected metropolitan areas
Comparing the CPI for an area with the U.S. CPI or with
the CPI for another area gives an indication of differences
among the areas’ rates of price change. In other words, such a
comparison indicates whether, over time, prices of items that
consumers in one area tend to buy have risen more or less
rapidly than the prices of items that consumers in another
area tend to buy. It does not indicate whether the average
level of prices in an area is higher or lower than the average
level in another area.

Seasonally adjusted indexes and percent changes. In addition to the originally computed indexes and percent changes,
which are called unadjusted indexes and unadjusted percent
changes, BLS calculates and publishes seasonally adjusted
series. The unadjusted numbers reflect the change in price
resulting from all causes, including normal seasonal price
movement due to regular changes—resulting, for example,
from weather, harvests, the school year, production cycles,
model changeovers, holidays, or sales—that recur every year.
For economic analysis and for other purposes, it is useful to

7 Steven Grandits, “Publication strategy for the 1998 revised Consumer
Price Index,” Monthly Labor Review, December 1996, pp. 26–30, http://
stats.bls.gov/opub/mlr/1996/12/art4full.pdf.
8Prior to January 1998, the CPI published data for medium and small metropolitan areas, which have been combined to form a single class.

9 The first release of C-CPI-U data took place on Aug. 16, 2002. At that
time, final data for the 12 months of 2000, interim data for the 12 months of
2001, and initial data for the first 7 months of 2002 were issued.

4

remove the estimated seasonal effects from the original indexes and percent changes. To produce the seasonally adjusted
indexes and percent changes, BLS uses seasonal adjustment
techniques that remove these effects. BLS seasonally adjusts
only those CPI series that pass certain statistical criteria and
for which there is an economic rationale for observed seasonality. For example, while the unadjusted CPI for All items
was unchanged from June 2013 to July 2013, the seasonally
adjusted 1-month percent change in the CPI was 0.2 percent.
Seasonally adjusted indexes are subject to annual revision and
therefore are not recommended for use in escalation contracts.
Seasonal adjustment is done only at the national level for the
U.S. city average CPI-U and CPI-W. Presently, the C-CPI-U
does not have sufficient historical data to permit calculation of
stable seasonal factors.

affects index point changes, but it does not affect percent
changes. The following tabulation shows how to compute
percent changes:
Index point change
CPI............................................................222.742
Less CPI for previous period................. 221.317
Equals index point change............................ 1.425

Percent change
Index point difference...................... 1.425
Divided by the previous index..... 222.742
Equals.............................................. 0.006
Average prices. For some food, beverage, and energy items,
Results multiplied by.............0.006 × 100
the CPI samples contain enough observations
of
unique
items
Equals percent
change ....................... 0.6
Index point difference
1.425
to make possible the computation and Divided
publication
of
meanby the previous index
222.742
ingful average retail prices. A list of what
is covered in the
Equals
0.06
Percent changes
periods other than 1 year often are expressed
published average price series is shownResults
in appendix
2. by 100.0
multiplied
0.06 ×for
100
Annualized percent changes indicate
Equals percent change as annualized percentages.
0.6
Correction policy. The CPI, unlike many other statistical what the change would be if the CPI continued to change at the
changes
other than 1 year often are expressed as annualized percentages. Annualize
series, does not rely on respondentsPercent
to transmit
datafortoperiods
the
same
each would
month be
over
a 12-month
period. to
These
are at
calcupercent
changes
indicate
what
therate
change
if the
CPI continued
change
the same rate eac
national office. CPI data collectors collect almost all data
month
over
a
12-month
period.
These
are
calculated
using
the
standard
formula
for
compound
growth:
lated
using
the
standard
formula
for
compound
growth:
needed for the CPI-U and CPI-W, so that routine revisions to

account for late-arriving data are not necessary. Virtually all
12/ m
data are received in time for the calculation of indexes for the
PC
 IX t m / IX t  – 1  100 ,
annual
appropriate month. In rare cases, however, when we discover
that we made an error collecting or compiling information, where
wherein accordance with
BLS issues corrections to the CPI series

IXt is the index in month t,
BLS policy and CPI practices.

IXt + m is the index m months after month t, and
IXt is the index in month t,
Corrections to the CPI-U and CPI-W. These series are final

PCannual is the annualized percent change.
when issued. The CPI-U and CPI-W are
commonly
usedm in
IXt+m
is the index
months after month t, and
escalation agreements and to adjust pensions and tax brackUses of the CPI
ets; consequently, revisions can be costly for the users of
PCannual is the annualized percent change.
these indexes. For this reason, there is a presumption in BLS
The CPI affects virtually all Americans because of the many
policy and practice against revisions to the CPI that extend
back over lengthy periods. When a mistake is discovered, ways in which it is used. Its major uses are as follows:
CPI staff evaluates the error in the context of BLS guidelines
• As an economic indicator. As the most widely used
of the
for issuing corrections to previously Uses
published
CPI CPI
data.
measure of retail inflation, the CPI is a major indicator
Corrections to the C-CPI-U. As previously
noted, virtually
C-CPI- all Americans
of thebecause
effectiveness
of Government
economic
policy.
The CPI affects
of the many
ways in which
it is used.
Its major uses ar
U indexes are not final when first issued.
They
are
routinely
The
President,
the
Congress,
and
the
Federal
Reserve
as follows:
revised, and are not final until the publication of data for the
Board use the movement of the CPI to help formulate
second January after initial publication. If theAsCPI-U
and
and monitor
the effect
fiscalmeasure
and monetary
an economic
indicator.
As the most
widelyofused
of retailpolicies.
inflation, the CPI is a
CPI-W series are corrected, the C-CPI-U seriesmajor
will be
cor- of the effectiveness
indicator
of Government
economic
President, the
Business executives,
labor leaders,
andpolicy.
other The
private
rected as well. Corrected C-CPI-U indexes will be
issued for
Congress,
and the Federal
Reserve
useindex
the movement
CPI toecohelp formulate an
citizens
alsoBoard
use the
as a guideofinthe
making
all series affected by the error, as far back as the
previous
monitor
the5effect of fiscal
anddecisions.
monetary policies. Business executives, labor leaders, and
nomic
years.
other private citizens also use the index as a guide in making economic decisions.
• Asincome
a means
of adjusting
payments.
Thethe
index
 As a means of adjusting
payments.
Theincome
index directly
affects
income of almos
10
directly
affects
the
income
of
almost
80
million
peoService pension
80 million people. Social Security benefits and military and Federal Civil
How to interpret the CPI payments are all indexed
ple.bySocial
Security
benefitssector,
and military
and Federal
the CPI.
In the10 private
many collective
bargaining
Movements of the indexes from one month to agreements
another usu-tie automatic wage increases to the CPI. Some private firms and individuals use
the index to keep rents, alimony, and child support payments in line with changing prices.
ally are expressed as percent changes rather than changes in
10 Specific information on the Social Security use of the CPI can be
 As a means of preventing
inflation-induced tax changes. Federal (and some state) income
found on the Social Security Administration website, http://www.
index points. The level of the index (relative to itstax
base
period)
brackets andsocialsecurity."gov/cola/."
other parameters are adjusted by the CPI. This prevents inflation from
automatically increasing taxes, a phenomenon called bracket creep.
 As a deflator of other economic series. Other statistical programs use the CPI or its
components to5adjust for price changes and produce inflation-free versions of their series.
Examples of CPI-adjusted series include components of the U.S. Department of Commerce





Civil Service pension payments are all indexed by the
CPI. In the private sector, many collective bargaining
agreements tie automatic wage increases to the CPI.
Some private firms and individuals use the index to
keep rents, alimony, and child support payments in
line with changing prices.
•

•

obtaining prices through personal observation whenever possible, and by correcting errors immediately upon discovery.
The economic assistants, technicians, and commodity specialists who collect, process, and analyze the data are trained
to watch for deviations in reported prices that might be due
to errors.
A full discussion of the varieties and sources of possible
error in the index is presented in part III of this chapter, “Precision of CPI Estimates.”

As a means of preventing inflation-induced tax
changes. Federal (and some state) income tax brackets and other parameters are adjusted by the CPI. This
prevents inflation from automatically increasing taxes, a phenomenon called bracket creep.

Experimental indexes

As a deflator of other economic series. Other statistical programs use the CPI or its components to adjust
for price changes and produce inflation-free versions
of their series. Examples of CPI-adjusted series include components of the U.S. Department of Commerce National Income and Product Accounts (such
as gross domestic product and personal consumption
expenditures) and retail sales measures and the BLS
hourly and weekly earnings series.

Population subgroups. The CPI also calculates and publishes
some indexes on an experimental basis only. For example, the
program provides experimental indexes for the elderly. Comparing indexes for such subgroups does not indicate whether
the prices they pay are higher or lower than the prices other
groups pay; this comparison indicates only whether prices of
their items have risen faster or slower than those for other
groups. Indexes for subgroups of the population are more difficult to construct than indexes for the whole. In particular,
making sure that samples refer to only part of the population
may be difficult or impractical. Moreover, making subgroup
indexes as precise as the national CPI would require that the
sample sizes be as large.

Limitations of the index
The CPI covers a wide variety of items that all urban consumers purchase, but—because most individuals concentrate
spending on a relatively small fraction of the total number of
items available in the market—it contains items that a given
individual does not purchase. The CPI must represent a composite consumer, and it does not necessarily represent the
price-change experience of any one individual, household, or
family. Similarly, the CPI may not be applicable to all questions about price movements for all population groups.

The experimental CPI for Americans 62 Years of age and
older (CPI-E). BLS occasionally issues a report on its experimental index for the elderly. This index, sometimes referred
to as the CPI for the elderly or CPI-E, is calculated monthly
and is available on request. It should be emphasized that the
CPI-E is merely a reweighting of the CPI basic indexes using
expenditure weights from households headed by someone 62
years of age or older. There is no attempt to recalculate the
basic indexes themselves so that they represent the retail outlets and consumption items of older consumers.11

As previously noted, CPI indexes cannot be used to determine relative living costs. The CPIs for various geographic
areas of the United States do not indicate the differences in
price level among them. The change in the CPI for an individual area measures the degree to which prices have changed
over time within that particular area. It does not show whether
prices or living costs are higher or lower in that area relative
to another area or to the United States as a whole. Comparing
indexes between one area and another indicates which area
has experienced more rapid price change—not which area has
a higher price level or higher living costs.

CPI research series. Over the years, BLS has made many
improvements to the CPI. When BLS changes its methods, it
always announces them in advance and, if possible, estimates
the impact the change would have had in recent periods. BLS
does not, however, revise previously published CPI data to
reflect the new methods. This practice means that the movement of the CPI reflects not only price change over time but
also changes to CPI methods. To assist users who wish to use
the CPI over long periods, BLS publishes the CPI-U Research
Series Using Current Methods (CPI-U-RS). It provides estimates, for the period since 1977, of what the CPI would have
been had the most current methods been in effect. Each time
there are new methods introduced into the CPI, the CPI-U-RS
is revised from 1978 forward.12

Sampling and non-sampling error. The CPI is estimated
from a sample of consumer purchases; it is not a complete
measure of price change. Consequently, the index results may
deviate slightly from those that would be obtained if all consumer transactions were covered. This is called sampling error. These estimating or sampling errors are statistical limitations of the index.
A different kind of error in the CPI can occur when, for example, a respondent provides BLS economic assistants with
inaccurate or incomplete information. This is called nonsampling error. BLS attempts to minimize these errors by

11 For more information, see Consumer Price Index Detailed Report (U.S.
Bureau of Labor Statistics, February 2000), pp. 5–7..
12 Kenneth J. Stewart and Stephen B. Reed, "CPI research series using
current methods, 1978–98," Monthly Labor Review, June 1999, pp. 29–38.

6

History of the CPI, 1919 to 2002
The CPI was initiated during World War I, when rapid increases in prices, particularly in shipbuilding centers, made
such an index essential for calculating cost-of-living adjustments in wages. To provide appropriate weighting patterns
for the index, so that it would reflect the relative importance
of goods and services purchased by consumers, studies of
family expenditures were conducted in 92 industrial centers
in 1917–1919. Periodic collection of prices was started and, in
1919, BLS began publication of separate indexes for 32 cities.
Regular publication of a national index, the U.S. city average,
began in 1921, and indexes were estimated back to 1913.13
Since its inception, the CPI has been comprehensively
revised on several occasions to implement updated samples
and weights, expanded coverage, and enhanced methodologies. For example, the 1998 revision introduced more timely
consumer spending weights; updated geographic and housing samples; a revised item classification structure; a new
housing index estimation system; computer-assisted price
collection; and a new Telephone Point-of-Purchase Survey
(TPOPS). BLS also has made important improvements to
the CPI beyond the major revision processes, an example
being the introduction of the geometric mean formula in
January 1999. Exhibit 1 provides a chronology of revisions
and improvements to the CPI, and appendix 3 displays

historical changes in base period, population coverage, and
other index characteristics.
The improvements introduced over the years have reflected not only the Bureau’s own experience and research, but
also the criticisms and investigations of outsiders. For example, in undertaking the 1940 comprehensive revision of the
CPI, BLS acted on recommendations made by an Advisory
Committee appointed by the American Statistical Association. Major studies were conducted during World War II by
the President’s Committee on the Cost of Living14 and in 1951
by the House Committee on Education and Labor.15
The 1961 report of the Price Statistics Review Committee (sometimes called the “Stigler Committee”) provided
impetus for subsequent changes in many aspects of the CPI,
including the sampling of outlets and items, the treatment of
quality changes in consumer durables, and the role of costof-living theory.16 Recent studies include the 1996 report
of the Advisory Commission to Study the Consumer Price
Index (the “Boskin Commission”)17 and the 2002 report, At
what price? Conceptualizing and measuring cost-of-living
and price indexes, by a National Research Council panel of
the National Academy of Sciences.18 A continuing flow of
articles in professional journals and books also has contributed to the assessment of the CPI’s quality and of the ways in
which it might be improved. For a list of published papers, see
the Technical References at the end of this chapter.

13
Collection of food prices back to 1890 had been initiated in 1903. During the course of the 1917–1919 expenditure survey, retail prices for other
items were collected in 19 cities for December of each year back to 1914,
and in 13 other cities back to December 1917 only. Retail prices of food and
wholesale prices of other items were used to estimate price change from
1914 back to 1913.

14
Report of the president's committee on the cost of living (Washington,
Office of Economic Stabilization, 1945).
15
Consumer Price Index, report of a special subcommittee of the committee on education and labor, Subcommittee Report No. 2 (U.S. Congress,
House of Representatives, 1951).
16
Government price statistics, hearing before the subcommittee on economic statistics, U.S. Congress, 871. Part 1 (U.S. Congress Joint Economic
Committee, January 24, 1961).
17
Final report of the advisory commission to study the Consumer Price
Index (The Boskin Commission Report) (U.S.Senate Committee on Finance, December 1996).
18
Charles Schultze and Christopher Mackie, eds. At what price? Conceptualizing and measuring cost-of-living and price indexes. (Washington,
DC: National Academy Press, 2002.

7

Exhibit 1. Chronology of changes in the Consumer Price Index
The Consumer Price Index to 1940
• Began publication of separate indexes for 32 cities (1919):
  Collected prices in central cities periodically
• "Developed weights from a study that BLS conducted in 1917–1919 of family expenditures in 92 industrial
centers, reflecting the relative importance of goods and services purchased by consumers"

  Reflected the relative importance of goods and services purchased by consumers

• Collected prices for major groups: Food, clothing, rent, fuels, house furnishings, and miscellaneous
• Limited pricing to items selected in advance to represent their categories
• Began regular publication of a national index, the U.S. city average (1921):
  Based index on an unweighted average of the city indexes
  Estimated U.S. city average back to 1913, using food prices only
The 1940 CPI revision: the first comprehensive revision
• Used weights based on 1934–1936 study of consumer expenditures
•
Collected prices in the 34 largest cities
Implemented a weighted average of cities for the U.S. city average CPI
Improvements made between the 1940 and 1953 revisions
• During World War II:
Discontinued the pricing of unavailable items, such as new cars and household appliances

Increased the weight of other items, including automobile repair and public transportation in 1951:

Adjusted weights in seven cities, using a 1947 and 1949 survey of consumer expenditures
Adjusted weights for the 1950 Census
Adjusted rent index to remove “new unit bias” caused by rent control
Added new items to the list of covered items, including frozen foods and televisions
The 1953 CPI revision: the second comprehensive revision
• Used weights from a 1950 expenditure survey conducted in central cities and attached urbanized areas
• Refined the target population to include urban wage earner and clerical worker families
• Added a sample of medium and small cities
• Updated the list of items that the index covered, adding restaurant meals
• Added new sources of price data
• Improved pricing and calculation methods
The 1964 CPI revision: the third comprehensive revision
• Based weights on 1960–1961 expenditure patterns in metropolitan areas
• Added single-person households to target population: urban wage earner and clerical worker households
• Extended pricing to the suburbs of sampled metropolitan areas
• Updated the sample of cities, goods and services, and retail stores and service establishments
Improvements made between the 1964 and 1978 revisions
•
Made quality adjustments for new vehicles at model changeover
•
Improved treatment of seasonal items
The 1978 CPI revision: the fourth comprehensive revision
• Added a new Consumer Price Index: the CPI for All Urban Consumers, or the CPI-U

8

Exhibit 1. Chronology of changes in the Consumer Price Index—continued
•
•
•
•
•
•
•
•

Renamed the older CPI [as] the CPI for Urban Wage Earners and Clerical Workers, or the CPI-W
Used weights from a 1972–1973 survey of consumer expenditures and the 1970 census
Expanded the sample to 85 areas
Increased minimum pricing frequency from quarterly to bimonthly
Implemented monthly pricing in five largest areas
Introduced probability sampling methods at all stages of CPI sampling
Introduced checklists that define each category of spending
Developed estimates of the CPI’s sampling error and optimal sample allocation to minimize that error

Improvements made between the 1978 and 1987 revisions
• Began systematic replacement of outlets and their item samples between major revisions (1981):
Implemented new Point-of-Purchase Survey (POPS)
Selected retail outlets with probability proportional to consumer spending therein
Eliminated reliance on outdated secondary-source sampling frames
Began rotating outlet and item samples every 5 years
Began rotating one-fifth of the CPI pricing areas each year
• Introduced rental equivalence concept (January 1983 for the CPI-U; January 1985 for the CPI-W)
The 1987 CPI revision: the fifth comprehensive revision
• Used weights from the 1982–1984 Consumer Expenditure Survey and the 1980 Census
• Updated samples of items, outlets, and areas
• Redesigned the CPI housing survey
• Improved sampling, data collection, data-processing, and statistical estimation methods
• Initiated more efficient sample design and sample allocation
• Introduced techniques to make CPI production and calculation more efficient
Improvements made between the 1987 and 1998 revisions
• Improved housing estimator to account for the aging of the sample housing units
• Improved the handling of new models of vehicles and other goods
• Implemented new sample procedures to prevent overweighting items whose prices are likely to rise
• Improved seasonal adjustment methods
• Initiated a single hospital services item stratum with a treatment-oriented item definition
• Discontinued pricing of the inputs to hospital services
The 1998 CPI revision: the sixth comprehensive revision
• Weights from the 1993–1995 Consumer Expenditure Survey and the 1990 census
• Updated geographic and housing samples
• Extensively revised item classification system
• Implemented new housing index estimation system
• Used computer-assisted data collection
• Added the Telephone Point-of-Purchase Survey (TPOPS):

Allows rotation of outlet and item samples by item category and geographic area, rather than by area alone
Improvements since the 1998 Revision
• Initiated a new housing survey based on the 1990 census (January 1999):
Estimated price change for owners’ equivalent rent directly from rents
Began using a geometric mean formula for most basic indexes (January 1999):
Mitigates lower level substitution bias

9

Exhibit 1. Chronology of changes in the Consumer Price Index—continued
•
•
•
•
•
•
•

Reflects shifts in consumer spending with item categories as relative price change
Extended the use of hedonic regression to estimate the value of items changing in quality
Directed replacement of sample items in the personal computer and other categories, to keep samples current
Implemented 4-year outlet rotation to replace 5-year scheme
Began within-outlet item rotation for prescription drugs and other item categories
Implemented biennial weight updates starting January 2002
Increased sample size of the Consumer Expenditure Survey, so that CPI weights could be based on just 2 years
of data

Added the Chained Consumer Price Index for All Urban Consumers (C-CPI-U) (August 2002)
•
•
•
•

Uses more advanced “superlative” index formula (the Törnqvist formula)
Corrects upper-level substitution bias
Expanded collection of price data to all business days of the month (Before 2004, prices were collected the first
18 business days of the month for the first 10 months of the year and the first 15 business days for November and
December.)
Began publishing indexes to three decimal places (January 2007)

10

Part II. Construction of the CPI

Sampling: areas, items, and outlets
The smallest geographic areas in which pricing is done
for the CPI are called primary sampling units (PSUs). Within
these areas, sales outlets are chosen where people shop and
live. The selected nonshelter outlets are matched to a sample
of items that these consumers buy. Appendix 4 lists the 87
PSUs selected for the 1998 revision and the counties contained therein. Prices from these were introduced into CPI
index calculation with the release of the January 1998 index.
Area sample
For the purpose of selecting the 1998 CPI PSU sample,19
the entire United States was divided into PSUs. First, BLS
used the U.S. Office of Management and Budget (OMB)
definition of Metropolitan Areas (MAs)20 to divide the country into metropolitan and nonmetropolitan areas. The PSUs
within the metropolitan area are, with five exceptions, always
OMB-defined MAs.21 In the nonmetropolitan areas, BLS defined the PSU boundaries. In general, a PSU is delineated by
county borders (with some exceptions in New England) and
can comprise several counties.
Each PSU was first classified by its size. All PSUs with
populations larger than 1.5 million were declared to be selfrepresenting and given the size type of A.22 The remaining
non-self-representing PSUs, metropolitan and nonmetropolitan, are called B and C PSUs, respectively.23 (To avoid confusion, it is important to recognize the distinction between the
naming conventions for PSUs and those for CPI size-class
indexes. In general, prices collected in B PSUs are used to
19 A more detailed description of the 1998 revision area sample selection
is contained in Janet Williams, Eugene F. Brown, and Gary R. Zion, “The
challenge of redesigning the Consumer Price Index sample,” Proceedings
of the Survey Research Methods Section, American Statistical Association, Business and Economic Statistics Section (Alexandria, VA: American
Statistical Association, 1993), pp. 200–205; and Janet L. Williams, “The
redesign of the CPI geographic sample.” Monthly Labor Review, December
1996, pp. 10–17, http://www.bls.gov/opub/mlr/1996/12/art2full.pdf.
20 MAs are Metropolitan Statistical Areas (MSAs), Primary Metropolitan
Statistical Areas (PMSAs), or Consolidated Metropolitan Statistical Areas
(CMSAs). For more information, see the Statistical Policy Office of the U.S.
Office of Management and Budget, Attachments to OMB Bulletin No. 9305, Metropolitan areas 1992, Lists I–IV..
21 The five PSU exceptions are the Los Angeles suburbs PSU, the three
PSUs that together form the New York–Northern New Jersey–Long Island
CMSA, and the Washington, DC, PSU.
22 Anchorage and Honolulu are A PSUs with smaller populations.
23 When planning began for the 1998 revision of the CPI, one potential
change envisioned was the publication of a Consumer Price Index for the total U.S.population (called the CPI-T). To accommodate this expanded CPI,
a larger number of PSUs was selected to cover the population living in rural
areas outside the metropolitan area.

compute the B/C CPI indexes, and prices in C PSUs are used
in the computation of the D CPI indexes. The exceptions are
the Anchorage and Honolulu metropolitan areas, which are A
PSUs but included in the B/C size class indexes.)
The second classification variable for PSUs is census region.
The next phase of the area selection was to stratify (group)
PSUs in each region-size class; for example, South B into strata
(groups) of similar PSUs based on their scores on several factors (called stratifying variables). Each A PSU is in a stratum
by itself, thus, the name “self-representing.” Selection of the
stratifying variables was based on linear regression modeling
of price change (sequentially finding sums of a constant and a
constant times each of a subset of 1990 census and geographic
PSU variables that best explain CPI price change over different periods). The variables (called geographic variables) used
in all stratifications except that for the South B PSUs were
percent urban, the normalized latitude and longitude of the
PSU’s geographic center, and normalized longitude squared.
In the stratification of the South B PSUs, percent urban and
variables used in the 1987 Revision—namely, mean interest
and dividend income per consumer unit (CU), mean CU wage
and salary income, percent housing units (HUs) heated by
electricity, percent HUs heated by fuel oil, percent owner-occupied HUs; percent black CUs, and percent CUs with retired
person—were used. The program employed to do the stratifications was a modified version of the Friedman-Rubin clustering algorithm, which puts PSUs in the same stratum based on
their similarities on the stratification variables, while keeping
the population sizes of the strata approximately equal.
A program was used to select one PSU per stratum so that
the selected PSUs were well-distributed over the States and
there were many 1988-sample PSUs among the newly selected ones.24 Prices from the 36 newly-selected non-1988-sample
PSUs were introduced into CPI index calculations in 1998.
Since 1998, indexes have been published monthly for the
New York, Los Angeles and Chicago Consolidated Metropolitan Statistical Areas (CMSAs). Indexes for the Washington-Baltimore CMSA, along with the next 10 largest A

24 In case the CPI-T was judged too costly, a selection was made from
nonmetropolitan PSUs that would have their urban parts included in the
CPI-U. Candidate PSUs had to contain some urban population. From these
candidates, a probability (proportional to the urban population of their stratum) sample for the CPI-U was selected in each region except the Northeast.
Long after the selection of the PSUs, a decision was made not to publish a
CPI-T because of its increased cost. At that time, 18 of the previously selected PSUs were dropped from the CPI sample and its increased cost. At that
time, 18 of the previously selected PSUs were dropped from the CPI sample
and designated as “Consumer Expenditure Survey only” PSUs.

11

PSUs (not including those contained in the aforementioned
CMSAs), are published bimonthly. Beginning in January and
July, semiannual indexes are published for the 12 smallest A
PSUs. Indexes are also published for the U.S. total as well as
for region and size class totals, with the exception of the D
indexes in the Northeast and West. Beginning in 2002, semiannual indexes have been published for Phoenix.
Replicates, which are used in variance calculation, are assigned to each A PSU based on population, with each A PSU
having either two or four replicates. B and C PSUs are paired
in each region, with each pair containing a PSU on the even
and odd monthly pricing cycles, except for single PSU pairs
in the region-size classes in which the number of PSUs is not
a multiple of 4. Publication of a region-size class requires at
least four PSUs (two replicates). The actual allocation of replicates is provided in the next section, along with the allocation of replicate panels.25

The primary objective of the Commodities and Services
(C&S) sample design is to determine an allocation of individual item and outlet selections, by item stratum and by PSU,
replicate, and POPS category (see list that follows), that minimizes the sampling variance of price change measured by the
all-cities C&S CPI, subject to certain budgetary and sample
size constraints. Models are used to project the sampling variance and data collection costs in terms of the decision variables
for the sample design. For these models, all commodities and
services item strata are grouped into 13 major groups:
•
•
•
•
•
•
•
•
•
•
•
•
•

Item and outlet samples
Commodities and services other than shelter
Item structure and sampling. The CPI item structure has four
levels of classification. The eight major groups are made up of
70 expenditure classes (ECs), which in turn are divided into
211 item strata. Major groups and ECs do not figure directly
in CPI sample selection, although ECs are used in smoothing
item stratum expenditure estimates during composite estimation. Within each item stratum, one or more substrata, called
entry-level items (ELIs), are defined. ELIs are the ultimate
sampling units for items as selected by the BLS national office. They represent the level of item definition from which
data collectors begin item sampling within each sample outlet. (See appendix 5 for a complete list of consumer ECs, item
strata, and ELIs.)
To enable the CPI to reflect changes in the marketplace,
new item and outlet samples are selected each year, on a rotating basis, for approximately 25 percent of the item strata in
each PSU. Each year, four regional item universes are tabulated from the two most recent years of CE data. Independent
samples of ELIs are selected from the corresponding regional
item universe for each item stratum PSU-replicate scheduled
for rotation that year. Within each sample PSU-replicate, each
item sample is based on a systematic probability-proportional-to-size (PPS) sampling procedure, in which each ELI has a
probability of selection proportional to the CPI-U population
expenditures for the region for the ELI within its stratum.

Food at home—non meat staples
Food at home—meat, poultry, fish
Food at home—fruits and vegetables
Other food at home, plus beverages
(alcoholic and nonalcoholic)
Food away from home
Fuels and utilities
Household furnishings and operations
Apparel
Transportation less motor fuel
Motor fuel
Medical care
Education and communication
Recreation and other commodities and services

In brief, the C&S sample allocation methodology is as follows: First, a variance function that projects the variance of
price change as a function of the preceding variables for the
commodity and service components is modeled. Second, a
cost function that predicts the total annual cost of the commodity and service components of the CPI is modeled. Third,
values for all coefficients of the two functions, including estimates of outlet sample overlap, are estimated. Fourth, nonlinear programming techniques are used to determine approximately optimal sizes for the item and outlet samples needed
to minimize the CPI variance under varying assumptions of
annual price change subject to cost constraints.
The variance and cost functions for the CPI are modeled
for 15 PSU groups:
PSU group name
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•

Item and outlet sample design. The CPI uses two separate
sample designs, one for rent and owners’ equivalent rent, and
one for all other commodities and services. The methodology used to determine the commodities and services item and
outlet sample design is presented here in brief. The design for
the rent and owners’ equivalent rent indexes is described later.

25 For the A PSUs, the number of replicate panels (or groups) is the same
as the number of replicates. Each non-self-representing PSU is allocated
one replicate panel.

12

New York City
New York City suburbs
Los Angeles City
Los Angeles suburbs
Chicago
Philadelphia and San Francisco
Detroit and Boston
Other large self-representing PSUs
Small self-representing PSUs
Medium-sized PSUs, Census Region 1
Medium-sized PSUs, Census Region 2
Medium-sized PSUs, Census Region 3
Medium-sized PSUs, Census Region 4
Small PSUs, Census Regions 1–4
Anchorage and Honolulu

A detailed discussion of the sample allocation methodology is provided in appendix 6. The allocation is resolved with each C&S sample rotation, which occurs twice
each year, and allocations change as sample frames are
refreshed and rotated. For ongoing pricing, about 27,000
outlets are visited each month, with prices collected for
about 83,400 commodities and services.
Outlet and price surveys. BLS economic assistants collect prices monthly for food at home, energy, and selected
other commodity and service item strata in all PSUs. Items
that are priced monthly typically are those with more
volatile and variable price movement. Commodities and
services priced monthly in all PSUs is given in exhibit 2.
Prices also are collected monthly for all commodity and
service item strata in the three largest publication areas:
New York, Los Angeles, and Chicago. Prices are collected
bimonthly in the remaining PSUs for the C&S item strata
not cited in the list. Those are assigned to either even- or
odd-numbered months for pricing.
Exhibit 2. Consumer Price Index (CPI) items
priced monthly everywhere
All food at home items
Housing at school, excluding board
Other lodging away from home, including hotels

and motels
Tenants’ and household insurance
Fuel oil
Propane, kerosene, and firewood
Electricity
Utility (piped) gas service
Used cars and trucks
Gasoline (all types)
Other motor fuels
Tires
Vehicle accessories other than tires
State and local registration, license, and motor vehicle

property tax
Parking and tolls
Newspapers and magazines
Recreational books
Postage
Delivery services
Land line telephone services
Wireless telephone services
Cigarettes
Tobacco products other than cigarettes
Telephone Point-of-Purchase Survey (TPOPS). The U.S.
Census Bureau conducts the TPOPS for BLS. The survey
furnishes data on retail outlets from which metropolitan
and urban nonmetropolitan households purchased defined
groups of commodities and services to be priced in the
CPI. Commodities and services are grouped into sampling categories, called POPS categories. (See appendix

7.) These categories are based on ELIs as defined in the
CPI classification structure. Some POPS categories consist of only one ELI, while others consist of multiple ELIs.
ELIs are combined into a single POPS category when the
commodities or services generally are sold in the same
outlets.
TPOPS uses random-digit dialing to select households
for participation in the survey. Within each PSU, banks
of landline telephone and cell phone numbers containing
at least some residential phone numbers are identified.
Among these identified banks, numbers are then randomly dialed. Inevitably, some of the dialed numbers such as
nonworking, ring-no-answer, business, and FAX machine
telephone numbers are ineligible for TPOPS interviewing.
Some numbers belong to ineligible households, such as
military households.
Eligible respondents include all civilian, noninstitutional persons residing in regular residences, boarding
houses, student or worker housing, mobile home parks,
and permanent-type living quarters in hotels and motels,
as well as staff residing in institutions. Not all of the eligible telephone numbers are productive, however, as respondent refusals are unavoidable.
BLS specifies a target number of completed TPOPS
interviews for each PSU. In the small and medium-sized
PSUs, that target number is 110 completed interviews.
For most of the self-representing PSUs, the target number
of completed interviews ranges from 200 to 400. In New
York City and Chicago, the target number of completed
interviews is 460. In Los Angeles, the specified goal is
500.
Upon first contact and after determining the eligibility
and willingness of the household, a Census Bureau interviewer asks a variety of administrative and demographic
questions. This information allows BLS to monitor how
well the selected households represent the overall population, as well as to analyze the shopping patterns of various
segments of the population.
Any given responsive household is called once a quarter for four successive quarters. Each time, the interviewer administers the survey to the original respondent, if
possible. During each interview, the respondent is asked
whether the household had expenditures for a set of POPS
categories over a duration of time called a “recall period.” Recall periods are POPS-category-specific and vary
from 1 week to 5 years. The recall period for a specific
POPS category is defined to produce a sufficient, but not
excessive, number of outlets for sampling purposes. For
instance, because households tend to purchase gasoline
frequently, a 1-week recall period is used. In contrast,
people tend to purchase cars and funeral services infrequently; therefore a 5-year recall period is assigned. If
the respondent reports expenditures for a particular POPS
category, the interviewer prompts the respondent for the
outlet name, location, and amount spent. At the end of
each quarter of interviewing, the Census Bureau sends
the TPOPS outlet frame data to BLS for processing. BLS
13

processes TPOPS data two quarters at a time. The primary objective of BLS processing of TPOPS data is to select a sample of outlets at which specific items ultimately
will be priced for inclusion in the CPI. The expenditure
amounts reported in TPOPS are used as outlet selection
probabilities.
TPOPS employs a quarterly rotating-panel sample design. On a quarterly basis, every PSU is assigned 1 of
16 TPOPS questionnaires. Each questionnaire consists
of up to 16 POPS categories. In a particular quarter and
for a particular PSU, the selected TPOPS respondents
are asked about expenditures made for some or all of
the POPS categories on the assigned questionnaire. During each subsequent quarter of TPOPS interviewing, the
given PSU is administered a different questionnaire until
each of the 16 questionnaires has been administered. It
takes 4 years of quarterly interviewing to rotate through
all 16 questionnaires. After the 4 years, the cycle for the
PSU starts over again. In this manner, all TPOPS samples
are refreshed once every 4 years. This practice is repeated
for each PSU. The quarterly pattern of assigned POPS categories varies from PSU to PSU in a strategic fashion to
ensure that every POPS group is assigned to at least a few
PSUs every quarter.

Table 1. Construction of replicate panels
Number
of
PSUs

Number of
replicate
panels

1. New York City

1

4

2. New York City suburbs

2

4

3. Los Angeles City

1

4

4. Los Angeles suburbs

1

2

5. Chicago

1

4

6. Philadelphia and San Francisco

2

4 (2 each)

7. Detroit and Boston

2

4 (2 each)

8. Other large self-representing PSUs

7

14 (2 each)

9. Small self-representing PSUs

12

24 (2 each)

10. Medium-sized PSUs, Census Region 1

8

8

11. Medium-sized PSUs, Census Region 2

10

10

12. Medium-sized PSUs, Census Region 3

22

22

13. Medium-sized PSUs, Census Region 4

6

6

14. Small PSUs, Census Regions 1–4

10

10

15. Anchorage and Honolulu

2

4 (2 each)

PSU group

Outlet sampling procedures. The design for TPOPS provides for the rotation of approximately one-quarter of the
items in each sample PSU during the course of each year.
With each rotation, item samples and outlet samples are
selected for the designated items and corresponding POPS
categories.
In self-representing PSUs, sample households for
each TPOPS rotation are divided into two or more independent groups. This process defines two or more
frames of outlets per category-PSU for outlet selection. The principal purpose of constructing these independent groups, or replicate panels, is for variance
estimation. A single subset of independently selected ELIs and outlets for all item strata within a PSU is
called a replicate. The number of replicates per PSU
group and the number of PSUs in each PSU group are
given in table 1.
Reported expenditures for each outlet within the frame
for each POPS category and PSU-replicate are edited
prior to sample selection. Sometimes, a purchase is reported for an outlet but the amount of expenditures is not
reported; to ensure a chance of selection for the outlet in
those cases, the mean expenditure for outlets for the POPS
category-PSU-replicate is assigned to the outlet. Large
expenditure totals for an outlet are edited, in some cases,
to be no greater than 25 percent of the total expenditure
reported for the POPS category-PSU-replicate. In cases in
which there are more than 20 outlets reported for a POPS
category-PSU-replicate, the largest reported expenditures
are trimmed to be no greater than 10 percent of the total
reported for that POPS category-PSU-replicate.

Source: U.S. Bureau of Labor Statistics.

Outlet samples are selected independently for each PSU,
replicate, and POPS category using a systematic PPS sampling procedure. Each outlet in a frame has a probability of
selection proportional to the total expenditures reported for
the outlet in the POPS category in the TPOPS survey. In each
PSU-replicate, all ELIs selected in the item sampling process
are assigned for pricing to each sample outlet selected from
the frame for the corresponding POPS categories. When multiple selections of a sample outlet occur, a commensurate increase is made in the number of quotes priced for the outlet.
Outlet sampling procedures for commodities and services
not included in the TPOPS. Some commodity and service
items are excluded from the TPOPS, either because existing
sampling frames are adequate or because it became apparent
that the TPOPS would not yield an adequate sampling frame.
(See appendix 8.) For each of these items (non-POPS), BLS
either constructs the sampling frame or acquires it from another source. Each non-POPS item has its own sample design.
The frames consist of all outlets providing the commodity or
service in each sample area. A measure of size is associated
with each outlet on the sampling frame. Ideally, this measure
is the amount of revenue generated by the outlet from the item
for the CPI-U population in the sample area. Whenever revenue is not available, an alternative measure of size, such as
employment, number of customers, or sales volume, is substituted. All samples are selected using systematic sampling techniques with probability proportional to the measure of size.

14

The source of the sampling frame, the definition of the
sampling unit, the measure of size employed, and the final
pricing unit for each non-POPS item are presented in appendix 8.
Merging item and outlet samples. Item and outlet samples,
which are selected independently, must be merged before data
collection. A concordance that maps ELIs to POPS categories
allows each sampled ELI to be assigned for price collection
to the outlet sample selected for the POPS category that contains it. The number of price quotes for an ELI in each outlet thus equals the number of times the ELI was selected for
pricing in the PSU- replicate during the item sampling process. The item-outlet sample merge determines the number
of price quotes assigned for collection in each sample outlet.
In the outlet sampling process, outlets with large expenditure
reports may be selected more than once from the frame for
a given POPS category. An outlet also may be selected from
the frame for more than one POPS category. If an outlet is
selected multiple times for a given POPS category, the same
multiple of price quotes is assigned for collection for each
sample ELI matching the category. If an outlet is selected for
more than one POPS category, price quotes are assigned for
collection for all ELIs selected in each category.
Selection procedures within outlets. A BLS economic assistant visits each selected outlet. For each ELI assigned to
the outlet for price collection, the economic assistant uses a
multistage probability selection technique to select a specific
item from among all the items the outlet sells that fall within
the ELI definition. The economic assistant first identifies all
of the items included in the ELI definition and offered for sale
by the outlet. When there are a large number of items in the
ELI, the assistant groups them by common characteristics,
such as brand, size, or type of packaging. With the assistance
of the respondent for the outlet, the economic assistant assigns probabilities of selection to each group.
The probabilities of selection are proportional to the sales
of the items included in each group. The economic assistant
may use any of the following four procedures, listed in order
of preference, for determining the proportion of sales:
•
•

•
•

Obtaining the proportions directly from a respondent
Ranking the groups by importance of sales as indicated by the respondent, and then obtaining the proportions directly or using assigned proportions
Using shelf space to estimate the proportions, where
applicable
Using equal probability

After assigning probabilities of selection, the economic assistant uses a random-number table to select one group. The
economic assistant then identifies all items included in the
selected group, forms groups of those items based on the incommon characteristics, assigns probabilities to each group,
and uses a random number table to select one. The economic
assistant repeats this process through successive stages until

reaching a unique item. The economic assistant describes the
selected unique item on a checklist for the ELI. Checklists
contain the descriptive characteristics necessary to identify
the item among all items defined within the ELI.
These selection procedures ensure that there is an objective and efficient probability sampling of CPI items other
than shelter. They also allow broad definitions of ELIs, so
that the same unique item need not be priced everywhere.
The wide variety of specific items greatly reduces the withinitem component of variance, reduces the correlation of price
movement between areas, and allows a substantial reduction
in the number of quotes required to achieve a given variance.
Another important benefit from the broader ELIs is a significantly higher likelihood of finding a priceable item within the
definition of the ELI in the sample outlet.
This selection process is completed during the visit to the
outlet to obtain the price for the selected item. Subsequently,
personal visits or telephone calls are made, either monthly or
bimonthly, to ascertain that the item is still sold and to obtain
its current price.
Computer-assisted data collection for commodities and services. A computer-assisted data collection (CADC) system
has been used in the C&S survey since September 2002. The
data collection instrument is composed of two main modules.
The interactive electronic checklists of item specifications
allow the data collector to identify the same item upon returning to an outlet, to substitute a similar item, or to initiate
a new item for pricing. Each ELI is subdivided into clusters,
with each cluster having its own set of specifications. Checklists contain descriptive information about items, including
features of the items themselves and components of the item
that might affect the price. A checklist can be a straightforward list of specifications, or it can be fairly complex, with
hierarchical dependencies among specifications and complicated mathematical formulas. The interactive electronic
checklist enforces rules regarding patterns of specifications
that may be necessary to identify an item. The checklist also
prevents inconsistencies.
The other module of the C&S CADC collection instrument
comprises some screens that make up the pricing form and
various functions pertaining to the task of data collection. For
instance, some screens enable the economic assistant to organize his or her work at the level of the outlet or the quote,
while some allow review of collected data or information
about the outlet and respondent. An economic assistant selects an action, such as substituting a new item for one that is
unavailable (only options that are appropriate for that action
are offered in the collection instrument). The pricing screens
allow the economic assistant to enter the price of the item as
well as relevant information about it, such as quantity, size,
unit of size, sales tax, and seasonality; the economic assistant
can also see the previous price and other data relevant to the
quote.
Electronic data collection improves data quality in part by
activating important rules at the moment the data are being
collected. For instance, a suspiciously large price change can

15

ule of the C&S CADC collection instrument comprises some screens that make up the
d various functions pertaining to the task of data collection. For instance, some screens
omic assistant to organize his or her work at the level of the outlet or the quote, while
iew of collected data or information about the outlet and respondent. An economic
be as
noted
immediately,
than
have been(only
sent options
equivalent
an action, such
substituting
a new rather
item for
oneafter
that data
is unavailable
that rent value within the segment, IRs. This gives segDC
and examined
by commodity
ments
for that actiontoareWashington,
offered in the
collection
instrument).
The pricinganalysts.
screens allow
thewith higher valued units (higher rent levels) a higher
The
collection
instrument
multiple
edits which
ant to enter the
price
of the item
as wellcontains
as relevant
information
abouteither
it, such probability
as quantity, of selection and a lower segment weight:
e, sales tax, and
seasonality;
the economic
can also
see theaspect
previous
warn
the economic
assistantsassistant
about some
important
of price and
ant to the quote.
a quote or prevent them from entering invalid information.
TCs= RCs+ OCs = R s × RR s+Os × IR s.
CPI data collection is scheduled in terms of business days
(that is,data
weekdays
2004,
data
Thethe
number of owned housing units, the number of rented
collection improves
qualityexcluding
in part by holidays).
activating Before
important
rules
at colthe moment
lection covered
three pricing
each can
comprising
busi- housing
units, and the average rent value were taken from
ollected. For instance,
a suspiciously
large periods,
price change
be noted6 immediately,
rather
have been sentness
to Washington,
DC
and examined
by commodity
analysts.
The the
collection
days in most
months
and 5 days
in November
and Dedecennial census. The estimated average owner equivains multiple edits
which
either warn the
the last
economic
assistants
about some
important
cember.
Consequently,
scheduled
data collection
was
alent rent value was determined by a linear regression on
e or prevent them
from
invalid information.
usually
theentering
18th business
day of the month. Beginning with Consumer Expenditure Survey property value, income, and
data for January 2004, the three pricing periods now are of number of rooms. The resulting regression coefficients were
variable
length
and end days
on the(that
last is,
business
day excluding
of the month.
applied
to decennial census values for the same independent
tion is scheduled
in terms
of business
weekdays
holidays).
Before
ection covered three pricing periods, each comprising 6 business days in most months
andto estimate the average owner’s equivalent rent for
variables
mber and December.
Consequently, the last scheduled data collection was usually
18th
eachthe
segment.
Shelter
the month. Beginning
with
data
for
January
2004,
the
three
pricing
periods
now
are
of
The following is the nonlinear regression that was used:
The CPI housing unit sample is the source of the data on
and end on theresidential
last business
day
of the
month. changes in rents for the rent
rents
used
to calculate
of primary residence (rent) index. The housing survey also
oerval = bo + (b1 × propval) + (b2 × propval2) +
uses these rent data in calculating changes in the rental value
(b3 × income) + (b4 × rooms).
of owned homes for the owners’ equivalent rent of primary
In this equation,
g unit sample residence
is the source
of the
dataThese
on residential
rents
used to
calculate
(OER)
index.
two shelter
indexes
account
for changes in

oerval = the value the home would rent for,
nt of primary residence
(rent) 30
index.
The of
housing
survey
also uses these rent data in
approximately
percent
the total
CPI weight.
nges in the rental value of owned homes for the owners’ equivalent rent of primary

propval = the market value of the home,
R) index. TheseWeighting
two shelterduring
indexessegment
account sample
for approximately
of the total CPI
selection. 30
In percent
the 1999

income = the income of the consumer unit, and
Housing Sample, segments were selected with probability
proportional to size, the size measure being estimated ex
rooms = the number of rooms in the house.
penditures.
In the
selection Sample,
process,segments
the segments
ng segment sample
selection.
In segment
the 1999 Housing
were selected with
arethe
ordered
within each
by county
and then by
portional to size,
size measure
beingPSU
estimated
expenditures.
In segment
the segment selection
The actual regression coefficients were determined
ments are ordered
withinwithin
each PSU
by county
andthe
thensegment
by segment
rent level
rent level
county.
Because
selection
is within
uniquely for each index area.
e the segment systematic,
selection is this
systematic,
this guarantees
not all or
high-rent
guarantees
that not allthat
high-rent
low-rentor low-rent
hosen and that segments
the segments
will be and
geographically
distributed
within
the PSU. Because rents are not volatile, the housing sample is
are chosen
that the segments
will be
geographically distributed within the PSU.
divided into panels; one panel is priced each month and
segment, of
s, selection,
was assigned
a probability
selection,
s, was assignedEach
a probability
P, within
the PSU,ofwhich
is the P,
ratioeach
of thepanel
cost is priced twice a year. For example, panel 1 is
the
PSU,
which
is
the
ratio
of
the
cost
of
housing
in
the
relative
to
the
cost
of
housing
in
the
PSU
times
the
number
of
segments
e segment TCwithin
priced
in January and July, panel 2 in February and Aus
segment relative to the cost of housing in the PSU times the gust, and so on through panel 6. The segments within the
fore,
number of segments selected. Therefore,
strata are assigned to these panels. These assignments are
made such that each panel has a representative subsample


Ps   TCs /
TCs   nPSU ,
of the PSU. Because each panel is representative of the ens∈PSU


tire sample, there is never an off-cycle month for the housing survey, a panel of data provides sufficient information
where
for monthly publication of the rent and rental equivalence
(REQ) indexes. Segments were selected within the PSUs
nPSU = number of segments chosen in the PSU,
in multiples of 6, so that each panel had the same sample
size within a PSU.
and
About17
10,000 segments were selected in the PSUs. The
TCs is as defined in the next paragraph.
housing sample is designed to consist of approximately
50,000 rental units. Sampling rates were computed for
Each segment also has a weight Ws , which is the reciprocal of each segment so that the sample design would be realized
the probability of selection. Therefore,
after the sampling and screening processes described next
Ws = 1/Ps .
were completed.



Sample allocation to PSUs. BLS allocated the sample to
PSUs based on the estimated total housing expenditure in
each PSU. The estimated total housing PSU expenditure
is the sum of the total cost of housing, previously defined,
across all segments:

The total cost of housing in the segment, TC is the cost of
rented housing in the segment, RCs, plus the cost of owned
housing in the segment, OCs. RCs is the number of rented
housing units in the segment Rs times the average rent value
within the segment (RR s). (OCs) is the number of owned housing units in the segment times an estimated average owner
16

allocation to PSUs. BLS allocated the sample to PSUs based on the estimated total housing
ture in each PSU. The estimated total housing PSU expenditure is the sum of the total cost of
, previously defined, across all segments:

field staff obtains answers to various (screening) questions
(through observation and through direct questioning of elis∈S
gible respondents) that determine whether an address is in
There are six collection panels. It was desired that the scope for the housing sample. The screening criteria consist
re six collection
panels.sample
It was desired
thewithin
segment
sample
size bepanel.
equal within
each(whether the unit is renter or owner occupied) and
segment
size be that
equal
each
collection
of tenure
on panel. Thus,Thus,
the segments
were
allocated
in
blocks
of
6
segments,
with
a
minimum
of 72 such as not being in public housing projects,
the segments were allocated in blocks of 6 segments, other criteria,
ts per PSU. For
PSUs
with
multiple
replicates,
it
was
desired
to
have
at
least
36
segments
per residence, and the tenant not being a relawith a minimum of 72 segments per PSU. For PSUs with mul- being a primary
e and an equaltiple
sample
size
in
each
replicate.
It
was
determined
that
a
minimum
of
108
segments
replicates, it was desired to have at least 36 segments tive of the landlord. With the computer, the skip patterns can
ded to supportper
publication
areas
were
published
semiannually
and that
of 180Because the computer has stored all of the
replicate in
and
an that
equal
sample
size in
each replicate.
It abeminimum
very efficient.
ts was needed was
for areas
that
are
published
bimonthly.
The
one
exception
was
Baltimore,
which
determined that a minimum of 108 segments was needed previously collected data, automated logic checks remove all
d 108 segments but is published bimonthly as part of the Washington–Baltimore CMSA. As the
to support publication in areas that were published semian- redundant question patterns, thereby reducing the field staff’s
size was previously about 10,000 segments and the budget for housing data collection was
nually and that a minimum of 180 segments was needed for work and the respondent’s burden. Automated data checking
able, multiples of 6 segments were chosen so that the total would be near 10,000 segments.
areas that are published bimonthly. The one exception was ensures that only correct data types are collected, other auBaltimore, which received 108 segments but is published bi- tomated logic checks ensure that collected data are consisng housing units.
Afterassegments
have
been chosen for each CMSA.
PSU, housing
areand
chosen
for
monthly
part of the
Washington–Baltimore
As theunits
tent,
the instrument
informs the field staff if any required
on within each segment. Lists of housing units are obtained for each segment and an equal
sample size was previously about 10,000 segments and the data have not been collected. These data checks are being
lity sample is chosen. In most cases, the target number of rental units from each segment is five.
budget for housing data collection was comparable, multiples performed at the time of collection, so errors and inconsissampling, the housing units are ordered by address and the sample taken is systematic, ensuring
of 6 segments were chosen so that the total would be near tencies can be corrected while the respondent is present. The
aphic spread of housing units selected within the segment. The sampling rate varies from segment
is that the data that are sent to Washington are as acent, depending10,000
on thesegments.
expected percentage of rental and owned units within theresult
segment.
curate as possible. (The collection instrument also automatiSampling housing units. After segments have been chosen cally determines appropriate “scope status”: permanently out
on. Collectionfor
includes
the screening
of theare
selected
units towithin
determine if the units are in
each PSU,
housing units
chosenhousing
for collection
of scope, temporarily out of scope, incomplete, or complete
or the housing each
sample.
If the unit
in scope,
it is
initiated.
Initiationforis each
the initial collection of
segment.
Listsis of
housing
units
are obtained
and in scope).
a, which consists
of theand
rentan
paid
and probability
the specificsample
housingisservices
are associated with the
segment
equal
chosen.that
In most
If the housing unit is found to be out of scope for some
d the rent paid.cases,
Thesethe
datatarget
are the
basis for
calculations
of rent
that occurs during the
number
of all
rental
units from
eachchange
segment
reason
that
is not likely to change, the collection instrument
he unit in the housing
sample.
initiation,
the housing
priced on
every
6months.
is five. Prior
to After
sampling,
the housing
unitsunit
areisordered
by panel
assigns
a
scope
is very similaraddress
to the initiation
process,taken
but some
previous answers
provided. The collectionstatus of “permanently out of scope” and the
and the sample
is systematic,
ensuringare
a geounit
is never
ousing data, and particularly the rent data, is independent. That is, the field staff
collects
thevisited
data again. (An example of this would be units
graphic spread of housing units selected within the segment. in public housing projects.) If the housing unit is found to be
giving the respondent the previous answer. Previous answers for some non-rent data are
The sampling rate varies from segment to segment, depend- out of scope for some reason that might change, its status is
d, so that the field staff can confirm certain changes with the respondent. Inherent
in all of the
ing on the expected percentage of rental and owned units “temporarily out of scope,” and another screening/initiation
ed housing questionnaires (screening, initiation, and pricing) are various flow determinations
within
segment.
attempt
tterns), such that
the the
answer
to one question determines the next question that must
be is made after a specified waiting period. (An example
of this might be when the unit is not the primary residence
nswered.
Collection. Collection includes the screening of the selected
for the current tenant, but may become the primary residence
housing units to determine if the units are in scope for the
for some
future tenant.) If the screening was incomplete, the
DC instrument.
The CADC
instrument
the screening/initiation
schedules
electronically.
housing
sample.
If the unitreceives
is in scope,
it is initiated. Initiahousing
unit is
ough the schedules
have
been
assigned
to
specific
panels,
the
field
staff
has
several
months
to assigned a scope status that results in another
tion is the initial collection of rent data, which consists of the
screening/initiation
attempt in 6 months. Selected addresses
he screening/initiation
schedules.
This
is
referred
to
as
the
non-monthly
period.
The
field
staff
rent paid and the specific housing services that are associated
thatquestioning
pass the screening
criteria are considered in scope for
answers to various
(screening)
questions
(through
observation
and
through
direct
of
with the unit and the rent paid. These data are the basis for
the housing
respondents) that determine whether an address is in scope for the housing sample.
The sample and are eligible for the next stage of the
all calculations of rent change that occurs during the life of
process,
initiation.
ng criteria consist of tenure (whether the unit is renter or owner occupied) and other
criteria,
such
the unit in the housing sample. After initiation, the housing
eing in public housing projects, being a primary residence, and the tenant not being a relative of
unit is priced on panel every 6 months. Pricing is very simi- Initiation. The CADC instrument automatically moves the
lord.
lar to the initiation process, but some previous answers are interviewer into the initiation portion of the instrument when
provided. The collection of the housing data, and particularly the instrument has determined that the screening is complete
e computer, the skip patterns can be very efficient. Because the computer has stored
all of the
the rent data, is independent. That is, the field staff collects and the housing unit is in scope. As previously mentioned,
sly collected data, automated logic checks remove all redundant question patterns,
thereby
the data without giving the respondent the previous answer.
collection
instrument handles the skip patterns, the autog the field staff’s work and the respondent’s burden. Automated data checking the
ensures
that only
Previous answers for some non-rent data are provided, so that mated data and consistency checks, the schedule completion
data types are collected, other automated logic checks ensure that collected data are consistent,
the field staff can confirm certain changes with the respon- checks, and the final initiation status. The screening and inident. Inherent in all of the structured housing questionnaires tiation data are then electronically transmitted to the housing
(screening, initiation, and pricing) are various flow determi- database in Washington,
19
DC.
nations (skip patterns), such that the answer to one question
determines the next question that must be asked/answered.
Pricing. During the non-monthly period, the screening/initiation may have occurred off-panel (or not at all). Therefore,
The CADC instrument. The CADC instrument receives the the housing units will have to be priced (or perhaps screened/
screening/initiation schedules electronically. Even though initiated) on panel. There must be two on-panel prices before
the schedules have been assigned to specific panels, the field the unit can be considered usable. The field staff receives,
staff has several months to collect the screening/initiation electronically, the housing units to price from the Washingschedules. This is referred to as the non-monthly period. The ton, DC, database. The CADC collection instrument automatiPSU expenditure 

∑TC .
s

17

cally moves the interviewer into the pricing portion of the instrument and, as mentioned, the collection instrument handles
the skip patterns, the automated data and consistency checks,
the schedule completion checks, and the final schedule status.
The pricing data are electronically transmitted to Washington,
where they are reviewed and corrected as necessary.
These data, along with the initiation or pricing data from 6
months earlier, are used in the housing pricing relative calculation (PRC) described in the section titled “Estimation of price
change for shelter.” Occasionally situations occur during pricing that affect the unit’s scope status and, on a scheduled but
infrequent basis, additional questions are asked to ensure that
the housing units are still in scope for the housing sample. If
changes occur, the units are treated as indicated in the section
titled Initiation, on the basis of their new scope status.

estimated quantities of the items purchased in its sampling
period serving as weights. In January 1999, most of the item
strata converted to the geometric mean index formula, which
is a weighted geometric mean of price ratios (an item’s current price divided by its previous price) with weights equal to
expenditures on the items in their sampling periods. Calculations for a limited number of strata, including the two shelter
strata, continue to use the Laspeyres formula, as shown in the
following list: 26
1. Selected shelter services (rent of primary residence;
owners’ equivalent rent of primary residence; and
housing at school, excluding board)
2. Selected utilities and government charges (electricity; residential water and sewerage maintenance; utility (piped) gas service; State vehicle registration and
driver’s license)

Estimation of price change in the CPI
As stated earlier, the CPI is calculated in two stages. In the
first stage, basic indexes are calculated for each of the 8,018
CPI item–area combinations. For example, the electricity index for the Boston CPI area is a basic index. The weights for
the first stage come from the sampling frame for the category
in the area. Then, at the second stage, aggregate indexes are
produced by averaging across subsets of the 8,018 CPI item–
area combinations. The aggregate indexes are the higher-level indexes; for example, the all-items index for Boston is the
average of all its 211 basic indexes. Similarly, the aggregate
index for electricity is the average of the basic indexes for
electricity in each of the 38 index areas. The U.S. city average
All items CPI is the average of all basic indexes. For the CPIU and CPI-W, the weights for the second stage are the baseperiod expenditures on the item category/areas from the CE.

3. Selected medical care services (physicians’ services; hospital services; dental services; services by
other medical professionals; and nursing homes and
adult day care.)
Since January 1999, most item strata have used an expenditure-share-weighted geometric average a,iRG[t;t-1]. The other
strata use the Laspeyres formula average, a,iRL[t;t-1], which all
strata used prior to 1999. The Laspeyres is a base-period,
quantity-weighted arithmetic average. Every month, the C&S
survey system uses the following formulas to compute price
relatives for each item–area combination (a,i):

a ,i

R

G

[t ;t 1]

=

∏
jε a ,i

Pj ,t

(W

j , POPS

∑Wk , POPS )

kεai

,

Pj ,t 1

Estimation of price change for commodities and
services other than shelter
The C&S survey is the CPI’s primary source of price
change data. Of the 209 C&S item strata, 185 are priced strata. The other 24 C&S strata, all of which have very small
weights, are, for a variety of reasons, unsampled or truncated
from pricing. The price movements of unsampled strata are
imputed from related priced strata.
For most priced C&S strata, the C&S survey is the primary
source, meaning that information on price change comes from
samples that the survey maintains. A few C&S strata, including those for airline fares, intercity train fares, and used vehicles, use secondary sources of data on prices for their samples.

a ,i

R

L

∑(W
=
∑(W
jε a ,i

[ t ;t 1]

jε a ,i

j ,POPS

j ,POPS

/ Pj ,POPS ) Pj ,t
/ Pj ,POPS ) Pj ,t

,
1

In these equations,
RG[t;t-1] and a,iR L[t;t-1], are, respectively, the geometric and
Laspeyres price relatives for area-item combination, a,i, from
the previous period, t – 1, (either 1 month or 2 months ago),
to the current month,
a,i

Pj,t

Pj,t-1
Pj,POPS

and

Price relatives. Each month, the processing of the C&S survey data yields a set of price relatives (a price relative is a
measure of short term price change) for all basic indexes. The
CPI uses an index number formula to obtain an average price
change for the items in each basic index’s sample. Prior to
January 1999, all CPI price relatives used a modified Laspeyres index number formula. This is a ratio of a weighted arithmetic mean of prices in the current period to the same average of the same items’ prices in the previous period, with

is the price of the jth observed item in month t for
area-item combination a,
is the price of the same item in time t – 1,
is an estimate of the item j’s price in the sampling
period when its POPS was conducted,

26
See “Planned change in the Consumer Price Index formula,” http://
stats.bls.gov/cpi/cpigm02.htm.

18

Wj,POPS is item j’s weight in the POPS and is defined in
detail next.
The product in the geometric mean formula and sums in
the Laspeyres formula are taken over all usable quotes in
area-item combination a,i. It is important that the price of
each quote be collected (or estimated) in both months in order to measure price change.
Quote weights. For each individual quote, the weight Wj,POPS
is computed as
Wj,POPS = AEfgη/BN,
where
A is the proportion of the total dollar volume of sales for
the ELI relative to the entire POPS category within
the outlet (called the outlet’s percent of POPS for the
ELI),
E is an estimate of the total daily expenditure for the
POPS category in the PSU by people in the U-population (called the basic weight),
f

is a duplication factor that accounts for any special
subsampling of outlets and quotes,

g is a geographic factor used to account for differences
in the index area’s coverage when the CPI is changing
from an area design based on an old decennial census
to a design based on a more recent census,
η

is the number of planned quotes for collection in the
ELI – PSU, which is also the sum of duplication factors for all quotes in the ELI – PSU;

y is the sum of duplication factors for uninitiated quotes
in an ELI – PSU;
N is 1 + (y/(N – y)), which is essentially the ratio of
planned quotes to quotes with usable prices in both
periods t – 1 and t for the ELI – PSU;
and
B is the proportion of the item stratum’s expenditure accounted for by the ELI in the region.
POPS-period prices. In the Laspeyres formula, the item expenditure weight is divided by an estimate of the item’s price
in the sampling period to convert the expenditure into an estimated quantity. An item’s POPS period occurred sometime
before its outlet’s initiation, so that one cannot observe its
POPS price directly. Instead the price is estimated from the
item’s price at the time the sample was initiated and the best
estimates of price change for the period from the POPS period to the initiation period. The formula is

where

Pj,0 is the price of the jth item at time 0 (when it was initi
ated or chosen for the sample),
IXj,0 is the value of the price index most appropriate for the
jth item in period 0, the time it was initiated,
and
IXj,pops is the value of the same price index in the POPS period
POPS).
Item replacement and quality adjustment
One of the more difficult problems faced in compiling a price
index is the accurate measurement and treatment of quality
change due to changing product specifications and consumption patterns. The concept of the CPI requires a measurement through time of the cost of purchasing an unchanging,
constant-quality set of goods and services. In reality, products
disappear, products are replaced with new versions, and new
products emerge.
When the data collector finds that he or she can no longer
obtain a price for an item in the CPI sample (most commonly
because the outlet permanently stops selling it), the data collector uses the CPI replacement procedure to find a new item.
As explained earlier in the section on CPI item and outlet
samples, each item stratum consists of one or more ELIs. CPI
staff economists, called commodity analysts, in Washington,
DC, have developed checklists that define further subdivisions of each ELI. When seeking a replacement in a retail
outlet, the data collector first uses the checklist for the ELI to
find the item the outlet sells that is “closest” to the previously
priced one. Then the data collector describes the replacement
item on the checklist, capturing its important specifications.
The commodity analyst assigned to the ELI reviews all replacements and selects 1 of 3 methods to adjust for quality
change and to account for the change in item specifications.
The following example describes the most common type
of quality adjustment problem. Assume that in period t a data
collector tries to collect the price for item j in its assigned outlet and is not able to do so because the outlet no longer sells
the item. A price for item j was collected in period t – 1. Following the procedure, the data collector finds a replacement
and collects a price for it. The replacement becomes the new
version (version v + 1) of item j. The decision as to how the
CPI treats the replacement is made by the commodity analyst
assigned to the ELI to which item j belongs. The commodity analyst has the descriptions of the two versions of item j.
In addition, he or she has the t – 1 price, Pvj,t-1, for the earlier
version (version v) and the period t price, Pv+1j,t, of the replacement version v + 1. The following matrix displays the price
information available to the commodity analyst:
Version
Old version
v
Replacement version
v+1

Pj,POPS = Pj,0/[IXj,0/IXj,POPS],

19

Period t-1 price

Period t price

Pvj,t-1

…

…

Pv+1j,t

To use the item in index calculation for period t, we need
an estimate of Pv+1j,t-1, the price of the replacement version, v + 1,
in period t – 1, or we need an estimate of Pvj,t, the earlier version v, in period t. If there is no accepted way of estimating
either Pv+1j,t-1 or Pvj,t, the observation for item j is left out of
index calculation for period t, meaning that the observation
is treated as a nonresponse handled by imputation (described
shortly).

In this method, which is sometimes referred to as “linking,” the item is effectively left out of the calculation for 1
month; the cell relative is computed without the observation.
The price relative (either R L ai,t,t-1 or RG ai,t,t-1) is computed with
one less useable quote.
When there is a new version of the item that is not comparable to the previous version, a price of the new version
(Pv+1j,t) is available. That price is not used in the calculations
for period t but, in the subsequent period Pv+1j,t,is used as the
previous price. If, on the other hand, the reason for the imputation was that the item was temporarily missing (meaning
that no price was collected), a period-t price must be estimated. For this purpose, the cell relative is used to estimate
the period-t price:
Pvj,t = Rai,t,t-1 × Pvj,t-1.

The commodity analyst chooses 1 of 3 methods to handle the
replacement
• Direct comparison
• Direct quality adjustment
• Imputation

Class-mean imputation. The C&S uses class-mean imputation for many noncomparable replacements, primarily in the
item strata for vehicles, for other durables including hightech items, and for apparel. The logic behind the class-mean
procedure is that, for many items, price change is closely associated with the annual or periodic introduction of new lines
or models. For example, at the introduction of new model
year vehicles, there are often price increases while, later in
the model year, price decreases are common. The CPI uses
the quality adjustment method as frequently as possible to
handle item replacements that occur when vehicle product
lines are updated. Class-mean imputation is employed in the
remaining replacement situations. In those cases the CPI estimates price change from the price changes of other observations that are going through item replacement at the same
time and were either quality adjusted directly or were judged
directly comparable. For class-mean imputation, the CPI estimates Pvj,t, which is an estimate of the current (t) price for
the old version (v), and uses this estimated current price in
the calculation of the price relative for period t. The estimated
current-period price is the previous-period (t – 1) price of the
old version times a specially constructed price relative for the
class:
Pv j,t = Pj,t-1 × cR[t;t-1].

Direct comparison. If the new and old items are essentially
the same, the commodity analyst deems them directly comparable, and the price comparison between the items is used
in the index. In this case, it is assumed that no quality difference exists.
Direct quality adjustment. The most explicit method for
dealing with a replacement item is to estimate the value of
the differences. The estimate of this value is called a quality
adjustment amount QAj,t-1. In this case,
Pv+1j,t-1 = Pvj,t-1 + QAj,t-1.
Chief sources of direct quality adjustment information are
manufacturers’ cost data and hedonic regression.
Imputation. Imputation is a procedure for handling missing
information. The CPI uses imputation for a number of cases,
including refusals, inability to collect data for some other reason (the item may be out of season), and the inability to make
a satisfactory estimate of the quality change. Substitute items
that can be neither directly compared nor quality adjusted are
called noncomparable. For noncomparable substitutions, an
estimate of constant-quality price change is made by imputation. There are two imputation methods: Cell-relative imputation and class-mean imputation.

cR[t;t-1] is computed with either the geometric mean or Laspey-

res formula over the subset of observations in the ELI to
which item j belongs. The subset is the class of interest—that
is, all the comparable and quality-adjusted replacement observations in the same ELI and PSU.

Cell-relative imputation. If there is no reason to believe that
price change for an item is different from those for the other
items in its cell or basic index, the cell-relative method is the
appropriate way to impute. This method is used for missing
values because in that case we have no knowledge about the
observation. For noncomparable substitutions, the cell-relative method is prevalent for food and service items. The price
change between the old item and the noncomparable new
item is assumed to be the same as the average price change of
all similar items in 1 month for the same geographic area—
that is, the same as the average price change for the cell for
that ELI and PSU.

Review and treatment of outlier price changes. All outlier
price changes are reviewed by commodity experts. Outlier
price changes, if accurate, are generally included in the calculation of price relatives. Extreme price changes are bounded, as the geometric mean formula performs poorly for zero
and near-zero prices.
Estimation of price change for shelter
The rent and OER indexes measure the change in the cost
of shelter for renters and owners, respectively. Price change

20

price in the calculation of the price relative for period t. The estimated current-period price is the
s-period (t – 1) price of the old version times a specially
constructed
price relative
for rents
the class:
Similarly,
the owners
equivalent
weight ( OWs ) is derived by multiplying the segment weight ( W
by thethe
number
of equivalent
owners in the
segment
the number
of renters sampled
in theweight
segment.
Similarly,
owners
rents
weightdivided
( OWs ) by
is derived
by multiplying
the segment
( Ws )
Pvj,t = Pj,t-1 ×bycRthe
[t;t-1].
Because
survey
collects divided
rents and
implicit
rents ofsampled
owners, in
thethe
ratio
of average
numberthe
ofhousing
owners in
the segment
bynot
thethe
number
of renters
segment.
implicit housing
rent to average
rent in the
segment
is the
alsoimplicit
included in the
owner’s the
equivalent
rent
weight:
survey
rents
andOER
not
owners,
average
data for these two indexes come Because
from thethe
CPI housing
sur- collects
The rent
and
estimators.rents
Theofrent
estimatorratio
usesofthe
implicit
rent
to average
rentsubset
in change
the of
segment
also included
in the
owner’s
equivalent
weight:
is computed with
geometric
mean or Laspeyres
formula
over
the
observations
vey.either
Each the
month,
BLS economic
assistants
gather
informain theis“economic
rent,”
which
is basically
the rent
“conO
IR
LI to which item
belongs.
Theunits
subset
is the
class
of interest—that
all on
the comparable
and
tionj from
renter
on the
rent
for the
current monthis,and
tract rent” adjusted
for
changes
of the houss in the
 s quality
OWany
,
s  Ws 
adjusted replacement
observations
in the same ELI and PSU..
what services
are provided.
ing unit. The OER estimator
the
change
in
the “pure
Os uses
nIR
RR
s
s
Ws cost of sany ,utilities
OWs  the
rent,” which excludes
included
in the
ns RR s
The rent
in the CPI
arechanges
“contract
contract.
and treatmentRent.
of outlier
priceestimates
changes.used
All outlier
price
arerents.”
reviewedrent
by commodity
The
and OER
estimators. The rent estimator uses the change in the “economic rent,” which is
They areifthe
payment
all services
the rent
landlord
provides
Outlier price changes,
accurate,
arefor
generally
included
in the calculation
of price relatives.
basically
the
“contract
adjusted
forchained
any changes
in the quality
of
theindex,
housing
The isOER
The formula
rentif and
OER
estimators.
rent
estimator
the change
in the
rent,”
which
exchange
For example,
theperforms
landlord
pro-rent”
The
6-month
For
the“economic
rent
theunit.
cure price changesinare
bounded,forasthe
the rent.
geometric
mean
poorly
for
zero
and
near- uses estimator.
estimator
uses
the
change
in
the
“pure
rent,”
which
excludes
the
cost
of
any
utilities
included
in the re
basicallyrent.
the “contract
for anyindex
changes
in the by
quality
of thethe
housing
unit.ofThe
vides electricity, it is part of the contract
The CPI rent”
item adjusted
rent month’s
is derived
applying
sixth root
the OER
ces.
contract.
estimator
uses
the
change
in
the
“pure
rent,”
which
excludes
the
cost
of
any
utilities
included
in
the
rent
expenditure weights also include the full contract rent pay- 6-month rent change to the index for the previous month. For
contract.
Thefor
CPIshelter
rents are calculated as the amounts the tenants the OER index, the current month’s index is derived by aption of pricement.
change
The 6-month
Forsixth
the rent
currentOER
month’s
index
is derived
pay their landlords plus any rent reductions
tenantschained
receiveestimator.
plying the
rootindex,
of the the
6-month
change
to the
index by applying
sixth
root
of
the
6-month
rent
change
to
the
index
for
the
previous
month.
For
the
OER
thethe
cur
The 6-month
chainedcalled
estimator.
the rent index,
the current month’s index is derived
by index,
applying
for performing services for the landlord
(sometimes
for For
the previous
month.
t and OER indexes measure the change in the cost
of
shelter
for
renters
and
owners,
respectively.
month’s
index
is
derived
by
applying
the
sixth
root
of
the
6-month
OER
change
to
the
index
for
the
sixth root
ofto
thethe
6-month
rent change
to
theestimator
index foruses
the previous
month.
the OERrent.”
index, the current
“rent as
pluscome
any subsidy
paid
landlord.
The
rent
the change
in theFor
“economic
ange data for these
twopay”)
indexes
from thepayment
CPI
housing
survey.
Eachby
month,
BLSthe
economic
previous
month.
month’s
index
is derived
applying
sixth
root structure
of the 6-month
OER
change to
the index
Reductions
for
any
other
reasons
are
not
considered
part
of
Because
of
the
panel
used
in
the
housing
sample,
the for the
ts gather information from renter units on the rent
for the month.
current month and on what services are
previous
the rent.
6-month change in rent is based on sampled, renter-occupied
d.
The rent estimator uses theunits
change
the “economic
rent.”rent
Because
of the
panel
thatinhave
usable 6-month
changes.
The
sumstructure
of the used in the
housing
sample,
the
6-month
change
in
rent
is
based
on
sampled,
renter-occupied
thatinhave
Owners’ equivalent rent (OER). The OER
approachuses
to mearent estimator
the change
in the
rent.”for
Because
of theunit
panelwithin
structure
used
the usab
current
(t) “economic
economic rents
each usable
aunits
seghe rent estimates
used
in
the
CPI
are
"contract
rents."
They
are
the
payment
for
all
services
the
6-month
rent
changes.
The
sum
of
the
current
(t)
economic
rents
for
each
usable
unit
within
a
segmen
housing
sample,
the 6-month
in rent is based
sampled,
renter-occupied
unitsisthat
suring price change for owner-occupied
housing
started
in the change
ment, itweighted
by theon
renter
weight
for that segment,
di-have usable
provides in exchange for the rent. For example,6-month
if the
landlord
provides
electricity,
is
part
of(t)
weighted
by
the
renter
weight
for
that
segment,
is
divided
by
the
sum
of
the
weighted
economic
rents
rent
changes.
The
sum
of
the
current
economic
rents
for
each
usable
unit
within
a
segment,
CPI-U in January 1983 and the CPI-W in January 1985. The vided by the sum of the weighted economic rents six months
ract rent. The CPI item expenditure weights alsoweighted
include
the
full
contract
rent
payment.
The CPI
months
earlier
(t
–
6).
This
ratio
is
used
to
represent
the
6-month
change
in
rent
for
all
renter-occupie
by
the
renter
weight
for
that
segment,
is
divided
by
the
sum
of
the
weighted
economic
rents
six
OER index is designed to measure the change in the rental earlier (t – 6). This ratio is used to represent the 6-month
e calculated as the amounts the tenants pay their months
landlords
plus
any
rentThis
reductions
tenants
units
in
the
segment.
earlier
(t
–
6).
ratio
is
used
to
represent
the
6-month
change
in
rent
for
all
renter-occupied
value of owner-occupied housing. In essence, OER measures change in rent for all renter-occupied units in the segment.
for performing services for the landlord (sometimes
as pay") plus any subsidy
unitscalled
inwould
the"rent
segment.
the change
in the amount
homeowner
pay
to rent, orpart of In
parallel calculation, the sum of the current (t) pure
t paid to the landlord.
Reductions
for anya other
reasons
are
not
considered
thearent.
In a parallel
calculation,
the sum
of the current (t) pure rents for sampled, renter-occupied units within
would earn from renting, his or her home
in a competitive
rents for sampled, renter-occupied units within a segment,
segment,
weighted
by
owner
weights,
is divided
by the
sum
of the weighted pure rents
months
In ainparallel
calculation,
thethe
sum
of the
current
pureweights,
rents
foris
sampled,
units six
within
a
market. It is a measure of the change
the price
of the shelter
weighted
by the (t)
owner
divided renter-occupied
by the sum of the
’ equivalent rent (OER). The OER approach to segment,
measuring
price
change
for
owner-occupied
earlier
(t
–
6).
This
ratio
is
used
to
represent
the
6-month
change
in
the
OER
index
for
all
ownerweighted
by
the
owner
weights,
is
divided
by
the
sum
of
the
weighted
pure
rents
six
months
service provided by owner-occupied housing.
weighted
rents six months earlier (t – 6). This ratio is
started in the CPI-U in January 1983 and the CPI-W
in(tJanuary
1985.
The
OER
ispurethe
occupied
in the
earlier
– 6). units
This
ratio
issegment.
used
toindex
represent
6-month
change
in theinOER
for all for
ownerused
to
represent
the 6-month
change
the index
OER index
d to measure the
change
in the rental
value
of owner-occupied
Indoessence, OER
occupied
insystems
the segment.
PRC
for Housing.
The
housing
and the units
C&Shousing.
allrenting,
owner-occupied
units in the segment.
s the change in the amount a homeowner would payThe
to rent,
or would
earn
from
or
functions
of the
PRC
have been his
designed
to make use of the parallel rent and OER computations
not directly calculate indexes. Instead, they
produce
price
The
functions
of
the PRC have been designed to make use
e in a competitive market. It is a measure of the change
in thethe
price
ofaggregates
the shelter the
service
general,
PRC
weighted to
rents
for theof
units (i)
in therent
Index
Area
for the current
relatives, and the index estimationThe
system
then uses
thePRC
price
functions
of
the
have
been
designed
make
parallel
OER(a)computations.
In
of
the
parallel
rent
and use
OER the
computations.
Inand
general,
the
d by owner-occupied housing.
ER
)
and
period
(t)
andaggregates
for ratios
6 months
previous
(t
–
6).
When
the
PRC
is
run
for
rent,
economic
rents
(
relatives for basic index calculation.
Price
relatives
are
general,
the
PRC
the
weighted
rents
for
the
units
(i)
in
the
Index
Area
(a)
for
the
current
i
PRC aggregates the weighted rents for the units (i) in the Inof price change from the previousperiod
month
(t –and
1) to
current
ER
)
and
(t)
forthe
6 (months
previous
(t
–
6).
When
the
PRC
is
run
for
rent,
economic
rents
(
RW
renter
weights
)
are
used:
dex
Area
(a)
for
the
current
period
(t)
and
for
6
months
prei
s
r Housing. The
housing
and basic
the C&S
systems
do not
directlythe
calculate
indexes.
Instead, they
month
(t), and
index
calculation
updates
last month’s
vious
– 6). When the PRC is run for rent, economic rents
weights
( RWrelatives
price relatives, and the index estimation systemrenter
then uses
the price
for
basic(t index
s ) are used:
indexes (t – 1) to the current month (t).
(ER
)
and
renter
weights (RWs ) are used:
ion. Price relatives are ratios of price change from the previous month (t – 1) to thei current
month
∑RWs  ER i ,t
basic index calculation
updates
month’s
(t – 1)begins
to the with
current
i∈a
Weighting
duringthe
thelast
PRC.
Eachindexes
calculation
a month (t).
.

RELRENT
RW
t -6 ,t ,a∑
s  ER i ,t
RWs  ER i ,t-6
segment weight (Ws ) based on the probability of selecting the
RENT
i∈a ∑
.
REL t-6,t ,a 
i∈a
ng during thesegment.
PRC. Each
beginstitled
with a“Weighting
segment weight
(W
(Seecalculation
earlier section
during
segs ) based on the
∑RW
s  ER i ,t -6
i∈a
ment
sample selection.”)
derivetitled
the “Weighting
renter weight
in the
ity of selecting
the segment.
(See earlierTo
section
during
segment sample
), the
weight
(Ws s),) the
is multiplied
by the ( Ws ) is multiplied by
segment weight
n.”) To derive segment
the renter(RW
weight
in segment
the segment
( RW
s
When the PRC is run for OER, pure rents (PR i ) and owner
of renters
in thebysegment,
divided
bythethe
number
ber or renters number
in the segment,
divided
the number
of renters
sampled
inoffor
the OER,
segment:
owner
When
PRC
is run
pure
rents
( PR
i ) and
weights
(OW
) are
used.
That
is, weights ( OWs ) are used. That is,
s
renters sampled in the segment:
When the PRC is run for OER, pure rents ( PR i ) and owner weights ( OWs ) are used. That is,

Rs
.
ns

∑
∑
∑
∑

OWs  PR i ,t
i∈A
.

RELOER
 PR i ,t
OW
t -6 ,t ,a
s
OWs  PR i ,t -6
OER
i∈A
.
RELt -6 ,t ,a 
i∈A
OW
Similarly, the owners' equivalent rents weight (OWs ) is des  PR i ,t -6
The index estimation
system needs a 1-month price relai∈A
rived by multiplying the segment weight (Ws ) by the number
tive,
not
a
6-month
price
relative;
the 6th
rootrelative;
of the therefore, th
The
index
estimation
system
needs
a
1-month
price
relative,therefore,
not a 6-month
price
of owners in the segment divided by the number of renters
REL
is
derived:
is
derived:
6th
root
of
the
REL
25
t
6
,
t
,
a
index estimation
system needs ta6,t,a
1-month price relative, not a 6-month price relative; therefore, the
sampled in the segment. Because The
the housing
survey collects
6th root the
of the
rents and not the implicit rents of owners,
ratioREL
of average
t 6 ,t ,a is derived:
implicit rent to average rent in the segment is also included in
RELt-1,t ,a  6 RELt-6,t ,a
the owner’s equivalent rent weight:

RWs  Ws 

26

O IR s then passed to the index estimation system for basic index computation for the rent and OER item
OWs = Ws × s × and
. 
and then passed to the index estimation system for basic instrata.
ns RR
s
dex computation for the rent and OER item strata.

Vacancy imputation. Vacant units that were previously occupied by renters are used in the calculation
Rt,t-1 and Rt,t-6. The vacancy imputation process incorporates several assumptions about the unobserved
rents of vacant units. It is assumed that rents tend to change at a different rate for units that become
21
vacant (and are, therefore, in the process of changing tenants) than for other units. The vacancy
imputation model assumes that, after an initial lease period, expected rents change at a steady rate unti

Vacancy imputation. Vacant units that were previously
occupied by renters are used in the calculation of Rt,t-1 and
Rt,t-6. The vacancy imputation process incorporates several
assumptions about the unobserved rents of vacant units. It
is assumed that rents tend to change at a different rate for
units that become vacant (and are, therefore, in the process
of changing tenants) than for other units. The vacancy imputation model assumes that, after an initial lease period, expected rents change at a steady rate until the old tenant moves
out of the unit. When there is a change in occupant or a unit
becomes vacant, the rent is assumed to jump at some rate,
referred to as the “jump rate.” In markets with generally rising rents, this jump rate is usually greater than the average
rate of change for occupied units. BLS estimates the jump
rate based on non-vacant sample units in the PSU that have
had a change in tenant between t – 6 and t. Rent changes for
nonvacant units without a tenant change are used to calculate
the average continuous rate of change. These values are used
to impute rents for vacant units for period t from their rent in
t – 6.27 The imputed rent, ri,t of the ith vacant rental unit in t is

or

teristics. 28The aging adjustment procedure was introduced
into the CPI in 1988.
Special pricing and estimation procedures for
medical care
Although third parties (mainly government agencies and
employers) pay much of the cost of medical care on behalf
of consumers, the medical care component of the CPI covers
only that part of healthcare commodities, services and health
insurance premiums that consumers pay for “out of pocket.”29
Medical insurance premiums constitute the largest part of
consumers’ out-of-pocket spending for medical care. Unlike
other forms of consumer insurance in the CPI, the data needed from insurers to hold the quality of the insurance policies
constant are so extensive and so closely held that BLS has not
been able to construct a constant-quality health insurance index. Consequently, the CPI has employed an indirect method
for pricing health insurance. In short, the CPI allocates most
of consumers’ out-of-pocket expenditures on health insurance premiums to the weights for other healthcare services
and commodities, placing the small remainder, which covers
the insurance companies’ costs and their profits, into a separate stratum.
Use of the indirect method for pricing health insurance has
two important effects on the CPI. First, the relative shares of
the weights for most of the other CPI medical care item strata
are increased, because they include their portions of the reallocated consumer expenditure for health insurance premiums.30 Second, the CPI approach to measuring price change
for medical care items reflects the fact that these items are,
for the most part, paid for by insurance companies and, therefore, the approach must take account of insurance arrangements such as type of reimbursement method.

ri,t = ri,t-1 J if the unit was not vacant in t – 6
ri,t = ri,t-1 C6 if the unit was vacant in t – 6,

where J is the 6-month jump rate calculated for the PSU, and
C is the 1-month steady rate of change.
The imputation of vacant rents ensures that the unobserved rent change that occurs when a unit becomes vacant
is reflected in the final rent index. The 6-month rent-change
estimates capture these changes once the units become occupied.
Non-interview imputations. Units that were previously responding, not currently responding, and are not vacant are
also imputed and used in the calculation of R t,t-1 and R t,t-6. All
units within a PSU are broken up into high, medium, and low
rent categories based on their rent level in t – 6. The rents of
nonresponding, nonvacant units are imputed forward into t
by using the average rent change of other housing units in
their respective category.

Medical care items and their prices. The movement of CPI
medical care indexes is based on the average change in the
prices of a sample of items selected to represent them. The
items are, for example, a prescription for a specific medicine
or a visit of a specified duration to a doctor or a hospital.
These are inputs to medical treatments addressing a specific

Aging adjustment. The aging adjustment accounts for the
small loss in quality as housing units age (or depreciate)
between interviews. The aging adjustment factors can be
thought of as 1/(1 – d) where d is the monthly rate of physical depreciation. BLS computes factors for each housing
unit with regression-based formulas. The formulas account
for the age of the unit and a number of structural charac-

28
For further information, see Walter F. Lane, William C. Randolph, and
Stephen A. Berenson, “Adjusting the CPI shelter index to compensate for
effect of depreciation,” Monthly Labor Review, October 1988, pp. 34–37.
29
As a consequence, the medical care portion of the CPI is much smaller
than its portion of the national accounts.
30
See U.S. Bureau of Labor Statistics, “Direct pricing of health insurance in the Consumer Price Index.” Paper presented at the Sixth Meeting
of the International Working Group on Price Indices, Canberra, Australia, April 2001, http://www.ottawagroup.org/Ottawa/ottawagroup.nsf/
home/Meeting+6/$file/2001%206th%20Meeting%20-%20U.S.%20
Bureau%20of%20Labor%20Statistics%20 -%20Direct%20Pricing%20of%20Health%20Insurance%20in%20the%20Consumer%20
Price%20Index.pdf/. In that study, the share of the CPI weight for health
insurance under direct health insurance pricing was 2.7 percent, compared
with 0.3 percent under indirect insurance pricing. At the same time, the
shares of the other medical care items were commensurately smaller (for
example, for hospitals, 0.4 percent versus 1.4 percent).

27

For more information on vacancy imputation, see J.P. Sommers and J.D.
Rivers, “Vacancy imputation methodology for rents in the CPI,” Proceedings of the American Statistical Association, Business and Economic Statistics Section (Alexandria, VA: American Statistical Association, 1983).

22

medical condition.31 The CPI data collectors, following CPI
sampling procedures, select the sample items by working
with respondents in pharmacies, doctors’ offices, hospitals,
and other outlets that provide medical care.
The CPI defines the transaction price for medical care
items as all payments or expected payments received from
eligible payers, including both the patient and appropriate insurers. In most cases, the field staff is able to collect transaction prices; if the respondent is unable or unwilling to provide
transaction prices, then cash or self-pay prices are collected,
except in the case of hospitals where “list prices,” or so called
charge master prices, are normally not collected unless associated with a self-pay patient.
CPI medical care indexes. The CPI medical care aggregate
index covers medical care commodities, which consist of prescription and over-the-counter (OTC) drugs and supplies, and
medical care services, which include professional services,
hospital services, and medical insurance.
The professional medical services expenditure category
serves as the umbrella for a series of stratum indexes: Physicians’ services, dental care, eye care, and services by other
medical professionals. The hospital and related services category includes item strata for hospital services, nursing home
services, and adult day care. Medical insurance, for which
the weight share is reduced due to indirect pricing, is the remainder of the medical care services category. Details on the
more difficult pricing issues associated with these item strata
follow.
Prescription drugs. In response to technological change and
the complex marketing of prescription drugs, the CPI program has developed a series of techniques to show the effects
of such trends. Field staff uses special procedures to handle
the expiration of a drug’s patent protection and the subsequent introduction of equivalent generic drugs, a prescription
drug’s conversion to OTC status, and the introduction of new
pharmaceutical products into the market place.
Brand vs. generic. Since 1995, a method has been in place
allowing generic versions of prescription drugs coming off
patent to have a chance for inclusion in the CPI. Typically,
6 months after the expiration of the patent for a particular
prescription drug, the CPI economic assistant (data collector) disaggregates among all the FDA-designated therapeutically equivalent versions of the medicine, including the brand
name, that are available in each outlet in which the original
drug is priced. This process allows the newer generics an opportunity to build sales in the individual pharmacy over a

31 A National Academy of Sciences panel that reviewed CPI practices suggested that BLS should experiment with using selected medical conditions
as the CPI items. Under that approach, the CPI would follow the movement to new treatments (such as drug therapy replacing surgery) and, to
the extent possible, show any price changes. (See Charles L. Schultze and
Christopher Mackie, eds., At what price? Conceptualizing and measuring cost-of-living and price indexes (Washington, DC: National Academy
Press, 2002), pp. 178–190.

6-month period, and then, through disaggregation, a probability-proportional-to-size statistical technique, the generic
versions of the drug have a one-time chance for selection in
proportion to sales volume at the particular outlet. Should
a generic drug be selected, any price change that occurred
from brand to generic is reflected in the index.
Prescription vs. over the counter. When a drug in the CPI’s
prescription sample loses its prescription status and is sold
as an OTC drug, the CPI retains the item as part of the prescription drug sample while using its OTC price. Thus, the
prescription drug index series shows any price change that
occurs as a result of drugs changing status from prescription
to OTC. The OTC version of the drug remains in the prescription drug sample until it rotates out during the next rotation scheduled for that item. Generally, each sample rotates
every 4 years. At future sample rotations, the OTC item is
eligible for initiation in the nonprescription drug stratum and
ineligible in the prescription drug stratum. Similarly, should
a drug in the nonprescription sample change its status (that is,
from OTC to prescription), the CPI would show the resulting
price change, if any, in the nonprescription drugs and supplies index.
Physicians’ services. This item stratum covers services that
are performed and billed by private-practice medical doctors. This includes all medical professionals with a Doctor of
Medicine (M.D.) degree except for ophthalmologists, whose
services are priced in the eye care stratum. It also includes
osteopaths (they are not MDs, but often have hospital privileges). House, office, clinic, and hospital visits are included
as long as the bill comes from the physician.32 At initiation of
a quote for physicians’ services, the CPI data collector first
establishes the practitioner’s specialty and then disaggregates
to an appropriate service. The data collector describes the
characteristics of the selected visit and any related procedures
using a CPI checklist specific to the medical specialty. Current Procedural Terminology (CPT) codes are used to help
describe the item precisely; this description remains fixed for
the 4 years during which the CPI program follows its price,
unless either the selected combination of services changes or
a CPT code definition is modified. Transaction prices in the
physicians’ services index may include Medicare Part B payments, as well as those fees that the doctor expects to receive
directly from the patient or from private insurance.
Services by other medical professionals. This stratum covers services performed and billed by medical practitioners
who are not Medical Doctors (lack an M.D. degree) and are
not covered in the dental stratum or the eye care stratum. Included here are chiropractic and physical therapy, podiatry,

32 Ina Kay Ford and Daniel B. Ginsburg, “Medical care in the CPI,” in E.
Berndt and D. Cutler, eds., Medical care output and productivity, NBER
Research Studies in Income and Wealth. (Chicago: University of Chicago
Press, 2001), p. 215.

23

audiology (including hearing aids), acupuncture, nursing,
nutritional counseling, occupational therapy, and psychology
and psychotherapy.

rent discounts and (for those fee-for-service reimbursements
based on the hospital charge master with applied discounts)
list prices.

Hospital services. Items in the hospital services stratum cover the hospital portion of a medical treatment, including inpatient and outpatient services. The pricing unit is the hospital
visit, defined by a date of admission and a date of discharge
as documented on a hospital bill and usually associated with
a specific diagnosis or medical condition.33 At initiation, the
CPI data collector works with the respondent to select a hospital bill based on revenues generated by eligible payers. The
data collector refers to the bill to describe the item in terms
of the bundle of goods and services consumed during a time
frame or visit for the purpose of bringing the patient to the
physical (or mental) state required for discharge from the hospital. The form that the hospital visit takes as the pricing unit
is that of its reimbursement method, the method used by the
insurer to pay the hospital for the services. There are several
possible types of reimbursement that insurers may write into
their contracts with providers: Fee-for-service, diagnosisrelated group, per diem, case rate, admission rate, package,
ambulatory patient group, service units, and capitation. With
the exceptions of fee-for-service and fee schedule, each type
of reimbursement reflects either a lump-sum payment based
on the diagnosis or type of procedure performed or a flat fee
per unit of service.

Health insurance. As previously noted, the CPI employs an
indirect method to measure price change for health insurance.34 This indirect approach decomposes medical insurance into three parts:

Current procedures for selecting hospital services to price in
the CPI involve the following basic steps:

Price movement over time for the health insurance index in
the CPI is determined by the movements of the other medical
care strata, adjusted by changes in the retained earnings ratio.
(See subsection titled “Retained earnings ratio.”) This process
yields a measure of price change for insurance of constant
coverage and utilization. That is, changes in benefit coverage
and utilization levels generally are offset by compensating
premium charges and thus do not significantly affect retention
rates. Implicit in the process is the assumption that the level of
service from the individual carriers is strictly a function of the
benefits paid. Other changes in the amount of service provided
for policyholders, such as more convenient claims handling,
affect the movement of the index even though, strictly speaking, they should be removed; still, the effects are probably
small.

•

Disaggregation by setting to reflect the relative proportions of inpatient services versus outpatient services at the individual hospital outlet level

•

Disaggregation by payer (for example, self-pay or insurance company)

•

Selection of hospital bills based on selected payers,
when the hospital administration will provide them

•

Request for the type of reimbursement method and the
actual or estimated payment for the described hospital
visit, based on the terms of the contract between the
provider and the insurer

•

Description of the hospital visit, including a bundle
of procedures, services, equipment use, supplies, and
materials typically associated with the hospital event
or episode, as defined by both the bill and the contract
(the visit).

1. Changes in the prices of medical care items covered
by health insurance policies
2. Changes in the cost of administering the policies
3. Changes in the cost of maintaining reserves and, as
appropriate, profits
Most of the expenditure for health insurance goes for the
first item—the part that reflects the insurers’ payments for
medical treatment. The CPI program allocates this part of
health insurance spending to the medical care indexes for
those treatments in proportion to claims paid out for them.35
The remaining weight for the other two parts of insurance
is for the overhead of the insurers; this is all that remains
in the CPI health insurance index. Note that it is only consumer-paid insurance that is in scope; out-of-scope or ineligible health insurance receipts include those from employers,
Medicare Part A (funded through payroll taxes), Medicaid,
and workers’ compensation.

Monthly pricing and bimonthly pricing consist of updating the reimbursement method and amount based on the
contract between provider and insurer, and maintaining cur-

Retained earnings ratio. BLS obtains calendar year data for
premium income, benefit payments, and retained earnings.
For each year, the ratio of retained earnings to benefit payments is calculated, yielding a retained earnings ratio. The
latest year’s ratio is divided by the previous year’s ratio to
obtain the annual relative of change in the ratios. This annual relative of change is converted to a monthly relative (by

33 At this time, neither the pricing of total treatment paths nor the measurement of patient outcomes is available for the CPI, given the structure of
the medical care component and the difficulty of evaluating the results of
treatments.

34 See “Measuring price change for medical care in the CPI,” Consumer
Price Index (U.S. Bureau of Labor Statistics, April 12, 2010), http://www.
bls.gov/cpi/cpifact4.htm.
35 Ford and Ginsburg, “Medical care in the CPI,” p. 216.

24

taking its 12th root) so the CPI can reflect the change month
by month over the calendar year. Because it is not feasible
to obtain the monthly change in price caused by changing
retention margins, spreading the annual change evenly over
the year is preferable to reflecting the entire annual change in
one month.

Special pricing for other items
New vehicles. Prices for new cars and trucks, selected for inclusion in the CPI, pose a special problem because the manufacturer’s suggested retail (sticker) price is not the transaction
price for most new vehicles. Most automotive dealers offer
customers concessions on the sticker price or, for models that
are in high demand, the dealers charge an additional markup
beyond the sticker price. When collecting the price of new vehicles, BLS economic assistants record all of the components
of the sticker price separately. This includes the base price
and the prices for options, dealer preparation, transportation,
and so forth. In addition, they obtain from the dealer the average rebate, concession, and/or markup during the preceding
30 days. This enables BLS to estimate the true transaction
price.
Quality adjustment is also common in calculation of the
new vehicles index. The most frequently cited example of
direct quality adjustment in the CPI deals with the annual
model changeover for new cars and trucks. Each year, price
adjustments are made to account for the quality differences
between the old and the new models. In some cases, the adjustments are based on the previous model’s retail price for
optional equipment. In other cases, the quality adjustments
must be derived from production cost data supplied by the
manufacturers. These data are adjusted by estimated manufacturer and retailer markup rates to derive retail values for
the quality changes.
Adjustments for quality change in the CPI new car index
include structural and engineering changes that affect safety,
reliability, performance, durability, fuel economy, carrying
capacity, maneuverability, comfort , and convenience. Since
1999, quality adjustments have not been made for changes
associated with pollution control mandates.36
The derivation of production cost-based quality adjustments for new cars is carried out in association with the BLS
Producer Price Index and International Price programs. The
adjustments exclude changes in style or appearance, such as
chrome trim, unless these features have been offered as options and purchased by customers. Also, new technology
sometimes results in better quality at the same or reduced
cost. Usually, no satisfactory value can be developed for such
a change. In such cases, the quality change is ignored, and
prices are compared directly.

36 “The treatment of mandated pollution control measures in the CPI,”
Consumer Price Index (U.S. Bureau of Labor statistics, October 16, 2001),
http://www.bls.gov/cpi/cpitreat.htm.

In addition to quality adjustments for physical changes to
cars and trucks, adjustments are made for changes in the warranty coverage provided by auto manufacturers when sufficient data are available to derive estimates of their values.
Vehicle leasing. The vehicle leasing index was first published
by BLS in January 2002. The prices used in the index are
monthly lease payments. As with new vehicles, the agreedupon purchase price of the vehicle must be estimated. BLS
economic assistants collect the base price and the prices for
options, dealer preparation, transportation, and so forth.
Also, any rebates available are included, along with the largest estimated concession or discount the dealer would allow
for the leased vehicle on the day of pricing. Then, the lease
terms are applied to obtain the residual value, depreciation
amount, rent charge, and the total monthly lease payment.
During the annual model changeover, the quality adjustments
developed for the CPI new car index are also used in the CPI
vehicle leasing index.
Used cars and trucks. Models that are from 2- to 7- years-old
are priced in the used car and truck index. Data on used vehicle prices are obtained from a secondary source. Once a year,
each sample vehicle is updated by one model year to maintain
the same age vehicle. The sample prices are adjusted for quality change by applying the same information used for quality
adjustment in the new vehicle index. This is done by figuring
the percentage that the quality adjustments represent of the
price of the vehicle when it was new. The quality adjustments
are then assumed to depreciate at the same rate as the car as
a whole.
Apparel. The special characteristics of apparel marketing
have historically caused a number of problems in the maintenance of a constant-quality market basket of apparel in the
CPI. Many apparel items are seasonal and inventory is constantly fluctuating in reaction to changing fashions. In addition, large price changes are common as marketing practices
for apparel generally entail introducing such goods at high
regular prices and discounting to lower sale prices throughout their product lifecycle.
When an outlet discontinues an apparel item, the BLS
economic assistant follows the CPI substitution procedures to
find the closest substitute that the outlet offers for sale. These
procedures are developed by applying the results of hedonic
regression models and ensure that the economic assistant
matches many of the price-determining quality characteristics between the substitute and discontinued items.
Hedonic regression modeling is the technique used to
determine the importance of the price-determining quality
characteristics that add or subtract value to a particular good.
In this approach, an item can be viewed as a collection of
characteristics that, taken together, provide satisfaction or
value to the consumer. For example, a woman’s suit can be
considered an aggregation of its components, such as a jacket
and skirt or pants, each of which contributes value to the suit
in the eyes of the consumer. In addition, characteristics of
25

the suit, such as its fiber content and its construction, add or
subtract value from the consumer’s standpoint.
Hedonic regression modeling is a tool that allows commodity analysts to estimate which characteristics are pricedetermining, and how these estimates influence the direction
and magnitude of a good’s price. This research has resulted
in improved data collection documents and procedures for
pricing apparel commodities. By noting the most important
quality characteristics on data collection documents, economic assistants who collect data for the CPI can hold pricedetermining characteristics constant when pricing seasonal
and fluctuating inventories. This improvement in data collection documents has enhanced the reliability of apparel price
estimation, since it increases the number of direct price comparisons or sample observations that can be used for index
calculations.
When the economic assistant must substitute to a new
good or item because the previously priced item is no longer
available in the retail outlet, the commodity analyst determines whether the items are comparable, noncomparable, or
can be quality-adjusted by means of applying characteristic
estimates developed in a hedonic regression model. For comparable items the price of the new item is directly compared
to the price of the old item and the price change or relative is
used for index estimation. For noncomparable items the price
change for the item is imputed via the class-mean imputation
method. For quality adjustments the price of the discontinued
item is adjusted based on the difference in characteristics between the discontinued and substitute items using the characteristics value estimates developed in the hedonic regression
model for the apparel item. For example, if a two-piece men’s
suit was priced in the apparel price sample and is now no longer available in the outlet, because the retailer will only stock
three-piece suits instead of two-piece suits, the value of a vest
included with a three-piece suit can be added to the price of
the old two-piece suit using the hedonic price estimate for
the vest and the quality-adjusted price of the old two-piece
suit can be reliably compared directly with the price of the
new three-piece suit for a constant quality measure of price
change. Alternatively, if fiber percentages vary between two
items, quality adjustment can account for the characteristic
difference to permit constant quality comparison of the prices of the two goods.
Other characteristic differences that have been found to be
statistically significant also have been factored out to permit
constant-quality price comparisons for apparel items. Hedonic quality adjustments have played a significant role in
increasing the number of constant quality price changes that
can be used for index calculations by accounting for differences in quality characteristics when substitutions occur.

priced each month, field staff collects the cost of that fixed
amount. This amount is defined as a fixed number of therms
(a therm is 100,000 British Thermal Units, or BTUs). When
the surveyed outlet delivers and bills its residential customers
by the number of therms consumed, the CPI uses the current price per therm to determine the prices of that outlet’s
observations. But, when piped gas is delivered and billed by
volume (for example, cubic feet), the CPI program must adjust each quote to account for the fact that the volume of gas
needed to produce a constant amount of energy or BTUs varies, depending on the quality of gas (BTUs/CF). In this case,
the amount of gas priced each month is adjusted based on the
heat value of gas delivered by the outlet as follows:
Current adjusted consumption = original consumption × (original heat value/current heat value).
This adjustment ensures that a constant amount of energy
is being priced from month to month for the utility (piped)
gas service index.
Special pricing for seasonal items. Seasonal items are those
commodities and services that are available only at certain
times of the year rather than year round. Down parkas, snow
skis, and fresh tangelos are examples of seasonal items.
Special procedures are employed when selecting and pricing items generally available only part of the year to ensure
that they are appropriately represented in the sample and that
price changes are correctly included in the calculation of the
CPI. In particular, the procedures prevent substituting away
from a seasonal item when it is out of season.
Although seasonal items can exist in any ELI, some ELIs
include an especially large percentage of such items and,
consequently, receive special treatment. These seasonal ELIs
include most apparel items, fresh fruit, indoor plants and
cut flowers, fans and air-conditioners, some sports and recreational equipment, and admission to sporting events. The
designation of an ELI as seasonal or nonseasonal is made at
the regional level, using the four geographic census regions
in the CPI design. Some items that exhibit a seasonal selling
pattern in the Northeast region, for example, may be sold year
round in the South. In practice, though, nearly all ELIs designated seasonal are seasonal in all four regions.
After the samples for these seasonal ELIs are selected following the normal sample selection procedures, the number
of quotes is doubled to ensure that, despite the seasonal disappearance of a substantial number of quotes, a large enough
number of in-season quotes remains to calculate the index.
The quotes in these ELIs are paired; that is, for each original quote that is selected, a second quote in the same ELI and
outlet is initiated and priced 6 months later. In the fresh fruit
ELIs, one quote of each pair is designated January–June, and
the other quote is designated July-December. In all other seasonal ELIs, one quote of each pair is designated fall/winter,
and one quote is designated spring/summer. The fall/winter
and spring/summer designations are used for the nonfood
quotes because these are the distinctions that are most com-

Natural gas. To measure a constant consumption amount for
the CPI’s utility (piped) gas service index, the data collector initiates a fixed level of energy or heat consumption for
each observation. The fixed consumption amount is selected
in Washington, based on household bill expenditure data as
reported in the CE. Subsequently, when the observation is
26

monly used by the retailing industry to categorize seasonal
merchandise. These seasonal designations are used to help
establish the specific items eligible for each quote so that
year-round items and items from each season are initiated in
their proper proportions.
Economic assistants attempt to price every item in each
period during which it is designated for collection, even during those months when the item may be out of its indicated
season. If the item is available, the price is collected and used
in the calculation of the CPI. A common practice in marketing seasonal items, particularly seasonal clothing, is to mark
down prices to clear the merchandise from the stores as the
end of each season approaches. During the period when a
seasonal item is unavailable, its price is imputed following
standard imputation procedures. When an item returns at
the beginning of its season several months later, the price is
directly compared with the item’s last price, as it has been
imputed forward. This completes the circle in a sense: having
followed the price of the item down to clearance price levels, BLS then follows the price back up to regular (or at least
higher) prices the following season. Keep in mind that, in this
context, the “following” season means the same season the
next year; that is, the following fall/winter season for the fall/
winter sample, and the following spring/summer season for
the spring/summer sample.
When an item becomes permanently unavailable, the standard procedure is to substitute the most similar item sold in
the outlet. In the case of a year-round item not in a seasonal
ELI, this process takes place as soon as the item is permanently unavailable. For items in seasonal ELIs and seasonal
items in ELIs that are not designated seasonal, however, the
period during which a substitution can take place is restricted
to those months when a full selection of appropriate seasonal
merchandise is available.
These special initiation, pricing, and substitution procedures are intended to ensure that an adequate sample of items
is available every month, and that the correct balance of seasonal and year-round items is maintained. As a result, the estimates of price movement for the ELIs that include seasonal
items correctly reflect price changes not just for items available year round but for the entire universe of items included
in those ELIs.
Other price adjustments and procedures
Bonus merchandise adjustments. Sometimes, products
are offered with free merchandise included with the purchase of the original item. Such “bonus” items may provide additional satisfaction to consumers, and BLS will,
therefore, make adjustments to the purchase price to take
into consideration the value of the bonus merchandise.
The adjustment made depends on the type of merchandise
offered and the perceived value of the bonus to the consumer. If the bonus merchandise consists of more of the
same item, the adjustment is reflected in the price of the
item. For example, if a manufacturer offers two ounces of
toothpaste “free with the purchase of the regular 6-ounce

tube,” the item’s price is adjusted to reflect a decrease in
the per ounce price. When the bonus is removed, the price
per ounce returns to its previous level, and a price increase
is recorded. In this instance, the value to the consumer is
assumed to be one-third greater during the bonus period.
If the bonus merchandise consists of an item that has some
significant value to the consumer, and the item is of a different genre, an adjustment is made to account for the value of the free item when it is feasible to do so. Bonuses that
are contingent on an additional unrelated purchase, such
as a free can of soup when purchasing a whole chicken
from the poultry case, are ignored.
Cents-off coupons. For a coupon to be used to reduce the
reported price of an item, the coupon must be either attached to the item, attached to the product’s display shelf,
dispensed by machines attached to the product’s display
shelf, located at promotional displays, or distributed to all
shoppers by product representatives standing in the immediate vicinity of the display shelf. All other coupons
presented by customers as purchase reductions at the time
of payment are ineligible.
Concessions. A concession is a deduction of a specific
amount from the proposed selling price for the item. The
usual CPI practice is to subtract from the proposed selling
price the average concession for the priced item over the
past 30 days.
Different day pricing. For a subset of items, if the selected
priced item is not available for sale at the time of collection, prices from up to seven days prior to the actual day
of collection are eligible. The item must have been offered
for sale during the previous 7 days and the most recently
available price is reported. The list of items eligible generally consists of specific items that may not be available
every day, such as a specific type of fresh fish.
Discounts. A discount price is a reduced price that is
available to only certain customers in a specific outlet. If
the discount is available only during the period of price
collection, such as that for a grocery-card discounted item,
the discount is included only if 50 percent or more of sales
for the affected item are discounted. If the discount is in
effect for more than one collection period and the discount
applies to 5 percent or more of the dollar sales of the item
in the outlet, a probability selection is made to determine
if the discount should be collected. For example, if the
regular cash price accounts for 84 percent of sales, senior
citizens’ discounts account for 10 percent and employee
discounts account for 6 percent of sales, a one-time probability-based selection is made among the three options to
determine which price to report.
Manufacturers’ rebates. When product manufacturers
offer customers cash rebates at the time of purchase for
purchases of items priced in the CPI, these rebates are reflected in the index as price reductions. When a rebate is
27

offered for a priced new vehicle, it is the estimated average
rebate over the past 30 days that is subtracted from the
vehicle’s reported price. For vehicle leasing, it is the rebate in effect as of the day the collected price is obtained.
For mail-in rebate offers, the price of the affected item is
reported without subtracting the amount of the rebate. An
attempt is made to determine the proportion of customers
who take advantage of the rebate, and prior to use in the
index, the reported price is then adjusted accordingly.

reflect current prices and price trends more accurately. Also
excluded are refunds that are paid directly to consumers in
a separate check and are not part of the bill. The utility indexes do include current-period credits that are based on
current consumption, such as purchased gas adjustments
and fuel adjustments.
Unit-priced food items. When food items that are sold
on a unit basis but lack a labeled weight are being priced,
two items are weighed to permit calculation of an average
weight for the item. This helps reduce the variability in size
that occurs among individual, loose items and is not overly
burdensome for the data collection process. For example, if
the item being priced is Red Delicious apples, and the price is
50 cents each, the BLS field staff reports the price of one apple
and the combined weight of two apples taken from the produce
rack. In computing the price per ounce, the combined weight is
divided by 2, and the 50-cent price of the Red Delicious apple is
divided by this average weight.

Membership retail outlets. Outlets that require a membership fee to be paid in order to be able to shop at the outlet
are eligible for pricing in the CPI. If the actual price paid
for products varies with the level of membership, a specific
membership is selected and the reported prices reflect that
membership level.
Quantity discounts. Many items in the CPI are sold both individually and in quantity. When consumers are able to purchase an amount greater than a single unit at a discounted
price, the first multiple-unit price is reported for use in the
CPI. For example, if the 12-ounce can of corn being priced
can be purchased at 25 cents for a single can, three cans for
69 cents, or five cans for $1, the price used in the CPI will be
the per ounce price of the three cans.

Container deposits. BLS collects information on container deposits for a variety of nonalcoholic and alcoholic beverages to
reflect the influence of changes in deposit legislation on price
change. Consumers who purchase throw-away containers are
considered to be purchasing both the product itself and the convenience of throwing the container away. When a local jurisdiction enacts deposit legislation and no longer allows stores to sell
throwaway containers, those consumers who were previously
purchasing throwaway containers may experience a change in
the price of this convenience. The price of the same-sized container of product plus its deposit establishes an upper bound for
the price change, because the consumer could retain the former
convenience by now purchasing returnables and simply throwing them away. In similar fashion, information about deposits
and the status of legislation can be used to estimate price change
when a container bill is repealed. Changes due to the enactment
or repeal of container bills are shown in data for the month in
which the legislation becomes effective.

Shoppers’ cards. If a priced outlet issues a card offering a
“card discount” on selected products purchased by cardholders, such discounts are treated as “temporary discounts” and
processed as follows. The discount is included only if 50 percent or more of sales for the affected item during the collection period are subject to the card discount.
Special-day prices. If a selected outlet has different prices
for priced items based on the day of the week when a purchase is made, a selection is made between special-day and
regular-day purchases, based on revenue. If the “special
day” is selected, the price collected is for the most recent
special-day price.

Sales taxes. The CPI includes all applicable taxes paid by consumers for services and products purchased. Many prices for
services and products used to calculate the CPI are collected
with taxes included because this is the manner in which they
are sold. Examples are tires and cigarettes. Other prices are collected excluding applicable taxes, with those taxes subsequently
added in the Washington office. The tax rates for these items are
determined from secondary sources based on the State, county,
and local tax structure governing the sale of the service or product at the point of purchase.

Utility refunds. Sometimes, public utility commissions require that utilities such as telephone, natural (piped) gas,
or electricity companies make rebates to their customers.
These rebates may arise from a number of different causes.
For example, a utility may be permitted to use a new rate
schedule temporarily until a final determination is made.
If the final rates set by the commission are lower than the
temporary ones, the difference must be refunded for consumption during the period. The CPI does not always view
such refunds as reflecting current period prices for utility
services. If all customers, both new and existing, are subject
to having the refund applied to their bill, then the refund is
included in the total price calculation. However, if the refund is only applied to those customers who were originally
subject to the overcharge (i.e., existing customers only) then
the refund is excluded. This procedure reduces the monthto-month volatility of utility indexes and ensures that they

Index calculation
As stated earlier, the CPI is actually calculated in two
stages. Earlier sections described the first stage of that calculation—how the CPI calculates the basic or elementary
indexes, which show the average price change of the items in
each of the 8,018 CPI item–area combinations.

28

The next section describes the second stage of calculation: how the aggregate indexes are produced by averaging
across the 8,018 CPI item–area combinations.

rotation schedule for updating the expenditure reference
period. Effective with the January 2004 index, the expenditure reference period changed from β = 1999–2000
to β = 2001–2002; effective with the January 2006 index,
it was updated again to 2003–2004; and so forth. It is
Estimation of upper level price change
worth noting that a change in the expenditure reference
results
in a change in the implicit quantity (Q)
ed earlier, the CPI
is
actually
calculated
in
two
stages.
Earlier
sections
described
the first
stage of
  Aggregation of elementary CPI data into published index- period
assigned
to
each
elementary index, but not the implicit
culation—howesthe
CPI
calculates
the
basic
or
elementary
indexes,
which
show
the
average
price
requires three ingredients: elementary indexes, elemenprice
component
(P)
of the aggregation weight (AW) of
of the items intary
each
of
the
8,018
CPI
item–area
combinations.
expenditures to use as aggregation weights, and a price
each
elementary
index.
index aggregation formula that uses the expenditures to ag-

xt section describes
thethe
second
stage
calculation:
how theinto
aggregate
indexes are produced by
gregate
sample
ofofelementary
indexes
a published
ng across the index.
8,018 CPI item–area combinations.
Table 2. Expenditure reference periods for the Consumer

Price Index, All Urban Consumers (CPI-U) and the Urban
Wage Earners and Clerical Workers (CPI-W), 1917–2015

ation of upper level price change

Input elementary price indexes
  The CPI-U, CPI-W, and all versions of the C-CPI-U are
ation of elementary
CPI data
published
indexes
requires three
ingredients: elementary
constructed
byinto
using
the same
combination
of Laspeyres
Expenditure
, elementary expenditures
to
use
as
aggregation
weights,
and
a
price
index
and geometric mean elementary indexes. In other words,
theaggregation formula
period
s the expenditures to aggregate the sample of elementary indexes into a published reference
index.
prices for each series are combined in the same way to form
1917–1919
the elementary price indexes.

elementary price indexes

Month
introduced

Terminal
month

1919

Dec. 1924

Jan. 1925

Dec. 1929

Jan. 1930

Dec. 1949

Jan. 1950

Dec. 1952

Jan. 1953

Dec. 1963

Jan. 1964

Dec. 1977

Jan. 1978

Dec. 1986

Jan. 1987

Dec. 1997

Jan. 1998

Dec. 2001

Jan. 2002

Dec. 2003

2001–2002

Jan. 2004

Dec. 2005

2003–2004

Jan. 2006

Dec. 2007

2005–2006

Jan. 2008

Dec. 2009

Avg. 1917–1919 and
1934–1936

Input
elementary
weights by using the same combination of
I-U, CPI-W, and
all versions
of theexpenditure
C-CPI-U are constructed
1934–1936

To
aggregate
elementary
indexes
into
published
indexes,
res and geometric mean elementary indexes. In other
words,
the prices
for each series are
aggregation
for each
1947–1949
ed in the sameanway
to form theweight
elementary
priceelementary
indexes. item–area combination is required. The function of the aggregation weight
1950
is to assign each elementary index a relative importance or
elementary expenditure weights
contribution in the resulting aggregate index. The aggrega1960–1961
tion weight corresponds to consumer tastes and preferences
regate elementary indexes into published indexes, an aggregation weight for each elementary
1972–1973
and resulting expenditure choices among the 211 elementary
ea combination is required. The function of the aggregation weight is to assign each elementary
items in the 38 elementary areas comprising the CPI sample,
1982–1984
relative importance or contribution in the resulting aggregate index. The aggregation weight
for a specified period.
onds to consumer tastes and preferences and resulting expenditure choices among the 211
1993–1995
tary items in the 38 elementary areas comprising the CPI sample, for a specified period.
1999–2000
CPI-U and CPI-W. In the CPI-U and CPI-W, aggregation
and CPI-W. In
the CPI-U
weights (AW) are defined as
weights
(AW)and
areCPI-W,
defined aggregation
as

i ,a , p

AW 

i ,a , p

 
( P Q )
100

,

2007–2008
Jan. 2010
Dec. 2011


P is item
P is the estimated
(i) purchased
in area
(a) by
(p) in period (α), and
where i ,aprice
the estimated
price
of item
(i) population
purchased
, p α of

2009–2010
Jan. 2012
Dec. 2013
in area (a) by population ( p) in period (α), and i ,a , p Qβ is
is the estimated quantity of item (i) purchased in area (a) by population
(p) in period (β). Period
the estimated quantity of item (i) purchased in area (a)
2011–2012
Jan. 2014
Dec. 2015
he base period by
of the
corresponding
item–area
population
( p) inelementary
period (β).
Periodindex.
(α) isFor
theexample,
base the “Sports
Note: 1985.
Prior toCPI
January 1953, previously published indexes often
ent” (ITEM =period
RC02) of
in the
Seattle
(AREA = A423)
index has
a base period
of α = June
corresponding
elementary
item–area
index.
wereperiod
revisedof
retroactively,
on the basis of more recent consumer
tary indexes have
varying
base
periods.
Most
published
indexes
have
an
index
base
α=
For example, the “Sports equipment” (ITEM = RC02) in
expenditure
data.
984.
Seattle (AREA = A423) index has a base period of α =
Source: U.S. Bureau of Labor Statistics.
June 1985. CPI elementary indexes have varying base peantity (β) corresponds
to thepublished
reference indexes
period ofhave
the expenditures
usedperiod
to derive the implicit
riods. Most
an index base
y weights needed
for
Laspeyres
aggregation.
As
of
2014,
the
CPI-U
and
CPI-W had an
of α = 1982–1984.
Aggregation weights for the CPI-U and CPI-W are deiture reference period of β = 2011–2012. Historically, the CPI expenditure reference period was
rived
from estimates of household expenditures collected
The quantity (β) corresponds to the reference period
d approximately every 10 years. (See table 3.) In 2002, BLS instituted a biennial rotation schedule
in
the
CE. Despite an increase in the CE sample size in
of the expenditures
used
to derive
quantity
ating the expenditure
reference period.
Effective
withthe
the implicit
January 2004
index, the expenditure
1999,
expenditure
weights
for Laspeyres
aggregation.
As ofwith
2014,
ce period changed
fromneeded
β = 1999–2000
to β = 2001–2002;
effective
the January 2006 index, estimates at the elementary item–area
would be unreliable due to sampling error without
the2003–2004;
CPI-U andand
CPI-W
hadIt an
expenditure
reference
pdated again to
so forth.
is worth
noting that
a changepein thelevel
expenditure
the
use
of statistical
smoothing procedures. BLS uses two
riod
β = 2011–2012.
Historically,
theassigned
CPI expenditure
ce period results
in of
a change
in the implicit
quantity (Q)
to each elementary index,
but
basic
techniques
to
minimize
the variance associated with
was
updated weight
approximately
every
10
implicit price reference
componentperiod
(P) of the
aggregation
(AW) of each
elementary
index.
each
elementary
item–area
base-period
expenditure estiyears. (See table 2.) In 2002, BLS instituted a biennial

,a , p

able 2
29

mate. First, data are pooled over an extended period in order to build the expenditure estimates on an adequate sample size. The current reference period (β) uses 24 months
of data.37 Second, elementary item–area expenditures are
averaged, or composite-estimated, with item-regional

expenditures.38 This has the effect of lowering the variance of
each elementary item–area expenditure at the cost of biasing
it toward the expenditure patterns observed in the larger geographical area. This process is summarized in the equations.
in exhibit 3.

Exhibit 3. Estimation of CPI-U elementary aggregation weights
Expenditure on item (i) in area (a) by population (p) in year (βn)

i ,a , p

i ,a , p
i , a∈a , p

Share of total expenditures for item (i) in area (a) for population (p)
in year (βn)

i ,a , p

i ,a , p

sβ =

( PQ) β

∑

( PQ)  
i ,m , p
n

i ,a

∑

i ,a , p
i , a∈m . p

s 
n

i ,m , p
i

( PQ) 

( PQ) 

i ,m , p
i , m∈m . p

( PQ) 

∑

i ,m , p
i , m∈m , p

i ,a , p

n

n

i

i ,m , p

n

( PQ) β

∑

Total expenditures in major area (m) by population (p) in year (βn)

Share of total expenditures for item (i) in area (m) for population (p)
in year (βn)

( PQ) β

i

n

i ,a , p
i ,a∈m , p

Expenditure on item (i) in major area (m) by population (p) in year (βn)

n

n

n

n

( PQ) 


s   i ,m , p s   (1   )i ,a , p s
n

 i

~~


i , a , p ( P Q ) n   ∑i , a , p ( PQ ) n  
 i , a∈a , p


Estimated expenditure on item (i) in
area (a) by population (p) in year (β n)

n

i

∑

Total expenditures in area (a) by population (p) in year (βn)

Composite-estimated share of total expenditures for
item (i) in area (a) for population (p) in year (βn)

( PQ ) 

n



n


s

i , a , p n

i,a

Raked expenditure on item (i) in
area (a) by population (p) in year (β n)

( Pˆ Qˆ ) β =
i ,a , p
n

∑

( PQ) β

∑

~~
( PQ) β

i,a,p

~~
i,a ∈ e , m
( PQ ) β × i,a
i ,a , p
n

i,a,p
i,a ∈ e , m

Estimated expenditure in expenditure reference period (β)

N
ˆ Qˆ ) = 1  ∑ ( Pˆ Qˆ )
(
P
i,a,p
β
i,a,p
β
N  n=1

n

n

n





38 Elementary areas are grouped into city-size classifications by region
for the purpose of composite estimation. There are four regions (Northeast,
Midwest, South, and West) and two city-size classifications (A-sized cities
and non-A-sized cities) for a total of eight regional city-size classifications.

Prior to 2002, the expenditure reference period was based on 36 months
of data (for example, β = 1993–1995 from 1998 to 2001 and β = 1982–1984
from 1987 to 1997).
37

30

Exhibit 3. Estimation of CPI-U elementary aggregation weights—continued
Cost weight in pivot month (v)

i,a,p

Aggregation weight

( PˆvQˆ β ) =

i,a,p

i ,a , p

 IX
( Pˆ Qˆ ) β ×  i,a,p a,v
 IX
 i,a,p a,β

( Pˆ Qˆ  ) 

i ,a , p






( PˆvQˆ  )

i ,a , p

IX ,v

where
p

=

population (urban or urban wage-earner)

a

=

CPI elementary area

i

=

CPI elementary item

e

=

expenditure class

m =

One of eight CPI major areas, defined by region and city-size classification.
Regions are Northeast, Midwest, South, and West; city-size types are selfrepresenting and non-self-representing

P

price

=

Q =

quantity

N =

number of years in the CPI-U expenditure reference period (NOTE: Currently,
N = 2.)

βn =

year belonging to expenditure reference period β (NOTE: n = 1 is 1999 and n = 2
is 2000 in the current CPI-U expenditure reference period.)

δ

=

weight assigned to major area (m), where 0 < δ < 1

α

=

lower-level index base period

v

=

year and month, usually December, prior to the month when expenditure weights
from reference period β are first used in the CPI

=

estimated expenditures (PQ) for item (i) in area (a) for population (p) as a percent
of total CPI expenditures in area (a) in period βn,

i,a IXα,β

=

lower-level index of price change from index base period (α) to expenditure
reference period (β) for item (i) in area (a)

i,a IXα,v

=

lower-level index of price change from index base period (α) to pivot-month (v)
for item (i) in area (a)

i ,a , p

S

n

31

.

CPI-U aggregation weights. Like the biennial data used for
CPI-U aggregation, adequacy of the underlying sample size
from which the expenditure weights are estimated is an issue
for C-CPI-U aggregation. To minimize the variance of the elementary item–area monthly expenditures, a ratio-allocation
procedure is adopted to estimate each item–area monthly expenditure from U.S. monthly item expenditures:

~~
The estimated expenditure i ,a , p ( P Q ) β for item (i) in area
(a) for population (p) in reference period (β) is derived from
a weighted average of the item’s relative importance in the
elementary area (a) and its relative importance in its corresponding region-size classification (m), for each year encompassing reference period (β). The weight (δ) assigned to the
region-size class (m) and the weight (1 – δ) assigned to the
elementary area (a) are a function of the variance in each area
and the covariance of each measure.39 The resulting average
share ( s ) is then multiplied by the sum of all expenditures
in the elementary area in the corresponding year to obtain a
revised item expenditure. In a process called “raking,” the
revised item expenditures are adjusted by a factor such that,
once summed, they equal the unadjusted expenditures at
the region-size class (m) expenditure class (e) level. Annual
item–area expenditures (βn) have a lower bound of 1 cent
($0.01). The raked item expenditures in each year of reference
period (β) are then averaged to obtain the estimated expenditure in (β). Finally, the estimated expenditure is adjusted by
the corresponding item–area index to obtain the aggregation
weight: an expenditure value with an implicit price of period
(α) and implicit quantity of period (β).
Because the initial version of the C-CPI-U is published
simultaneously with the CPI-U, it uses expenditure data from
the same expenditure reference period (β) as the CPI-U as
aggregation weights. Unlike those in the CPI-U, however, the
expenditures are not adjusted forward to a December pivot
month and rebased so that the implicit price corresponds to
the item–area index base period. Rather, the estimated expenditure weights with implicit prices of period (β) and implicit quantities of period (β) are used as aggregation weights.
Before 2015, the interim version of each monthly C-CPI-U
index was published in February of the ensuing year. Hence,
if the ensuing year was one in which the weight was updated,
then the interim version of each monthly C-CPI-U was based
on more contemporaneous expenditures than its initial version. For example, 2012 initial indexes produced in 2012 used
β = 2009–2010. Interim indexes for 2012 were produced in
2013 and likewise used β = 2009–2010. Initial indexes for
2013 also used β = 2009–2010. However, 2013 interim indexes produced in 2014 (a weight update year) were constructed
using β = 2011–2012.40
n

Estimation of monthly expenditures at the elementary
level
Estimated monthly expenditures
T

( Pˆ Qˆ ) t =
i ,a , p

∑

a

∑

i ,a , p
i ,a∈i , A

( PQ) t

t∈T
a
T

i ,a , p

∑∑
i ,a∈A t∈T

( PQ) t

,  

( PQ) t
i ,a , p

where
p

= population (NOTE: C-CPI-U is produced for the

urban population only.)
a = CPI elementary area
i = CPI elementary item
A = all CPI elementary areas (“U.S. city average”)
P = price
Q = quantity
t = month
T = period covering month (t) and 11 months prior to

month (t)
The monthly expenditure for an item in an elementary area
is derived in two steps: First, the monthly expenditure for the
item is summed across all 38 areas to obtain a U.S. monthly
item expenditure. Second, the U.S. monthly item expenditure
is allocated among all 38 elementary areas, according to each
area’s relative expenditure share for the item during the current and preceding 11 months. Note that

i, A

( PQ) t  i , A ( Pˆ Qˆ ) t . 

The estimated monthly item–area expenditures have
a lower bound of 1/12th of a penny ($0.000833), and when
summed over the calendar year, they have a lower bound
($.01) equivalent to that of the annual data in the CPI-U expenditure reference period.

Final C-CPI-U. For the final C-CPI-U, which uses the Törnqvist index for upper-level aggregation in a monthly chained
construct, monthly expenditure estimates for each elementary item–area combination are required as aggregation
weights. These are derived from the same CE data as the

Aggregation formula
A Laspeyres price index is used to aggregate elementary indexes into published CPI-U and CPI-W indexes. The Laspeyres index uses estimated quantities from the predetermined
expenditure reference period (β) to weight each elementary
item–area index. These quantity weights remain fixed for a
2-year period, and then are replaced in January of each even
year when the aggregation weights are updated. In a Laspey-

39 For more information on composite estimation, see Michael P. Cohen
and John P. Sommers, “Evaluation of the methods of composite estimation of cost weights for the CPI,” Proceedings of the American Statistical
Association, Business and Economic Statistics Section (Alexandria, VA:
American Statistical Association, 1984.) pp. 466–471.
40 Starting in 2015, BLS will begin issuing four preliminary estimates of
the C-CPI-U, by quarter, with final data issued approximately 1 year after
the reference month.

32

res aggregation, consumer substitution between items is assumed to be zero. The aggregate index for any given month
is computed as a quantity-weighted average of the current
mputed as a quantity-weighted
average
of the
current
month index divided
by the
index
valuemonth
in the index
index base pee index base period.
Month-to-month
price
change
is
then
calculated as
as aa rariod. Month-to-month price change is then calculated
ndexes. The relevant
equations
are
as
follows:
tio of the long-term monthly indexes. The relevant equations
are as follows:

vel aggregation formula

CPI-U and CPI-W upper-level aggregation formula
Long-term price change

I , A, p

IX[Lz ;t ] 

I , A, p

IX[Lz ;v ] 




i ,a , p
i ,aI , A
i ,a , p
i ,aI , A

Month-to-month price change
I , A, p

I , A, p

IX [Lt-1;t ] =

IX [Lt

I , A, p

1L;t ]

IX [ z ;t ]

i ,a , p

IX[ ;t ]

AW 

i ,a , p

IX[ ;v ]

IX [Lz ;t-1]
I , A, p

Lo rG

I , A, p

,   I , A, p

Final C-CPI-U upper-level aggregation formula
Long-term price change

I , A, p

Lo rG

AW 



by the Törnqvist formula; rather, it is implicitly accounted
for by use of current- and base-month expenditures. An index
of 1-month price change is calculated and then multiplied by
the index value for the previous month to obtain the currentmonth index value. Following are the relevant equations:

IX

IX

L
[ z ;t ]

L
[ z ; t -1]

,

,

I , A, p

IX [Tt -1;t ]

IX[Tz ;t -1] ×

IX[ z ,v ]

CPI-U index of price change from
= aggregate-level
level index of price change
from period (α) to pivot-month (v) for item
period (z) to pivot month (v) for aggregate item (I)

rea (a)
in aggregate area (A) for population (p)

In contrast,
theperiod
C-CPI-U
built(i)by
gation weight from
reference
(β) foris item
in chaining
area (a) together indexes of 1-month price change. For the final CCPI-U index,
each monthly index is computed using the Törnqvist formula
gate-level CPI-U index of price change from period (z) to pivot month
with monthly weights from both the current and the previaggregate item
(I)month.
in aggregate
area (A)
for population
(p) is not assumed
ous
Consumer
substitution
behavior

i ,a , p

s

or

t -1

i , a , p

2

s

t

or

,

p
=

a
=

population (Note: the C-CPI-U is calculated for the urban consumer population only.)

A
=
i
=

aggregate area

CPI elementary area
CPI elementary item

I X[α;t] =

lower-level index of price change from pe-
riod (α) to month (t) for item (i) in area (a)

IX[α;t-1] =

S =
i,a t

S =
i,a t-1

lower-level index of price change from pe-
riod (α) to month (t – 1) for item (i) in area
(a)

IXT[z;t] =

aggregate-level C-CPI-U Törnqvist index
of price change from period (z) to month
(t) for aggregate item (I) in aggregate area (A)

i,a

[α;t]

= lower-level index of price change from period (α) to
nd month, usually
to (i)
thein month
L December,
  month (t) prior
for item
area (a)when expenditure
I , A , p IXperiod
[ ,v ]
s from reference
(β) are first used in the CPI

= lower-level index of price change from period (α) to

pivot-month (v) for item (i) in area (a)
AWβ change from period (α) to month (t) for item (i) in
level index ofi ,a , pprice
= aggregation weight from reference period (β) for
a)
L item (i) in area (a)

∏

L G
IX[ ;t ] 
i ,a , p

L G
IX[ ;t -1] 
i ,a , p


IX[Tt -1;t ] ,

I
=
aggregate item

z
=
base period of the aggregate index (NOTE:

the U.S. city average, all-items C-CPI-U in

dex has a base-period of z = December

1999.)
a
=
base period of the elementary index (i) in

area (a)
t
= month

base period
elementary
indexall-items
(i) in area
(a)
α = index
eriod of the aggregate
(Note: of
thethe
U.S.
city average,
CPIx has a base period
= 1982–1984.)
v of= zyear
and month, usually December, prior to the
month when expenditure weights from reference
eriod of the elementary
index (i) in area (a)
IX period (β) are first used in the CPI





i ,a∈I , A


I , A, p

where

t = month
mentary items (“all-items”)
z = base period of the aggregate index (Note: the U.S.
city average, all-items CPI-U index has a base pe riod of z = 1982–1984.)

I , A, p

I , A, p

Month-to-month price change

where
A = all elementary areas (“U.S. city average”)
mentary areas (“U.S. city average”)
a = CPI elementary area
ementary area p = population (the C-CPI-U is calculated for the U-
population only.)
ation (the C-CPI-U
is
calculated for the U-population only.)
i = CPI elementary item
ementary item I = all elementary items (“all-items”)

i,a

pIX[Tz ;t ] =

i, a

I,A

42
33

expenditure in month (t) for item (i) in
area (a) as percentage of total expenditures
in month (t) for aggregate item (I) in aggre
gate area (A)
expenditure in month (t-1) for item (i) in
area (a) as percent of total expenditures in
month (t – 1) for aggregate item (I) in ag-
gregate area (A)

Starting in 2015, BLS began revising the Chained Consumer
Price Index for All Urban Consumers (C-CPI-U) quarterly,
and the Constant Elasticity of Substitution (CES) formula
will replace the adjusted geometric mean formula for the
calculation of the preliminary versions of that index.
The initial version of the C-CPI-U will continue to be
released concurrently with the CPI-U for each calendar month.
The C.E.S. formula will be used to calculate the C-CPI-U.
The final version of the index will be released approximately
10–12 months later, according to the publication schedule
outlined in the following table:

C-CPI-U quarterly release schedule
Index month
1 (Feb)

as Interim (1), Interim (2), and Interim (3) in the table. The
1-month price change for each interim release will be the
same as the initial version. The interim versions reflect only
updates to index levels—that is, the value of the index in a
given month relative to the value in its base period. These
updates result from the conversion of 1-month price changes
from initial to final value in preceding months in the monthly
chained series.
The CES uses an estimate of consumer substitution that
lies between the estimates assumed in the geometric mean
and Laspeyres formulas, and represents a model that is closer
to actual consumer behavior. This estimate of consumer
substitution, sigma (σ), is called the elasticity of substitution.

Quarterly release
2 (May)
3 (Aug)
4 (Nov)

�
���IX �������

5 (Feb)

y – 1, 1

Final

Final

Final

Final

Final

y – 1, 2

Final

Final

Final

Final

Final

y – 1, 3

Final

Final

Final

Final

Final

y – 1, 4

Interim3

Final

Final

Final

Final

y – 1, 5

Interim3

Final

Final

Final

Final

y – 1, 6

Interim3

Final

Final

Final

Final

y – 1, 7

Interim2

Interim3

Final

Final

Final

y – 1, 8

Interim2

Interim3

Final

Final

Final

y – 1, 9

Interim2

Interim3

Final

Final

Final

y – 1, 10

Interim1

Interim2

Interim3

Final

Final

y – 1, 11

Interim1

Interim2

Interim3

Final

Final

y – 1, 12

Interim1

Interim2

Interim3

Final

Final

y, 1

Initial

Interim1

Interim2

Interim3

Final

y, 2

Interim1

Interim2

Interim3

Final

y, 3

Interim1

Interim2

Interim3

Final

y, 4

Initial

Interim1

Interim2

Interim3

y, 5

Interim1

Interim2

Interim3

y, 6

Interim1

Interim2

Interim3

y, 7

Initial

Interim1

Interim2

y, 8

Interim1

Interim2

y, 9

Interim1

Interim2

y, 10

Initial

Interim1

y, 11

Interim1

y, 12

Interim1

y + 1,1

Initial

�
�����

�����
�
�����������
IX
� �∑������� ��
�
� � ����� �
��
�
IX�������
∑������� �����������
�
�
�
��
�
����� �
�
�����
�
�
�
�����������
IX
�∑������� ��
� � ������� �
��
�
IX�������
∑������� �����������
�
�

New formulas and quarterly release schedule

The CES
month-to-month index relative for a biennial period

is
�
��,�,�,��,�

�

���,� ���,�

IX�,�,���
�
�
IX�,�,����

�����

,

where

C = C.E.S index/weight,

i = elementary item stratum,

I = aggregate item,

a = elementary index area,

A = aggregate index area,

IX = component index,

T = current calendar month in calendar year y (e.g.,
if IXyymm = IX0308, then t = August and y =
2003),

t –1 = calendar month previous to calendar month t,

x = reference period of index (initially, x =

December 1999),
௜ǡ௔
ܲ௕  = biennial expenditure reference period,

= during biennial expenditure reference period
ܳ݅ǡܽ
ܾ  

σ = sigma for component and aggregate index

periods and for weight period, and

V = pivot month index period.

In the table, the lightface italic gray text indicates that the
final index has been released earlier. Thus, reading down
the first column shows that, in February of the current
year (y, 2), the final version will be released for the first 3
months (January, February, and March) of the previous year
(y – 1, 1; y – 1, 2; and y – 1, 3) and the initial version will
be released for January of the current year (y, 1). Column 5
of the current year corresponds to column 1 of the previous year. The other columns are read similarly. A blank cell
indicates that it is too early to have an index for that month
at that time.
The final index will continue to be calculated with the
Törnqvist formula. In between the initial release and the
final release, there will be three quarterly updates, noted

Calculation of seasonally adjusted indexes
Seasonal adjustment. Seasonal adjustment removes the estimated effect of changes that normally occur at the same time
every year (such as price movements resulting from changing
climatic conditions, production cycles, model changeovers,
holidays, and sales). CPI series are selected for seasonal adjustment, if they pass certain statistical criteria and if there is
an economic rationale for the observed seasonality. Seasonal
factors used in computing the seasonally adjusted indexes
are derived, using X-13ARIMA-SEATS seasonal adjustment software. X-13ARIMA-SEATS is an extension of the

34

However, for affected series, the resulting seasonal factors better represent the true seasonal patt
factors calculated without these techniques. These seasonal factors are applied to the original un
series. Level shifts and outliers, removed in calculating the seasonal factors, remain in the result
seasonally adjusted series.

X-12 variant of the Census Method II Seasonal
Adjustment
and
aggregative
adjustment.
Each
year, BLS
In recent
years, BLS Direct
has used
intervention
analysis
seasonal
adjustment
forseasonvarious indexes—ga
methodology. In some cases, intervention analysis
seasonal
ally
adjusts
eligible
lower
level
CPI
index
series
directly
with electricity, ut
fuel oil, new vehicles, women's and girls’ apparel, educational books and supplies,
adjustment is carried out using X-13ARIMA-SEATS,
to
dethe
X-13ARIMA-SEATS
software
using
unadjusted
indexes
(piped) gas service, water and sewerage maintenance, nonalcoholic beverages and beverage mat
rive more accurate seasonal factors. Consumer
indexes
latest 5 Series
to 8 calendar
years.using
CPI intervention
index series analysis
are ad- techniques w
and price
whiskey
at homefor
arethe
examples.
are adjusted
may be adjusted directly or aggregatively, depending
onare
theclearly
justed
using theAfter
multiplicative
model.
interventions
identified.
a number of
years, series may revert to adjustment usin
level of aggregation of the index and the behavior
of the
com- In addition,
Most high-level
are adjusted
byisthe
aggregastandard
methods.
for some index
series,series
intervention
analysis
used,
and the resulting se
ponent series.41
method,
which
is more
appropriate
broad
not show a clear and tive
stable
seasonal
pattern.
In these
cases, theforseries
is categories
not seasonally adjusted.
whose component indexes show strongly different seasonal
Intervention analysis seasonal adjustment. Some index patterns. Under the aggregative method, direct adjustment is
Direct and aggregative adjustment. Each year BLS seasonally adjusts eligible lower level CPI i
series show erratic behavior due to non-seasonal
economic
first
applied to indexes at software
lower levels
of unadjusted
detail, and thereafter
series directly
with the
X-13ARIMA-SEATS
using
indexes for the latest 5
events (called interventions) or methodology
changes.
These
the adjusted
detail
is aggregated
yield the higher
level seacalendar years.
CPI index
series are
adjusted
using thetomultiplicative
model.
events, which can be one-time occurrences or recurring sonally adjusted indexes. If intervention analysis is indicated,
events that happen at infrequent and irregular intervals, ad- it will be used in adjusting selected lower level indexes prior
Most high-level index series are adjusted by the aggregative method, which is more appropriate
versely affect the estimate of the seasonal component of the to aggregation. For those series that have not been selected
categories whose component indexes show strongly different seasonal patterns. Under the aggre
series.
for seasonal
adjustment,
the original,
unadjusted
dataand
arethereafter the a
method, direct adjustment
is first applied
to indexes
at lower levels
of detail,
Intervention analysis seasonal adjustmentdetail
allows
non-sea- to
used
in the
the higher
aggregation
process. adjusted indexes. If intervention analysis
is aggregated
yield
level seasonally
sonal economic phenomena, such as outliersindicated,
and levelitshifts,
will be used in adjusting selected lower level indexes prior to aggregation. For those
to be factored out of indexes before calculation
of seasonal
Revision.
The seasonal
factorsthe
areoriginal,
updatedunadjusted
annually. data
Eachare used in the
that have
not been selected
for seasonal
adjustment,
adjustment factors. (An outlier is an extreme
value
for
a
paryear
in
February,
BLS
recalculates
and
publishes
revised
seaaggregation process.
ticular month. A level shift is a change or shift in the price sonally adjusted indexes for the previous 5 years. Seasonally
level of a CPI series caused by an event, such
as an excise
adjusted
become
final inEach
the last
and
year ofBLS
revi-recalculates an
Revision.
The tax
seasonal
factorsindexes
are updated
annually.
year
in 5th
February,
increase or oil embargo, occurring over onepublishes
or more months.)
sion.
Seasonal
factors
for
the
past
year
are
used
to
generate
revised seasonally adjusted indexes for the previous 5 years. Seasonally adjusted inde
An index series whose underlying trend has
experienced
a last
seasonally
indexes Seasonal
for the current
become
final in the
and 5th adjusted
year of revision.
factorsyear
for starting
the past with
year are used to ge
sharp and permanent shift will generate seasonally
distorted results
the
release
of
the
January
CPI.
adjusted indexes for the current year starting with the release of the January CPI.
when adjusted using the standard X-13ARIMA-SEATS procedure. X-13ARIMA-SEATS regression techniques are used Calculation of annual and semiannual average
Calculation of annual and semiannual average indexes
to model the distortions and account for them as part of the indexes
seasonal adjustment process. The result is an adjustment
CPI annual
average indexes
annual average
use 12 successive
monthsuse
of 12
CPIsuccessive
values as months of
based on a representation of the series withCPI
the seasonal
pat- indexes
CPI values as
tern emphasized. Intervention analysis seasonal adjustment
12
also makes it possible to account for seasonal shifts, resulting

/ 12 .
12av
t ,0
in a better seasonal adjustment in the periods before and after
t 1
the shift occurred. Not all CPI series are adjusted using interSemiannual average indexes are computed for the first half
vention analysis seasonal adjustment techniques. However,
Semiannual average indexes are computed for the first half of the year (January to June) and for
of the year (January to June) and for the second half of the year
for affected series, the resulting seasonal factors
better
represecond half of the year (July to December) using six successive months of CPI values as
sent the true seasonal pattern than factors calculated without (July to December) using six successive months of CPI values as
these techniques. These seasonal factors are applied to the
6
original unadjusted series. Level shifts and outliers, removed

/6,
6 av
t ,0
in calculating the seasonal factors, remain in the resulting
t 1
seasonally adjusted series.
where the value of each monthly index is real or interpolated,
In recent years, BLS has used intervention analysis sea- depending on availability.42
sonal adjustment for various indexes—gasoline, fuel oil, new
For bimonthly indexes, the intermediate indexes are calvehicles, women’s and girls’ apparel, educational books and culated using a geometric mean of the values in the months
supplies, electricity, utility (piped) gas service, water and adjacent to the one being estimated.
sewerage maintenance, nonalcoholic beverages and beverage materials, and whiskey at home are examples. Series Average prices
are adjusted using intervention analysis techniques when inAverage prices are estimated from CPI data for selected
terventions are clearly identified. After a number of years, food and beverage items, utility (piped) gas, electricity, gasseries may revert to adjustment using standard methods. In oline, automotive diesel fuel, and fuel oil #2 to support the
addition, for some series, intervention analysis is used, and research and analytic needs of CPI data users. (See appendix
the resulting series does not show a clear and stable seasonal 2.) Average food prices are published without tax, while the
pattern. In these cases, the series is not seasonally adjusted.
other average prices are published with tax included.

I

∑I

I ∑I

41
J.A. Buszuwski and S. Scott, "On the use of intervention analysis in
seasonal adjustment,'' Proceedings of the American Statistical Association,
Business and Economics Section. (Alexandria, VA: American Statistical
Association, 1988).

42
To be published, a semiannual average must have at least two noninterpolated index values with sufficient samples. An annual average must
have at least four noninterpolated index values with sufficient samples.

35

For each food and beverage item, the average price for a
specified unit of size (for instance, pound or gallon) is published monthly for the U.S. city average and for the four regions: Northeast, Midwest, South, and West. Metric-equivalent sizes are shown, as well.
Average prices for utility (piped) gas, electricity, and gasoline are published monthly for the U.S. city average, the four
regions, the three population size classes, 10 region/size-class
cross-classifications, and the 14 largest local index areas. For
utility (piped) gas, average prices per therm are published.
For electricity, average prices per kilowatt-hour (kWh) are
published. For gasoline, the average price per gallon is
published. Average prices for commonly available grades
of gasoline are published, as well as the average price
across all grades.
Average prices per gallon for automotive diesel fuel and
fuel oil #2 are published monthly for the U.S. city average,
the four regions, the three population size classes, and 10
of 12 region/size-class cross-classifications.

All eligible prices are converted to a price per normalized quantity. These prices are then used to estimate a
price for a defined fixed quantity. For example, prices for
a variety of package sizes for flour are converted to prices
per ounce. An average price per ounce of flour is then estimated and multiplied by 16 to yield a price per pound, the
published quantity.
The average price for collection period t is estimated as

P

t

=

∑W P P
∑W P
i

it

it

i

it

ib

, 

ib

where Wit is the quote-level expenditure weight of items used
in the average price estimation for the ELI/PSU/replicate.
Dividing the expenditure weight by the base price, Pib,
for a given quote yields an implicit estimate of quantity.
Thus, the average price is, conceptually, a weighted average of prices, Pit, where the weights are quantity amounts.
Imputed prices are used in estimating average prices.

36

Part III. Precision of CPI estimates

An important advantage of probability sampling methods
is that a measure of the sampling error of survey estimates can
be computed directly from the sample data. The CPI sample
design accommodates error estimation by making two or more
selections (replications) of items and outlets within an index
area. Therefore, two or more samples of quotes in each selfrepresenting PSU and one in each non-self-representing PSU
are available. With this structure, which reflects all stages of
the sample design, variance estimation techniques using replicated samples can be used.

areas. Self-representing areas are large metropolitan areas,
such as the Boston metropolitan area, the St. Louis metropolitan area, and the San Francisco metropolitan area. Nonself-representing areas are collections of smaller metropolitan areas. For example, one non-self-representing area is a
collection of 32 small metropolitan areas in the Northeast region (Buffalo, Hartford, Syracuse, Burlington, and others) of
which 8 have been randomly selected to represent the entire
set. Within each of the 38 areas, price data are collected for
211 item categories called item strata. Together, the 211 item
strata cover all consumer purchases.
Multiplying the number of areas by the number of item
Sources of error
We divide the total error into two sources: sampling error strata gives 8,018 (= 38 × 211) different area-item combinaand nonsampling error. Sampling error is the uncertainty in tions for which price indexes need to be calculated. Separate
the CPI caused by the fact that a sample of retail prices is used price indexes are calculated for each one of these 8,018 areato compute the CPI, instead of using the complete universe of item combinations. After calculating all 8,018 of these basicretail prices. The sampling variance attributable to the esti- level indexes, the indexes are then aggregated to form higher
mation of expenditure weights (see chapter 16 for more detail level indexes, using expenditure estimates from the CE as
on consumer expenditure weights) is directly incorporated in their weights.
CPI variances are primarily computed with a stratified
the variance estimates computed for the CPI, due to the fact
that these expenditures are independently estimated for each random groups method, for 1-, 2-, 6- and 12-month percent
replicate. Nonsampling error is the rest of the error, and will changes. From 1978 to 1998, the BLS computed CPI varibe discussed at the end of this section. Incorrect information ances by using a first-order Taylor approximation of the ratio
given by survey respondents and data processing errors are of cost weights. This methodology was replaced, beginning
in January 1998, by the stratified random groups method, in
examples of nonsampling error.
BLS constantly tries to reduce error in the CPI. Vari- which variances are computed separately for certain subsets
ance and sampling error are reduced by using samples of of areas and items, and then those individual variances are
CPI variances
computed
with athe
stratified
random
groups
for 1-, 2-, 6- and 1
combined
to produce
variance
of the
entiremethod,
item–area
retail prices and samples of consumer expenditures
that are
areprimarily
percent
changes.
From
1978
to
1998,
the
BLS
computed
CPI
variances
by
using a first-orde
as large as possible, given resource constraints. The Bureau combination. Subsets of items are formed by the intersection
approximation
of
the
ratio
of
cost
weights.
This
methodology
was
replaced,
beginning
in Janua
has developed a model that optimizes, on a 2-year basis, the of the item category with each 1 of the 8 major groups.
by
the
stratified
random
groups
method,
in
which
variances
are
computed
separately
for
certain
s
allocation of resources. The model indicates the number of
areas and items, and then those individual variances are combined to produce the variance of t
prices that should be observed in each geographic area and Variance estimation using replicates
areaU.S.
combination.
Subsets
of items
are formed
by the
intersection
of A,
theitem
item category wi
item
each item category to minimize the variance
of–the
city
Let
IX(A,I,f,t)
denote
the index
value
for area =
of
the
8
major
groups.
average all-items index. The Bureau reduces nonsampling er- category = I, in month = t, where f indicates that it is the fullror through a series of computerized and professional data sample value, and let IX(A,I,f,t – k) denote the value of the
Variance
estimation
reviews, as well as through continuous survey
process
im- same using
index inreplicates
month = t – k. The uppercase letter A denotes
provements and theoretical research.
a set of areas, such as the Northeast or Midwest region of the
the for
uppercase
letter
strata,
Let IX(A,I,f,t) denotecountry,
the indexand
value
area = A,
itemI denotes
categorya=set
I, of
in item
month
= t, where f indica
is
the
full-sample
value,
and
let
IX(A,I,f,t
–
k)
denote
the
value
of
the
same
index
such
as
all
items
or
all
items
less
food
and
energy,
or
even
a in month = t
Sample design
uppercase
letter
A
denotes
a
set
of
areas,
such
as
the
Northeast
or
Midwest
region
of the country
single
item
stratum.
Also,
let
IX(A,I,r,t)
and
IX(A,I,r,t
–
k)
be
the
Starting in 1978 the CPI’s sample design has accommouppercase
letter
I
denotes
a
set
of
item
strata,
such
as
all
items
or
all
items
less
food
and energy,
corresponding
index
values
for
replicate
=
r.
Most
areas
have
dated variance estimation by using two or more independent
single
item
stratum.
Also,
let
IX(A,I,r,t)
and
IX(A,I,r,t
–
k)
be
the
corresponding
index
values for
two
replicates,
but
some
have
more.
Then
the
full-sample
ksamples of items and outlets in each geographic area. This
=
r.
Most
areas
have
two
replicates,
but
some
have
more.
Then
the
full-sample
k-month
percen
month
percent
change
between
months
t
–
k
and
t
is
computed
allows two or more statistically independent estimates of
between
months
t
–
k
and
t
is
computed
by
dividing
IX(A,I,f,t)
by
IX(A,I,f,t
–
k),
subtractin
by
dividing
IX(A,I,f,t)
by
IX(A,I,f,t
–
k),
subtracting
1,
and
the index to be made. The independent samples are called
multiplying by 100:
replicates, and the set of all observed prices is called the multiplying by 100:
full sample.

 IX( A, I , f , t )
As discussed earlier, BLS calculates CPI indexes for 38
– 1  100 .
PC ( A, I , f , t , t – k )  
geographic areas across the United States. The 38 areas con
 IX( A, I , f , t – k )
sist of 31 self-representing areas and 7 non-self-representing

Every index has an aggregation weight AGGWT(A I, f) or AGGWT(A, I, r) associated with it,
used to combine the index with other indexes to produce indexes for larger geographic areas a
item categories. For example, the aggregation weights are used to combine all 8,018 basic-leve
37
into higher level indexes
such as the U.S. city average-all-items index. The product of an inde
weight is called a cost weight:

IX(A,I,f,t) denote the index value for area = A, item category = I, in month = t, where f indicates that it
e full-sample value, and let IX(A,I,f,t – k) denote the value of the same index in month = t – k. The
ercase letter A denotes a set of areas, such as the Northeast or Midwest region of the country, and the
ercase letter I denotes a set of item strata, such as all items or all items less food and energy, or even a
le item stratum. Also, let IX(A,I,r,t) and IX(A,I,r,t – k) be the corresponding index values for replicate
Most areas have two
replicates,
some
more. Then
the full-sample
k-month
change
Every
indexbuthas
an have
aggregation
weight
AGGWT(A
I, f)percent
or the
211 item categories. (Note: Item aggregation I can be as
ween months t –AGGWT(A,
k and t is computed
by dividing
IX(A,I,f,t)
by IX(A,I,f,t
– k),
subtractingsmall
1, andas one item stratum or may comprise one or more major
I, r) associated
with
it, which
is used to
combine
tiplying by 100:

the index with other indexes to produce indexes for larger groups.)
geographic areas and larger item categories. For example, the
For the stratified random groups method, a replicate per
 IX( A, I , f , t )
  100
– 1all
. basic-level cent change is defined as follows: At each item–area replicate
PC ( A, I ,weights
f , t , t – k )are
  used to combine
aggregation
8,018
 IX( A, I , f , t – k )
indexes into higher level
indexes such as the U.S. city aver- level, the individual full sample cost weight, CW(a, i, f, •) is
age-all-items index. The product of an index and
its stratified
weight israndom
subtracted
from the
full sample
costchange
weightis CW(A,
•), and At each item–
For the
groups method,
a replicate
percent
defined I,
asf,follows:
ry index has an aggregation weight AGGWT(A I, f) or AGGWT(A, I, r) associated with it, which is
area
replicate
level,
the
individual
full
sample
cost
weight,
CW(a,
i,
f,
),
is
subtracted
from
called
a
cost
weight:
a
replicate
cost
weight,
CW(a,
i,
r,
•),
is
added
back
in.
The the full sample
d to combine the index with other indexes to produce indexes for larger geographic areas and larger
cost
CW(A,ForI,the
f,replicate
),
and arandom
replicate
cost
weight,
CW(a,
r,item
), is
addedasback
in. At
The
replicate
percent
stratified
groupschange
method,
a for
replicate
percent
issubset
defined
each
item–
percent
area
= i,a,change
=follows:
i, replicate
m categories. For example, the aggregation weights are used to combine
all weight
8,018 basic-level
indexes
area
replicate
level, the
full sample
cost weight,
CW(a, i,t f,–),k isand
subtracted
from computed
the full sampleas follows:
change
area
a, item
subset
= i,individual
replicate
=–rkbetween
months
t as
is then
higher level indexesCW(A,
such asI,the
city average-all-items
index. The
an =index
and
its
f, t)U.S.
= IX(A,
I, f, t) × AGGWT(A,I,
f,product
t). for of
=
r
between
months
t
and
t
is
then
computed
follows:
cost weight CW(A, I, f, ), and a replicate cost weight, CW(a, i, r, ), is added back in. The replicate percent

ght is called a cost weight:

change for area = a, item subset = i, replicate = r between months t – k and t is then computed as follows:


 item–
CW( A, I , f , t ) – CW(a, i, f , t )  CW(a, i, r , t )
A cost weight is an estimate of the total cost in areaPC= A
for For the stratified
random groups method, a replicate percent change is defined as follows:
At each

 100

 –1 
CW(A, I, f, t) = IX(A, I, f, t) × AGGWT(A,I, f, t)
CW( A, I , f , t ) – CW(a, i, f , t )  CW(a, i, r , t )
S ( a, i , r , t , t – k )  
full
area
replicate
level,
the
individual
full
sample
cost
weight,
CW(a,
i,
f,
),
is
subtracted
from
the
sample


PC (a, i, rCW
, t , t – (kA
) ,I, f , t – k ) – CW(a, i, f , t – k )  CW(a, i, r , t ––1 k
consumption of item category I in month t. A replicate cost
 )100 
f , t – ki,) r,
 CW
, I , f , t – kcost
) – CW
(a, i,CW(a,
, i, r , t back
– k ) in. The
cost weight CW(A, I, f, ),
and(aAreplicate
weight,
), is(aadded
 CW
 replicate percent
weight ofwould
becost
indexed
r instead
of f.ofBecause
the agchange for area = a, item subset = i, replicate = r between months t – k and t is then computed as follows:
ost weight is an estimate
the total
in areawith
= A for
consumption
item category
I in month
t. A
areas. For non-self-representing
areas,
another replicate
percent
changechange
for area =for
a, area = a,
weights
not indexed
by time
(except
across
icate cost weight gregation
would be indexed
withare
r instead
of f. Because
the aggregation
weightspivarefor
notself-representing
indexed
for self-representing
areas.
For non-self-representing
areas,
another
replicate
percent


item category = I, replicate = r between
months
is computed
t ) –tCW
 CW(a, i, r , t )
CW( At, I–, kf ,and
(a, i, f , t ) as
for self-representing
For
non-self-representing
areas,–1an  100
a, i, r , t , t – k )months
  areas.
PCr (between
ime (except across
pivot months;
see section
titled
“Bridging across
across item
pivot
months”),
preceding
= the
I, replicate
=
t
–
k
and
t
is
computed
as
ot months;
see section
titled
“Bridging
pivotcategory
months”),
 CW( A, I , f , t – k ) – CW(a, i, f , t – k )  CW(a, i, r , t – k )
ent change formula
equivalent to
other replicate
percent
change for area = a, item category
= I,
theispreceding
percent change formula is equivalent to


CW( A, I , f , t ) – CW(a, I , f , t )  CW(a, I , r , t )
PC (a, I , r , t , t – k )  
– 1   100 ,
for
self-representing
areas,
for area = a,
rCW
months
t(another
f ,–t –non-self-representing
k ) – (CW
kCW
I,,rr ,,replicate
tt–) k ) percent
)  CW
(,aI,computed
as change
replicate

Iareas.
a, I(a,, Ift,,–ft,)tk–and
ais
CW=(A
,between
,( Af,,It, )For
CW
S

S

N

 category = I, replicate = r between months t – k and t is computed as
– 1   100 ,
 PC (a, I , r , t , t – k )  item
 CW( A, I , f , t )
PC( A, I , f , t , t – k )  
– 1  100 ,N
 CW( A, I , f , t – k ) – CW(a, I , f , t – k )  CW(a, I , r , t – k )

where
CW
(
,
,
,
–
)
A
I
f
t
k




CW( A, I , f , t ) – CW(a, I , f , t )  CW(a, I , r , t )
PC (a, I , r , t , t – k )  
– 1   100 ,
N

ch is equivalent to
which is equivalent to

 CW( A, I , f , t – k ) – CW(a, I , f , t – k )  CW(a, I , r , t – k )


CW( A, I ,,)  ∑∑CW(a, i,,) .

where

a∈A

where




i∈I

where
The symbol α  A means that the sum is over all basic-level areas within area = A, and the symbol I  I
CW
( A,groups
I ,,CW
)within
 (∑∑
(a=CW
, iI.,(,a,)i,.,) .
A, Iitem
,,)category
CW
means that the sum is over
major
∑∑

∑∑
∑∑
∑∑



CW(a, i, f , t )


a∈A i∈I
 ,
PC( A, I , f , t , t–k )    a∈A i∈I CW ( a, i, f , t –) 1  100

The variance is computed with the following stratified random groups variance estimation formula:
a∈CW(
Ai∈I a, i, f , t – k )


The symbol α  A means that the sum is over all basic-level areas within area = A, and the symbol I  I
PC ( A, I , f , t , t –k )  
– 1  100 ,
 a∈A i∈I CW( a, i, f , t – k ) The symbol α  A means
means that
that the
sum
is over
within item category
I.
the
sum
is major
over groups
all basic-level
areas =within
area = A, and the symbol I  I
α  A
1

V
[
PC
(
A
,
I
,
f
,
t
,
t
–
k
)]

PC
(
a
,
i
,
r
,
t
,
t
– k )is
– PC
( A, I ,all
f , t , tbasic– k )
∑∑
∑
 that the
 a∈ Ai∈I
means
sum
is
over
major
groups
within
item
category
=
I.
symbol
the sum
over

Rmeans
( R – 1stratified
) that random
The The
variance
is computed with the following
groups
variance
estimation
formula:
i  I 
ause cost weights are additive from the lowest area-item level up to the highest U.S. city average-alllevel areas within area A, and the symbol
means that the
ms level. The lowercase letter a denotes one of the 38 basic-level areasThe
included
in area
= A, and the
R
1
variance
is computed
with
the
random
groups
estimation
V [over
PC
( A, following
Imajor
, f ,1t , t – kgroups
)] stratified
(a, i, r ,variance
t , t – kI.
) – PC
( A, I2 , f , t , t – kformula:
)
because
weights
are additive
from
lowest
is
item
category
∑∑
∑PC
ercase letter i denotes
one ofcost
the 211
item categories.
(Note:
Item the
aggregation
Iarea-item
can be as smallsum
as one
PCNwithin
 ∑
(R
a, (IR, r ,–t1,)t – k ) – PC ( A, I , f , t , t – k )  ,
∑
R
(
R
–
1
)
up one
to the
highest
city average-all-items level. The
Thea⊂variance
is r computed
with the following stratified
a
a
A∩N
1
m stratum or may level
comprise
or more
majorU.S.
groups.)
a∈A

∑∑

i∈I

Ra

2

S

i⊂I a⊂A∩S

a

r 1

a

Ra

a

2

S

i⊂I a⊂A∩S

a

r 1

a

Ra

R
2
lowercase letter a denotes one of the 38 basic-level areas
in- I , frandom
variance
formula:
1 1 estimation
PC
V [PC( A,where
,Stand
, t –Nkare
)]groups
 ∑∑
– kareas
, Ik,)fgeographic
sets
of all self-representing
and
non-self-representing
PC
2, t, , t – k ) 
∑
the
, t, i–, rk,)t ,–t PC
()A–, IPC
,inf the
, t(,A
tCPI’s
–
∑
∑
S ,(ta
N ( a, I , r
R
(
R
–
1
)
R
(
R
) r the
cluded in area = A, and the lowercase letter i denotes one ofsample, respectively;a⊂iand
aSS and
a a A a
r–
1 1are
N
⊂AI ∩aNA
⊂A
∩
1 sets of all self-representing and nonself-representing
a

areas within area = A. The number Ra is the number of replicates in area = a.



where S and N are the sets of all self-representing and non-self-representing areas in the CPI’s geographic
R

sample,
respectively;
andlonger
A  Sspans
and A
 Nthan
are one
the major
sets ofgroup,
all self-representing
and nonself-representing
49
When
the item
category I ano
more
the preceding formula
reduces
2 to
areas within area = A. The number Ra is the number of replicates in area = a.

∑ R ( R1 – 1) ∑PC

(a, I , r , t , t – k ) – PC ( A, I , f , t , t – k )  ,
∑R ( R – 1) ∑

N
R
1
a
a
a⊂A∩N
r 1
2
WhenV the
than
reduces to
[PCitem
( A, Icategory
, f , t , t – kI)]no longer spans morePC
(aone
, I , r ,major
t , t – kgroup,
) – PC the
( A, Ipreceding
, f , t , t – k formula
) .
a⊂A

a

a

r 1

a

1
where S and N are the sets of Vall[PC
self-representing
areas
geographic
( A, I , f , t , t – k )]  ∑and non-self-representing
( A, I ,in
f , tthe
, t – kCPI’s
) .
∑PC (a, I , r, t, t – k ) – PC
Variance
estimation
withoutPrice
replicates
R Index,
( R – 1)
Table 3. Response rates for commodities
and
services
for
the
Consumer
Allself-representing
Urban Consumers
sample, respectively; and A  S and A  N are the sets of all
and nonself-representing
areas2014
within area =
A.Variance
The number
is without
thespecial
number
replicates
area
a. the item stratum level
Ra for
BLS
computes
index
series
85
(SRC)of
item
categories, in
which
are=below
(CPI-U), U.S. city average, by major group,
estimation
replicates
Ra

a⊂A

Commodities and
services
Outlets
Total quotes

Eligible
When the
305,184
1,180,932

a

a

2

r 1

and thus do not have accompanying replicate index values. (CE weights are produced only down to the
item-stratum level in each
index area.) The stratified random in
groups methodology
requiresin
a replicate
Percent
Percent
BLS computes index
series for 85 special (SRC)Used
item categories, which are
below the item
stratum level
Collected
structure.
So,
for
these
SRC
items more
(such asthan
butterone
orindex
pork
or
newgroup,
cars),
anthe
alternative
varianceonly
estimation
item
category
I no
spans
major
preceding
and thus
dolonger
not have
accompanying
replicate
values.
(CE weights
are producedformula
downreduces
to the
collected
estimation
estimation
method
is
needed.
Given the availability
(at
the
regional
and
higher
area
levels)
of
independent
estimates
item-stratum level in each index area.) The stratified random groups methodology requires a replicate
for these
SRC
items,
the
jackknife
variance
estimation
methodology
can
be
employed.
Each
area
full
structure. So, for these SRC items (such as butter or pork or new cars), an alternative variance estimation

to

284,730
93.3
271,931
89.1
R (at the regional and higher area levels) of independent estimates
method is needed. Given the availability
1
2
jackknife variance
can (be
PCestimation
V [PC970,546
( A, I for
, f ,these
t , t –SRC
k )] items,
 the82.2
(a944,414
, I , r ,methodology
t , t – k ) – PC
A80.0
,employed.
I , f , t , t –Each
k ) area
. full
50
R
(
R
–
1
)
a⊂A
r 1
a
a
418,060
89.8
410,615
88.2
a

∑

∑

Food and beverages

465,398

Housing (less shelter)

146,466

Apparel

135,950
75,102
55.2
70,700
52.0
BLS computes index
series for 85 special
(SRC) item categories,
149,721
134,956
90.1
131,796which are below
88.0the item stratum level
and thus do not have accompanying replicate index values. (CE weights are produced only down to the
76,631
39,534
51.6 requires a replicate
item-stratum level40,847
in each index area.)53.3
The stratified random
groups methodology
structure. So, for 68,427
these SRC items (such81.2
as butter or pork or65,064
new cars), an alternative
84,227
77.2 variance estimation
method is needed. Given the availability (at the regional and higher area levels) of independent estimates
78,164
68,872
88.1 estimation methodology
67,039
for these SRC items,
the jackknife variance
can be 85.8
employed. Each area full

Transportation
Medical care
Recreation
Education and
communication
Other goods and services

125,300

85.5

Variance estimation without replicates

44,375

38,982

Source: U.S. Bureau of Labor Statistics.

38

87.8

121,502

38,164

83.0

86.0

50

50

V [PC( A, I , f , t , t – k )] =

V [PC( A, I , f , t , t – k )] 

∑∑
i⊂I a⊂A∩S



Ra

∑ R ( R1 – 1) ∑PC

a⊂A∩N

Ra

1
∑PC S (a, i, r , t , t – k ) – PC ( A, I , f , t , t – k )2
R a ( R a – 1) r 1

a

a

(a, I , r , t , t – k ) – PC ( A, I , f , t , t – k )  ,  

Ra

∑R ( R1 – 1) ∑(PC (a, I , r , t , t – k ) – PC ( A, I , f , t, t – k ))

2

a∈A

a

a

r =1

2

N

r 1

where S and N are the sets of all self-representing and nonself-representing areas in the CPI’s geographic sample, respectively; and A ∩ S and A ∩ N are the sets of all self-representing and nonself-representing areas within area = A. The
number is the number of replicates in area = a.
When the item category IV. Design
The ATUS sample is a stratified three-stage sample drawn
from the population of households that participated in the
eighth interview of the CPS. In the first stage of selection, the
CPS oversample in the less populous states is reduced. Unlike
the CPS, the ATUS does not have a state reliability requirement. To improve the efficiency of the national estimates derived from the survey, the ATUS sample is chosen from the
population of households that responded to the CPS. In that
way, the sample is distributed across states so that it is representative of the proportion of the national population that
each state represents. The second stage of selection stratifies
households according to race and ethnicity, the presence and
age of children in the household, and the number of adults
in adults-only households. Sampling rates vary within each
stratum. Eligible households with a Hispanic or non-Hispanic
Black householder are oversampled to improve the reliability
of time-use data for these demographic groups. To ensure adequate measures of childcare, households with children are
also oversampled; equivalently, households without children
are undersampled. In the third stage of selection, an eligible
person from each household selected in the second stage is
randomly selected to be the designated person for the ATUS.
Civilian household members at least 15 years of age are eligible. All eligible persons within a sample household have
the same probability of being selected the designated person.
The monthly sample is divided into four randomly selected panels, one for each week of the month. To ensure good
measures of time use on weekdays and weekend days, the
sample is also split evenly between weekdays and weekend
days. Ten percent of the sample is allocated to report about
each weekday, and 25 percent of the sample is allocated to report about each weekend day. Designated persons from each
household are randomly assigned a day of the week about
which to report.
For more information on the ATUS survey design, see
chapter 3 of the User’s Guide.
no longer spans more than one major group, the preceding
formula reduces to

Variance estimation without replicates
BLS computes index series for 85 special (SRC) item categories, which are below the item stratum level and thus do not
have accompanying replicate index values. (CE weights are
produced only down to the item-stratum level in each index
area.) The stratified random groups methodology requires a
replicate structure. So, for these SRC items (such as butter or
pork or new cars), an alternative variance estimation method
is needed. Given the availability (at the regional and higher
area levels) of independent estimates for these SRC items,
the jackknife variance estimation methodology can be employed. Each area full sample cost weight can be subtracted
from the all-area full sample cost weight to provide a jackknife replicate estimate. By taking the ratio of these replicate
cost weight estimates at times t and t – k, subtracting 1, and
multiplying by 100, one obtains the required jackknife replicate percent change value. (For the U.S. city average special
item estimates, there are 38 independent index areas, and so
38 jackknife replicate estimates to work with.)
The full-sample percent change is computed as before (except that item category = I here is smaller even than an item
stratum):

 CW( A, I , f , t )

PC ( A, I , f , t , t – k )  
– 1   100 .  
 CW( A, I , f , t - k )

The jackknife replicate percent change is computed as follows:


CW ( A, I , f , t ) – CW (a, I , f , t )
PC ( A – a, I , r , t , t – k )  
– 1  100 .  
 CW ( A, I , f , t – k ) – CW (a, I , f , t – k )


Then the variance for the k-month percent change is computed in the usual jackknife form:
V [PC ( A, I , f , t , t – k )] =

N A –1
∑(PC ( A – a, I , r, t, t – k ) – PC ( A, I , f , t, t – k ))2 .  
N A a∈A

Bridging across pivot months
Every 2 years, BLS updates its set of aggregation index
weights based on CE data collected from the t – 2 and t – 3
years. In January 2012, BLS replaced its old set of aggregation weights with a new 2-year set of weights from expenditure data collected in 2009–2010. In January 2014, this set
of weights was replaced by an updated set of weights from
expenditure data collected in 2011–2012, and so on.
Whenever the variance estimates cross the pivot month (as
they did in December 2011 and December 2013), a bridging
factor has to be introduced into any variance calculation that
crosses the pivot month anywhere between t and t – k months
(including month t – k, but not including month t). The bridg-

39

.

ing factor is then applied directly to the individual ratio of
cost weights, for both full-sample and replicate values, inside
each percent change calculation. Thus, in its most general
form,

Table 4. Response rates for shelter for the Consumer
Price Index, All Urban Consumers (CPI-U), U.S. city
average, 2014
Shelter

 CW (,,, t )

CW (,,, old)
PC (,,, t , t  k )  

 1  100  
CW
(

,

,

,
t

k
)
CW
(

,

,

,
new
)



Number of total
units

for every combination of area and item, and for full-sample
and replicate values, with the bridging factor defaulting to 1
whenever not applicable.
The bridging factor, CW(•,•,•, old)/CW((•,•,•, new), essentially allows the old aggregation weight in the bridge’s numerator to cancel out the old aggregation weight in the t – k
cost weight, while the new aggregation weight in the bridge’s
denominator cancels out the new aggregation weight in the
t cost weight, leaving IX(•,•,•,t)/X(•,•,•,t – k) free to move
this level’s percent change without disruption. Note that
IX(•,•,•,old)/IX(•,•,•,new) = 1 at all times.

Percentage of
eligible units

Eligible

Collected,
data
reported

No data at
collection
or other

99,383

72,966

26,417

100

73.4

26.6

Source: U.S. Bureau of Labor Statistics.

Response error results from the collection and use in estimation of incorrect, inconsistent, or incomplete data. Response
error may arise because of the collection of data from inappropriate respondents, respondent memory or recall errors,
deliberate distortion of responses, interviewer effects, misrecording of responses, pricing of wrong items, misunderstanding or misapplication of data collection procedures, or
misunderstanding of the survey needs and/or lack of cooperation from respondents. The pricing methodology in the
commodities and services component of the CPI allows the
previous period’s price to be available at the time of collection. This dependent pricing methodology is believed to reduce response variance for measuring change, but may cause
response bias and lag. The housing component of the CPI
employs an independent pricing methodology specifically to
avoid potential response bias.

Nonsampling error
CPI estimates are subject to nonsampling error as well
as sampling error. Surveys involve many operations, all of
which are potential sources of nonsampling error. The errors
arise from the survey process, regardless of whether the data
are collected from the entire universe or from a sample of
the population. The most general categories of nonsampling
error are coverage error, nonresponse error, response error,
processing error, and estimation error.

Processing error arises from incorrect editing, coding, and
data transfer. Price data are collected by CADC. Automated
data checking ensures that only correct data types are collected; other automated logic checks remove all redundant
question patterns, and the instrument informs the field staff
when not all required data have been collected. In both systems, errors also can result from software problems in the
computer processing that cause correctly entered data to be
lost. Computer screening and professional review of the data
provide checks on processing accuracy. Occasional studies
of these processing errors in the CPI have shown them to be
extremely small.

Coverage error in an estimate results from the omission of
part of the target population (undercoverage) or the inclusion
of units from outside of the target population (overcoverage).
Coverage errors result from the omission of cities, households, outlets, and items that are part of the target populations from the relevant sampling frames or from their double-counting or improper inclusion in the frames. A potential
source of coverage error is the time lag between the TPOPS
and the initiation of price collection for commodities and services at sampled outlets. Because of the time lag, the products offered by the outlet at the time pricing is initiated may
not coincide with the set from which the TPOPS respondents
were purchasing.

Estimation error results when the survey process does not
accurately measure what is intended. Such errors may be
conceptual or procedural in nature, arising from a misunderstanding of the underlying survey measurement concepts or a
misapplication of rules and procedures.

Nonresponse error results when data are not collected for
some sampled units because of the failure to interview households or outlets. This can occur when selected households
and outlets cannot be contacted or refuse to participate in the
survey. Nonresponse rates during monthly pricing for the CPI
C&S and housing surveys are shown in tables 3 (page 38) and
4 (page 40).

Substitutions and adjustments for quality change in the items
priced for the CPI are possible sources of estimation error due
to procedural difficulties. Ideally, CPI data collection forms

40

and procedures would yield all information necessary to determine or explain price and quality differences for all items
defined within an ELI. Because such perfect information is
not available, BLS economists supplement directly collected
data with secondary data. Estimation error will result, if the
BLS adjustment process—which may require significant
judgment or lack key data—is misapplied, or if it consistently
overestimates or underestimates quality change for particular
kinds of items.
The effect of the aging of housing units is an example of
potential estimation error, which is similar to the issue of
quality change in commodities and services. Until 1988, BLS
did not adjust for the slow depreciation of houses and apartments over time. BLS research indicates that annual changes
for the residential rent and owners’ equivalent rent indexes
would have been 0.1 to 0.2 percent larger if some type of aging adjustment had been included.
The total nonsampling error of the CPI results from errors
in the type of data collected, the methods of collection, the
data processing routines, and the estimation processes. The
cumulative nonsampling error can be much greater than the
sampling error.

in estimation divided by the sum of (1) the number of eligible
sample units and (2) the number of sample units with eligibility
not determined.
Commodities and services items (any except rent and owner’s equivalent rent) are further broken down into outlets and
quotes. An “outlet” is a generic term used to describe places
where prices are collected. A “quote” is a specific item to be
priced in a specific outlet. There may be from 1 to more than
50 quotes priced in an outlet. Table 3 shows the relatively low
percentages of quotes reported collected and used in estimation
for apparel. Low rates for these items largely can be attributed
to the design of the apparel sample. Because apparel items are
commonly in stores only at certain times of the year, most of the
apparel sample is doubled, with each half of the sample designated for pricing during part of the year. Thus, at any particular time of the year many apparel quotes, although eligible, are
designated “out of season,” and prices are not collected. For additional information, see the earlier section on seasonal items.
The response rates for housing (shelter) shown in table 4 include categories for renters only. Owners are out of scope for
the CPI housing sample. A unit qualifies as renter if its tenure
status is known either by previous knowledge or is collected in
the current interview period. The response rates at the data collection phase for housing (shelter) are separated into three categories. If usable information is obtained, the unit is designated
eligible and data reported. If the assigned unit is located but is
unoccupied, the unit is designated “eligible, found vacant.” In
instances where the unit is eligible but no data are available (for
example refusals), the unit is designated “eligible, other.” The
response rates at the estimation phase are units that are used in
either rent or REQ.

Response rates
Response rates are calculated for the CPI at the data collection phase and at the index estimation phase for ongoing pricing. The response rate at the data collection phase is the number
of responding sample units divided by the sum of (1) the number of eligible sample units and (2) the number of sample units
with eligibility not determined. A sample unit is eligible if it
belongs to the defined target population and responses should
be collected from the unit for one or more items. The response
rate at estimation is defined as the number of sample units used

Technical references
Aizcorbe, Ana M., and Patrick C. Jackman. “The commodity
substitution effect in CPI data, 1982–91.” Monthly Labor Review, December 1993, pp. 25–33. http://www.
bls.gov/opub/mlr/1993/12/art3full.pdf.
Cage, Robert. “New methodology for selecting outlet samples.” Monthly Labor Review, December 1996, pp.
49–61. http://www.bls.gov/opub/mlr/1996/12/art7full.pdf.
Cardenas, Elaine M. “Revision of the CPI hospital services
component.” Monthly Labor Review, December 1996,
pp, 40–48. http://www.bls.gov/opub/mlr/1996/12/

art6full.pdf. The Conference Board. Measuring
Prices in a Dynamic Economy: Re-examining the CPI.
New York: The Conference Board, 1999.
Dalton, Kenneth V., John S. Greenlees, and Kenneth J. Stewart. “Incorporating a geometric mean formula into the
CPI.” Monthly Labor Review, October 1998, pp. 3–7.
http://www.bls.gov/opub/mlr/1998/10/art1full.pdf.
Diewert, W. E., and A. O. Nakamura, eds. Essays in index
number theory, vol. 1. Amsterdam: North-Holland
Publishing Co., 1993.

41

Final report of the advisory commission to study the Consumer Price Index. U.S. Senate, Committee on
Finance, 104th Cong., 2d sess., 1996.

ence, “Price indices and the measurement of quality
change,” University of Mannheim, Germany, April
25–26, 2002.

Fixler, Dennis, Charles Fortuna, John Greenlees, and Walter Lane. “The use of hedonic regressions to handle
quality change: the experience of the U.S. CPI.” Paper presented at the Fifth Meeting of the International
Working Group on Price Indices, Reykjavik, Iceland,
August. 26, 1999.

Moulton, Brent R. “Basic components of the CPI: estimation of price changes.” Monthly Labor Review, December 1993, pp. 13–24. http://www.bls.gov/opub/
mlr/1993/12/art2full.pdf.
Moulton, Brent R., and Karin E. Moses. “Addressing the
quality exchange issue in the Consumer Price Index.”
Brookings Papers on Economic Activity, Washington:
The Brookings Institution, 1997.

Gillingham, Robert. “A conceptual framework for the Consumer Price Index.” Proceedings of the American
Statistical Association, Business and Economics
Section. Alexandria, VA: American Statistical Association, 1974, pp. 46–52.

Pollak, Robert A. The theory of the cost-of-living index. Oxford, U.K: Oxford University Press, 1989.
Ptacek, Frank, and Robert M. Baskin. “Revision of the CPI
housing sample and estimators.” Monthly Labor Review, December 1996, pp. 31–39. http://www.bls.gov/
opub/mlr/1996/12/art5full.pdf.

Gillingham, Robert F. “Measuring the cost of shelter for
homeowners: theoretical and empirical considerations.” The Review of Economics and Statistics, vol.
65, no. 2, May 1983, pp. 254–265.

Schultze, Charles L., and Christopher Mackie, eds. At What
Price? Conceptualizing and measuring cost-of-living
and price indexes. National Research Council Panel
on Conceptual, Measurement, and Other Statistical
Issues in Developing Cost-of-Living Indexes, Committee on National Statistics, Commission on Behavioral and Social Sciences and Education. Washington:
National Academies Press, 2002.

Grandits, Steven. “Publication strategy for the 1998 revised
Consumer Price Index.” Monthly Labor Review, December 1996, pp. 26–30. http://www.bls.gov/opub/
mlr/1996/12/art4full.pdf.
Greenlees, John S., and Charles C. Mason. “Overview of the
1998 revision of the Consumer Price Index.” Monthly
Labor Review, December 1996, pp. 3–9. http://www.
bls.gov/opub/mlr/1996/12/art1full.pdf.

Shoemaker, Owen J., and William H. Johnson.aaa “Estimation of variance components for the U.S. Consumer
Price Index.” Proceedings of the American Statistical
Association, Government Statistics and Social Statistics sections. Alexandria, VA: American Statistical
Association, 1999, pp. 298–303.

Greenlees, John S., and Robert B. McClelland. “Addressing
misconceptions about the Consumer Price Index.”
Monthly Labor Review, August 2008, pp. 3–17. http://
www.bls.gov/opub/mlr/2008/08/art1full.pdf.
Lane, Walter F. “Changing the item structure of the Consumer Price Index.” Monthly Labor Review, December 1996, pp. 18–25. http://www.bls.gov/opub/
mlr/1996/12/art3full.pdf.

Stigler, George, ed. “The price statistics of the federal government.” In Report to the Office of Statistical Standards, Bureau of the Budget. New York: National Bureau of Economic Research, 1961.

Leaver, S. G., and others. “Sample redesign for the introduction of the telephone point-of-purchase survey frames
in
the commodities and services component of the U.S. Consumer Price Index.” Proceedings of the American Statistical Association, Government Statistics and Social
Statistics sections. Alexandria, VA: American Statistical Association, 1999, pp. 292–297.

U.S. Bureau of Labor Statistics. “Measurement Issues in
the Consumer Price Index,” Statistical Journal of the
United Nations ECE, 15 (1998), pp. 1–36.
Williams, Janet L. “The redesign of the CPI geographic
sample.” Monthly Labor Review, December 1996, pp.
10–17. http://www.bls.gov/opub/mlr/1996/12/art2full.pdf.

Leaver, Sylvia, and Richard Valliant. “Statistical problems in
estimating the U.S. Consumer Price Index.” In Brenda G. Cox et al., eds. Business Survey Methods. New
York: John Wiley & Sons, Inc., 1995.

Williams, Janet L., Eugene F. Brown, and Gary R. Zion. “The
challenge of redesigning the Consumer Price Index
area sample.” Proceedings of the American Statistical
Association, Survey Research Methods Section. Alexandria, VA: American Statistical Association, 2003,
pp. 200–205.

Liegey, Paul. “Hedonic quality adjustments in the U.S. CPI:
a statistical agency perspective.” Paper presented
at Center for European Economic Research confer-

42

Appendix 1. List of published indexes
CPI-U and CPI-W Indexes Published at the U.S. City Average (National) Level
All items
Food and beverages
Food
Food at home
Cereals and bakery products
Cereals and cereal products
Flour and prepared flour mixes
Breakfast cereal
Rice, pasta, cornmeal
Rice*
Bakery products
Bread
White bread*
Bread other than white*
Fresh biscuits, rolls, muffins
Cakes, cupcakes, and cookies
Fresh cakes and cupcakes*
Cookies*
Other bakery products
Fresh sweetrolls, coffeecakes, doughnuts*
Crackers, bread, and cracker products*
Frozen and refrigerated bakery products, pies, tarts, turnovers*
Meats, poultry, fish, and eggs
Meats, poultry, and fish
Meats
Beef and veal
Uncooked ground beef
Uncooked beef roasts
Uncooked beef steaks
Uncooked other beef and veal
Pork
Bacon, breakfast sausage, and related products
Bacon and related products*
Breakfast sausage and related products*
Ham
Ham, excluding canned*
Pork chops
Other pork, including roasts and picnics
Other meats
Frankfurters*
Lunchmeats*
Lamb and organ meats*
Lamb and mutton*
Poultry
Chicken
Fresh whole chicken*
Fresh and frozen chicken parts*
Other poultry, including turkey
Fish and seafood
Fresh fish and seafood
Processed fish and seafood
Shelf stable fish and seafood*
Frozen fish and seafood*

43

Appendix 1. List of published indexes—continued

Eggs

Dairy and related products
Milk
Fresh whole milk*
Fresh milk other than whole*
Cheese and related products
Ice cream and related products
Other dairy and related products
Fruits and vegetables
Fresh fruits and vegetables
Fresh fruits
Apples
Bananas
Citrus fruits
Oranges, including tangerines*
Other fresh fruits
Fresh vegetables
Potatoes
Lettuce
Tomatoes
Other fresh vegetables
Processed fruits and vegetables
Canned fruits and vegetables
Canned fruits*
Canned vegetables*
Frozen fruits and vegetables
Frozen vegetables*
Other processed fruits and vegetables, including dried
Dried beans, peas, and lentils*
Nonalcoholic beverages and beverage materials
Juices and nonalcoholic drinks
Carbonated drinks
Frozen noncarbonated juices and drinks
Nonfrozen noncarbonated juices and drinks
Beverage materials, including coffee and tea
Coffee
Roasted coffee*
Instant and freeze-dried coffee*
Other beverage materials, including tea
Other food at home
Sugar and sweets
Sugar and artificial sweeteners
Candy and chewing gum
Other sweets
Fats and oils
Butter and margarine
Butter*
Margarine*
Salad dressing
Other fats and oils, including peanut butter
Peanut butter*
Other foods
Soups
Frozen and freeze dried prepared foods
Snacks

44

Appendix 1. List of published indexes—continued

Spices, seasonings, condiments, sauces

Salt and other seasonings and spices*
Olives, pickles, relishes*
Other condiments
Baby food
Other miscellaneous foods
Prepared salads
Food away from home
Full service meals and snacks
Limited service meals and snacks
Food at employee sites and schools
Food at elementary and secondary schools
Food from vending machines and mobile vendors
Other food away from home
Alcoholic beverages
Alcoholic beverages at home
Beer, ale, and other malt beverages at home
Distilled spirits at home
Whiskey at home*
Distilled spirits, excluding whiskey, at home*
Wine at home
Alcoholic beverages away from home
Beer, ale, and other malt beverages away from home*
Wine away from home*
Distilled spirits away from home*

Housing
Shelter
Rent of primary residence
Lodging away from home
Housing at school, excluding board
Other lodging away from home, including hotels and motels
Owners’ equivalent rent of residences
Owner’s equivalent rent of primary residence
Tenants’ and household insurance
Fuels and utilities
Household energy
Fuel oil and other household fuels
Fuel oil
Propane, kerosene, firewood
Energy services
Electricity
Utility (piped) gas service
Water and sewer and trash collection services
Water and sewerage maintenance
Garbage and trash collection
Household furnishings and operations
Window and floor coverings and other linens
Floor coverings
Window coverings
Other linens
Furniture and bedding
Bedroom furniture
Living room, kitchen, and dining room furniture
Other furniture
Infants’ furniture
45

Appendix 1. List of published indexes—continued

Appliances
Major appliances
Laundry equipment*
Other appliances
Other household equipment and furnishings
Clocks, lamps, and decorator items
Indoor plants and flowers
Dishes and flatware
Nonelectric cookware and tableware
Tools, hardware, outdoor equipment and supplies
Tools, hardware and supplies
Outdoor equipment and supplies
Housekeeping supplies
Household cleaning products
Household paper products
Miscellaneous household products
Household operations
Domestic services
Gardening and lawn care services
Moving, storage, freight expense
Repair of household items

Apparel
Men’s and boys’ apparel
Men’s apparel
Men’s suits, sport coats, and outerwear
Men’s furnishings
Men’s shirts and sweaters
Men’s pants and shorts
Boys’ apparel
Women’s and girls’ apparel
Women’s apparel
Women’s outerwear
Women’s dresses
Women’s suits and separates
Women’s underwear, nightwear, sportswear and accessories
Girls’ apparel
Footwear
Men’s footwear
Boys’ and girls’ footwear
Women’s footwear
Infants’ and toddlers’ apparel
Jewelry and watches
Watches
Jewelry

Transportation
Private transportation
New and used motor vehicles
New vehicles
New cars and trucks*
New cars*
New trucks*
Used cars and trucks
Leased cars and trucks
Car and truck rental
46

Appendix 1. List of published indexes—continued

Motor fuel
Gasoline (all types)
Gasoline, unleaded regular*
Gasoline, unleaded midgrade*
Gasoline, unleaded premium*
Other motor fuels
Motor vehicle parts and equipment
Tires
Vehicle accessories other than tires
Vehicle parts and equipment other than tires*
Motor oil, coolant, and fluids*
Motor vehicle maintenance and repair
Motor vehicle body work
Motor vehicle maintenance and servicing
Motor vehicle repair
Motor vehicle insurance
Motor vehicle fees
State and local registration and license fees
Parking and other fees
Parking fees and tolls*
Automobile service clubs*
Public transportation
Airline fare
Other intercity transportation
Intercity bus fare*
Intercity train fare*
Ship fare*
Intracity transportation
Intracity mass transit*

Medical care
Medical care commodities
Medicinal drugs
Prescription drugs
Nonprescription drugs
Medical equipment and supplies
Medical care services
Professional services
Physicians’ services
Dental services
Eyeglasses and eye care
Services by other medical professionals
Hospital and related services
Hospital services
Inpatient hospital services*
Outpatient hospital services*
Nursing homes and adult day services
Care of invalids and elderly at home
Health insurance

Recreation
Video and audio
Televisions
Cable and satellite television and radio service
Other video equipment

47

Appendix 1. List of published indexes—continued

Video discs and other media, including rental of video and audio

Video discs and other media
Rental of video or audio discs and other media
Audio equipment
Audio discs, tapes, and other media
Pets, pet products and services
Pets and pet products
Pet food*
Purchase of pets, pet supplies, accessories*
Pet services, including veterinary
Pet services*
Veterinarian services*
Sporting goods
Sports vehicles, including bicycles
Sports equipment
Photography
Photographic equipment and supplies
Film and photographic supplies*
Photographic equipment*
Photographers and film processing
Photographer fees*
Film processing*
Other recreational goods
Toys
Toys, games, hobbies and playground equipment*
Sewing machines, fabric and supplies
Music instruments and accessories
Other recreation services
Club dues and fees for participant sports and group exercises
Admissions
Admission to movies, theaters, and concerts*
Admission to sporting events*
Fees for lessons or instructions
Recreational reading materials
Newspapers and magazines
Recreational books

Education and communication
Education
Educational books and supplies
College textbooks*
Tuition, other school fees, and childcare
College tuition and fees
Elementary and high school tuition and fees
Child care and nursery school
Technical and business school tuition and fees
Communication
Postage and delivery services
Postage
Delivery services
Information and information processing
Telephone services
Wireless telephone services
Land-line telephone services

48

Appendix 1. List of published indexes—continued

Information technology, hardware, and services

Personal computers and peripheral equipment
Computer software and accessories
Internet services and electronic information providers
Telephone hardware, calculators, and other consumer information items
Other goods and services
Tobacco and smoking products
Cigarettes
Tobacco products other than cigarettes
Personal care
Personal care products
Hair, dental, shaving, and miscellaneous personal care products
Cosmetics, perfume, bath, nail preparations and implements
Personal care services
Haircuts and other personal care services
Miscellaneous personal services
Legal services
Funeral expenses
Laundry and dry cleaning services
Apparel services other than laundry and dry cleaning
Financial services
Checking account and other bank services*
Tax return preparation and other accounting fees*
Miscellaneous personal goods
Stationery, stationery supplies, gift wrap*
Infants’ equipment*

Special aggregate indexes
All items—old base
All items less energy
All items less food
All items less food and energy
All items less medical care
All items less shelter
All items less food and shelter
All items less food, shelter, and energy
All items less food, shelter, energy, and used cars and trucks
Apparel less footwear
Commodities
Commodities less food
Commodities less food and beverages
Commodities less food and energy commodities
Commodities less food, energy, and used cars and trucks
Domestically produced farm food
Durables
Education and communication commodities
Education and communication services
Energy
Energy commodities
Household furnishings and supplies
Information technology commodities
Nondurables
Nondurables less food
Nondurables less food and apparel
Nondurables less food and beverages
49

Appendix 1. List of published indexes—continued
Nondurables less food, beverages, and apparel
Other goods
Other personal services
Other services
Recreation commodities
Recreation services
Rent of shelter
Services
Services less energy services
Services less medical care services
Services less rent of shelter
Transportation commodities less motor fuel
Transportation services
Utilities and public transportation
Video and audio products
Video and audio services
Purchasing power of the consumer dollar
Purchasing power of the consumer dollar–old base
* Special index based on a substantially smaller sample.

50

Appendix 1. List of published indexes—continued
CPI-U and CPI-W Indexes Published at the Regional City-Size Class and Local Area Levels
All items
Food and beverages
Food
Food at home
Food away from home
Alcoholic beverages
Housing
Shelter
Rent of primary residence
Owners’ equivalent rent of residences
Owner’s equivalent rent of primary residence
Fuels and utilities
Household energy
Energy services
Electricity
Utility (piped) gas service
Household furnishings and operations
Apparel
Transportation
Private transportation
Motor fuel
Gasoline (all types)
Gasoline, unleaded regular*
Gasoline, unleaded midgrade*
Gasoline, unleaded premium*
Medical Care
Recreation
Education and communication
Other goods and services
Special Aggregate indexes
All items—old base
All items less energy
All items less food and energy
All items less medical care
All items less shelter
Commodities
Commodities less food
Commodities less food and beverages
Durables
Education and communication commodities
Education and communication services
Energy
Household furnishings and supplies
Nondurables
Nondurables less food
Nondurables less food and beverages
Other goods
Other personal services
Services
Services less medical care services
Services less rent of shelter
Transportation commodities less motor fuel
* Special index based on a substantially smaller sample.
51

Appendix 1. List of published indexes—continued
C-CPI-U Indexes Published at the U.S. City Average (National) Level
(The C-CPI-U is issued for national averages only, and employs a December 1999 = 100 reference base.)
All items
Food and beverages
Food
Food at home
Food away from home
Alcoholic beverages

Housing
Shelter
Fuels and utilities
Household furnishings and operations

Apparel
Transportation
Private transportation
Public transportation
Medical care
Medical care commodities
Medical care services
Recreation
Education and communication
Education
Communication
Other goods and services
Special aggregate indexes
All items less food and energy
Commodities
Durables
Energy
Nondurables
Services

52

Appendix 1. List of published indexes—continued
CPI-U and CPI-W Indexes Published at the National City-Size Class
All items
Food and beverages
Food
Food at home
Food away from home
Alcoholic beverages

Housing
Shelter
Rent of primary residence
Owners’ equivalent rent of residences
Owner’s equivalent rent of primary residence
Fuels and utilities
Household energy
Energy services
Electricity
Utility (piped) gas service
Household furnishings and operations

Apparel

Transportation
Private transportation
New and used motor vehicles
New vehicles
New cars and trucks*
New cars*
Used cars and trucks
Motor fuel
Gasoline (all types)

Gasoline, unleaded regular*
Gasoline, unleaded midgrade*
Gasoline, unleaded premium*

Medical care
Medical care commodities
Medical care services
Professional services

Recreation
Education and communication
Other goods and services
Special aggregate indexes
All items—old base
All items less energy
All items less food
All items less food and energy

53

Appendix 1. List of published indexes—continued
All items less medical care
All items less shelter
Commodities
Commodities less food
Commodities less food and beverages
Commodities less food and energy commodities
Durables
Education and communication commodities
Education and communication services
Energy
Energy commodities
Household furnishings and supplies
Nondurables
Nondurables less food
Nondurables less food and apparel
Nondurables less food and beverages
Nondurables less food, beverages, and apparel
Other goods
Other personal services
Other services
Recreation commodities
Recreation services
Rent of shelter
Services
Services less energy services
Services less medical care services
Services less rent of shelter
Transportation commodities less motor fuel
Transportation services
* Special index based on a substantially smaller sample.

54

Appendix 2. List of average retail price series published by the Bureau of Labor Statistics
(“X” indicates published)
United
States

Region

Population
size class

Area

Utility natural gas per therm

X

X

X

X

Fuel oil #2 per gallon (3.785 liters)

X

X

X

Electricity per kWh

X

X

X

X

Gasoline, all types, per gallon/3.785 liters

X

X

X

X

Gasoline, unleaded regular, per gallon/3.785 liters

X

X

X

X

Gasoline, unleaded midgrade, per gallon/3.785 liters

X

X

X

X

Gasoline, unleaded premium, per gallon/3.785 liters

X

X

X

X

Automotive diesel fuel, per gallon/3.785 liters

X

X

X

Flour, white, all purpose, per lb (453.6 gm)

X

X

Rice, white, long grain, uncooked, per lb (453.6 gm)

X

X

Spaghetti and macaroni, per lb (453.6 gm)

X

X

Bread, white, pan, per lb (453.6 gm)

X

X

Bread, French, per lb (453.6 gm)

X

X

Bread, whole wheat, pan, per lb (453.6 gm)

X

X

Cookies, chocolate chip, per lb (453.6 gm)

X

X

Crackers, soda, salted, per lb (453.6 gm)

X

X

Ground chuck, 100% beef, per lb (453.6 gm)

X

X

Ground beef, 100% beef, per lb (453.6 gm)

X

X

Ground beef, lean and extra lean, per lb (453.6 gm)

X

X

All uncooked ground beef, per lb (453.6 gm)

X

X

Chuck roast, USDA Choice, bone-in, per lb (453.6 gm)

X

X

Chuck roast, graded and ungraded, excluding USDA Prime and
Choice, per lb (453.6 gm)

X

X

Chuck roast, USDA Choice, boneless, per lb (453.6 gm)

X

X

Round roast, USDA Choice, boneless, per lb (453.6 gm)

X

X

Round roast, graded and ungraded, excluding USDA Prime and
Choice, per lb (453.6 gm)

X

X

Rib roast, USDA Choice, bone-in, per lb (453.6 gm)1

X

X

Title
Energy

Cereals and bakery products

Beef and veal

All uncooked beef roasts, per lb (453.6 gm)

X

X

2

Steak, T-bone, USDA Choice, bone-in, per lb (453.6 gm)

X

X

Steak, rib eye, USDA Choice, boneless, per lb (453.6 gm)3

X

X

Steak, round, USDA Choice, boneless, per lb (453.6 gm)

X

X

Steak, round, graded and ungraded, excluding USDA Prime and
Choice, per lb (453.6 gm)

X

X

Steak, sirloin, USDA Choice, bone-in, per lb (453.6 gm)4

X

X

55

Appendix 2. List of average retail price series published by the Bureau of Labor Statistics
(“X” indicates published)—continued
United
States

Region

Steak, sirloin, graded and ungraded, excluding USDA Prime and
Choice, per lb (453.6 gm)

X

X

Steak, sirloin, USDA Choice, boneless, per lb (453.6 gm)

X

X

Short ribs, any primal source, bone-in, per lb (453.6 gm)

X

X

Beef for stew, boneless, per lb (453.6 gm)

X

X

All uncooked beefsteaks, per lb (453.6 gm)

X

X

All uncooked other beef (excluding veal), per lb (453.6 gm)

X

X

Bacon, sliced, per lb (453.6 gm)

X

X

Chops, center cut, bone-in, per lb (453.6 gm)

X

X

Chops, boneless, per lb (453.6 gm)

X

X

All pork chops, per lb (453.6 gm)

X

X

Ham, rump, or shank half, bone-in, smoked, per lb (453.6 gm)

X

X

Ham, boneless, excluding canned, per lb (453.6 gm)

X

X

All ham (excluding canned ham and luncheon slices), per lb (453.6
gm)

X

X

Ham, canned, 3 or 5 lbs, per lb (453.6 gm)

X

X

Shoulder picnic, bone-in, smoked, per lb (453.6 gm)

X

X

All other pork (excluding canned ham and luncheon slices), per lb
(453.6 gm)

X

X

Sausage, fresh, loose, per lb (453.6 gm)

X

X

Frankfurters, all meat or all beef, per lb (453.6 gm)

X

X

Bologna, all beef or mixed, per lb (453.6 gm)

X

X

Lamb and mutton, bone-in, per lb (453.6 gm)

X

X

Chicken, fresh, whole, per lb (453.6 gm)

X

X

Chicken breast, bone-in, per lb (453.6 gm)

X

X

Chicken breast, boneless, per lb (453.6 gm)

X

X

Chicken legs, bone-in, per lb (453.6 gm)

X

X

Turkey, frozen, whole, per lb (453.6 gm)

X

X

Tuna, light, chunk, per lb (453.6 gm)

X

X

Eggs, grade A, large, per doz

X

X

Eggs, grade AA, large, per doz

X

X

Milk, fresh, whole, fortified, per ½ gal (1.9 lit)

X

X

Milk, fresh, whole, fortified, per gal (3.8 lit)

X

X

Milk, fresh, low fat, per ½ gal (1.9 lit)

X

X

Title

Pork and other meats

Poultry, fish, and eggs

Dairy products

56

Population
size class

Area

Appendix 2. List of average retail price series published by the Bureau of Labor Statistics
(“X” indicates published)—continued
United
States

Region

Milk, fresh, low fat, per gal (3.8 lit)

X

X

Butter, salted, grade AA, stick, per lb (453.6 gm)

X

X

American processed cheese, per lb (453.6 gm)

X

X

Cheddar cheese, natural, per lb (453.6 gm)

X

X

Ice cream, prepackaged, bulk, regular, per ½ gal (1.9 lit)

X

X

Yogurt, natural, fruit flavored, per 8 oz (226.8 gm)

X

X

Apples, Red Delicious, per lb (453.6 gm)

X

X

Bananas, per lb (453.6 gm)

X

X

Oranges, Navel, per lb (453.6 gm)

X

X

Oranges, Valencia, per lb (453.6 gm)

X

X

Cherries, per lb (453.6 gm)

X

X

Grapefruit, per lb (453.6 gm)

X

X

Grapes, Thompson, Seedless, per lb (453.6 gm)

X

X

Lemons, per lb (453.6 gm)

X

X

Peaches, per lb (453.6 gm)

X

X

Pears, Anjou, per lb (453.6 gm)

X

X

Strawberries, dry pint, per 12 oz (340.2 gm)

X

X

Potatoes, white, per lb (453.6 gm)

X

X

Lettuce, iceberg, per lb (453.6 gm)

X

X

Lettuce, romaine, per lb (453.6 gm)

X

X

Tomatoes, field grown, per lb (453.6 gm)

X

X

Broccoli, per lb (453.6 gm)

X

X

Cabbage, per lb (453.6 gm)

X

X

Carrots, short trimmed and topped, per lb (453.6 gm)

X

X

Celery, per lb (453.6 gm)

X

X

Corn on the cob, per lb (453.6 gm)

X

X

Cucumbers, per lb (453.6 gm)

X

X

Onions, dry yellow, per lb (453.6 gm)

X

X

Peppers, sweet, per lb (453.6 gm)

X

X

Applesauce, any variety, all sizes, per lb (453.6 gm)

X

X

Orange juice, frozen concentrate, 12 oz can, per 16 oz (473.2 mL)

X

X

Beans, dried, any type, all sizes, per lb (453.6 gm)

X

X

Corn, canned, any style, all sizes, per lb (453.6 gm)

X

X

Potatoes, frozen, French fried, per lb (453.6 gm)

X

X

Peaches, any variety, all sizes, per lb

X

X

Title

Fresh fruits and vegetables

Processed fruits and vegetables

57

Population
size class

Area

Appendix 2. List of average retail price series published by the Bureau of Labor Statistics
(“X” indicates published)—continued
United
States

Region

Sugar, white, all sizes, per lb (453.6 gm)

X

X

Sugar, white, 33–80 oz pkg, per lb (453.6 gm)

X

X

Margarine, stick, per lb (453.6 gm)

X

X

Margarine, soft, tubs, per lb (453.6 gm)

X

X

Shortening, vegetable oil blends, per lb (453.6 gm)

X

X

Peanut butter, creamy, all sizes, per lb (453.6 gm)

X

X

Cola, nondiet, cans, 72-oz 6 pk., per 16 oz (473.2 mL) {deposit may
be included in price.}

X

X

Cola, nondiet, per 2 liters (67.6 oz) {deposit may be included in
price.}

X

X

Coffee, 100 percent, ground roast, all sizes, per lb (453.6 gm)

X

X

Coffee, 100 percent, ground roast, 13.1–20 oz can, per lb (453.6 gm)

X

X

Coffee, instant, plain, regular, all sizes, per lb (453.6 gm)

X

X

Potato chips, per 16 oz (453.6 gm)

X

X

Malt beverages, all types, all sizes, any origin, per 16 oz (473.2 mL)

X

X

Vodka, all types, all sizes, any origin, per 1 liter (33.8 oz)

X

X

Wine, red and white table, all sizes, any origin, per 1 liter (33.8 oz)

X

X

Title
Other food items

Alcoholic Beverages

1 Last in published 1997.
2 Last published in April 2002.
3 Last published in October 2001.
4 Last published April 1994.

58

Population
size class

Area

CPI Appendix 3. Characteristics of the Consumer Price Index, 1890 to date
Date

Survey providing
expenditure weight
Group
weights

Base period

Item weights

1890 1

None

1901

Varied

1919

1917–1919

1917–1919

1913

Census
providing
population
weights

None

Family
composition

Two or more
persons.

Earnings of
chief earner

Source and
amount of family
income

Salaried worker
earning $1,200
or less during
No limitation.
year. No
limitation on
wage earners.

Economic level,
length of
Length of
residence,
employment
nativity, and
race

No limitation.

32

Minimum of
husband, wife,
and one child
who was not a
boarder or lodger.
No boarders nor
more than three
lodgers present.

33

At least $500. Less
than one-fourth
from interest,
At least $300.
dividends,
Salaried worker
royalties,
earning less
Two or more
speculative gains,
than $2,000
persons. Not
rents, gifts, or
At least 1,008
during year or
more than two
income in kind.
hours spread
less than $200
boarders or
No rent in
over 36 weeks.
lodgers, or guests during any
payment of
for more than 26 month. No
services. Less than
upper limitation
guest-weeks.
3 months’ free
on wage
rent. No subsidiary
earners.
clerical worker
earning $2,000 or
more.

Salaried worker
earning $2,000
or less. No
limitation on
wage earners.

At least 75 percent
from principal
earner or others
who contributed
all earnings to
family fund.

Title(s)

No limitation.

No slum or
charity families;
White only; in
area entire year
and in United
Cost of
States 5 years or living.
more; no nonEnglishspeaking
families.

3
Average
1920–1930

Sept. 1935
Dec. 19354

1923–1925

6

1934–1936

1935–1939

1930

May 19418
July 1943

Varied

2

Feb. 1921

Aug. 19405 1934–1936

Number of
areas
included

7

No relief
families, either
on direct or
work relief;
White only,
except where
Black
population was
significant part
of total; in area
9 months or
more.

Indexes of
the cost of
living of
wage earners
and lower
salaried
workers in
large cities.

34
9

1940
Consumer
Price Index
for ModerateIncome
Families in
Large Cities.

Sept. 1945

59

CPI Appendix 3. Characteristics of the Consumer Price Index, 1890 to date—Continued
Survey providing
expenditure weight

Date

Group
weights

Jan. 1951 10 1947–1949

Jan.1953 12

13

Item weights

11

13

1950

1950

17

1960–1961

17

14

1947–1949

15

1957–1959

Family
composition

Two or more
persons.

Earnings of
chief earner

50

Source and
amount of family
income

Family income
under $10,000
after taxes in the
No limitation. survey year. No
(Family income minimum income
not in excess of limit, except that
$10,000.)
families with no
income from
wages or salaries
were excluded.

Length of
employment

Economic level,
length of
residence,
nativity, and
race

Families of two
or more persons
and single
workers; at least
one full-time
wage earner.

56
1967

60

No limitation.

Title(s)

No exclusion for
receipt of relief
as such, but
only families
with wage or
Family head
must have been salary earnings
included. No
employed at
least 26 weeks. length of
residence,
nativity, or
racial
limitations.

Short title:
Consumer
Price Index.
Complete
name: Index
of Change in
Prices of
Goods and
Services
Purchased by
City WageEarner and
ClericalWorker
Families to
Maintain
Their Level
of Living.

No specific
requirement,
but major
portion of
income of
family head
must be from
employment as
wage earner or
salaried clerical
worker.

46

1960

1960–1961

Jan. 1966 18
Jan. 1971 19

Number of
areas
included

1950

1934–1936

Jan. 1962

Jan. 1964 16

Base period

Census
providing
population
weights

More than half of
A minimum of
combined family
income from wage- 37 weeks for at
earner or clerical- least 1 family
worker occupation. member.

No restrictions
other than the
wage-earner and
clerical-worker
definition.

Consumer
Price Index
for Urban
Wage
Earners and
Clerical
Workers.

CPI Appendix 3. Characteristics of the Consumer Price Index, 1890 to date—Continued
Survey providing
expenditure weight

Date

Group
weights

Jan. 1978 20

21

Base period

Item weights

22

1972–1973

Census
providing
population
weights

1970

1974

Number of
areas
included

Family
composition

85

For earner and
clerical-worker
indexes, families
of two or more
persons and
single workers; at
least one fulltime wage
earner.. No
limitation for
urban consumer
index.

Jan.1983 24

CPI-U

Jan. 1985 25

CPI-W

Jan. 1987 26

Families of two
or more persons
and single
workers; at least
one full-time
wage earner.
Students residing
in collegeregulated housing
are treated as
separate family
units.

27

1982–1984

28

29

Jan. 1988
Jan. 1998

30 31

Jan. 2002 33

34

1993–1995

1992–1996

1990

1999–2000

35

1997–2001

1990

Dec.1999

Jan. 2004 37

38

2001–2002

39

2000–2003

Jan. 2006 40

41

2003–2004

42

2002–2005

2005–2006

45

2004–2007

43 44

Jan. 2010 46

47

2007–2008

48

2006–2009

Jan. 2012 49

50

2009–2010

51

2008–2011

2011–2012

54

2010–2013

Jan. 2014

52 53

91

Source and
amount of family
income

Length of
employment

Economic level,
length of
residence,
nativity, and
race

For wage-earner For wage-earner For wage-earner
and clericaland clericaland clericalworker indexes, worker indexes,
worker indexes,
more than half of a minimum of no restrictions
combined family 37 weeks for at other than the
income from wage- least one family wage-earner and
clerical-worker
earner or clerical- member. No
definition. No
worker occupation. employment
limitation for
No limitation for required for
urban consumer urban consumer
urban consumer
index.
index.
index.23

Title(s)

1. Consumer
Price Index
for Urban
Wage
Earners and
Clerical
Workers
(CPI-W).
2. Consumer
Price Index
for All Urban
Consumers
(CPI-U).

1982–1984

32

July 200236 1999–2000

Jan. 2008

1980

1985–1989

Earnings of
chief earner

Same as CPI-U
population.

61

Chained
Consumer
Price Index
for All Urban
Consumers
(C-CPI-U).

CPI Appendix 3. Characteristics of the Consumer Price Index, 1890 to date—Continued

1

Food price indexes only.

2

For 19 cities, data were available back to December 1914, and for 13 cities, back to 1917. For the United States, data were available back to the 1913
annual average.

3

Indexes between 1918 and 1929 were recomputed retroactively with population weights based on the average of the 1920 and 1930 censuses.

4

Index published in December 1935 for July 15, 1935; Indexes were also calculated on the 1913 = 100 base.

5

Indexes between 1925 and 1929 were recomputed retroactively, with group weights based on the average of 1917–1919 and 1934–1936. Indexes between
March 15, 1930, and March 15, 1940, were recomputed retroactively with 1934–1936 group weights.
6

During World War II, weights were adjusted to account for rationing and shortages.

7

There were 51 to 56 cities included in the food index.

8
9

Index published in May 1941 for March 14, 1941. Food indexes were based on 51 cities.
Census data for 1940 were supplemented by ration book registration data.

10

Index published in March 1951 for January 1951.

11

All-item and group indexes bteween January 1950 and January 1951 were revised retroactively. Rent and all-item indexes were corrected for new-unit
bias from 1940. Old series also published through 1952.

12

Item weights were revised for only the seven cities for which 1947–1949 expenditure data were available. Index was published in February and January
1953. Linked to old series as of December 1952. Old series also published during a 6-month overlap period.

13

Data were adjusted to 1952 for weight derivation.

14

Indexes also were calculated on the base of 1935–1939 = 100 through December 1957.

15

Index published in February for January 1962. Indexes also were calculated on bases of 1947–1949 = 100 and 1939 = 100.

16

Index published March 3 for January 1964. Linked to old series as of December 1963. Old series also published during a 6-month overlap period.

17

Data were adjusted to December 1963 for weight derivation.

18

Index published in February for January 1966. Linked to old series as of December 1965.

19

Index published in February for January 1971. Indexes were also calculated on the 1957–1959 = 100 base.

20

Index published in February for January 1978. Linked to old series as of December 1977. Old series also published during a 6-month overlap period.

21

Data were adjusted to December 1977 for weight derivation.

22

Item weights based on Points of Purchase Survey in 1974.

23

Coverage was expanded to include wage earners and clerical workers in the entire nonfarm parts of metropolitan areas, in addition to those living within
the urbanized areas of metropolitan areas and urban places of 2,500 or more inhabitants.
24

Changed homeowners’ costs from asset approach to flow-of-service approach (rental equivalence).

25

Changed homeowners’ costs from asset approach to flow-of-service approach (rental equivalence).

26

Index published in February for January 1987. Linked to old series as of December 1986. Old series also published during a 6-month overlap period.

27

Data were adjusted to December 1986 for weight derivation.

28

Item weights based on Continuing Point of Purchase Survey (CPOPS) from 1985–1989; first “rolling revision.”

29

Index published in February for January 1988. Indexes also calculated on the 1967 = 100 base.

30

Index published in February for January 1989. Linked to old series as of December 1997. Old series also published during a 6-month overlap period.

31

Data adjusted to December 1997 for weight derivation.

32

At time of revision, the CPOPS covered 1992–1996. The revised TPOPS was introduced in 1998.

33

Index published in February for January 2002. Linked to old series as of December 2001. Old series also published during a 6-month overlap period.

34

Data were adjusted to December 2001 for weight deviation.

35

TPOPS on 4-year rotation; at the time of expenditure weight update, largely reflected data from 1997 forward.

36

New index. Elementary indexes aggregated by using an adjusted geometric mean for the preliminary versions and a Törnqvist formula for the final
version. Until 2014, first issued in preliminary form (initial); subject to revision in February (interim) and again the following February (final). Starting in
37

Index published in February for January 2004. Linked to old series as of December 2003.

38

Data were adjusted to December 2003 for weight deviation.

39

TPOPS on 4-year rotation; at the time of expenditure weight update, largely reflected data from 1999 forward.

40

Index published in February for January 2006. Linked to old series as of December 2005.

41

Data were adjusted to December 2005 for weight deviation.

62

CPI Appendix 3. Characteristics of the Consumer Price Index, 1890 to date—Continued

42

TPOPS on 4-year rotation; at the time of expenditure weight update, largely reflected data from 2001 forward.

43

Index published in February for January 2008. Linked to old series as of December 2007.

44

Data were adjusted to December 2007 for weight deviation.

45

TPOPS on 4-year rotation; at the time of expenditure weight update, largely reflected data from 2003 forward.

46

Index published in February for January 2010. Linked to old series as of December 2009.

47

Data were adjusted to December 2009 for weight deviation.

48

TPOPS on 4-year rotation; at the time of expenditure weight update, largely reflected data from 2005 forward.

49

Index published in February for January 2012. Linked to old series as of December 2011.

50

Data were adjusted to December 2011 for weight deviation.

51

TPOPS on 4-year rotation; at the time of expenditure weight update, largely reflected data from 2007 forward.

52

Index published in February for January 2014. Linked to old series as of December 2013.

53

Data were adjusted to December 2013 for weight deviation.

54

TPOPS on 4-year rotation; at the time of expenditure weight update, largely reflected data from 2009 forward.

63

CPI Appendix 4. Sample Areas and Pricing Cycles
PSU

Sample areas and counties

Pricing cycle
(months)
Odd
Even

B102

Reading, PA MSA
Berks

X

B104

Syracuse, NY MSA
Cayuga, Madison, Onondaga,
Oswego

X

B106

Buffalo–Niagara Falls, NY MSA
Erie, Niagara

X

B108

Hartford, CT MSA
Hartford (part), Litchfield
(part), Middlesex (part), New
London (part), Tolland (part),
Windham (part)

X

B110

Burlington, VT MSA
Chittenden (part), Franklin
(part), Grand Isle (part)

X

B112

Sharon, PA MSA
Mercer

X

B114

Johnstown, PA MSA
Cambria, Somerset

X

B116

Springfield, MA MSA
Franklin (part), Hampden
(part), Hampshire (part)

X

X

X

A110

New York–Connecticut suburbs
NY Portion:
Dutchess, Nassau, Orange,
Putnam, Rockland, Suffolk,
Westchester
CT Portion:
Fairfield, Litchfield (part),
Middlesex (part), New
Haven (part)

X

X

New Jersey–Pennsylvania
suburbs
NJ Portion:
Bergen, Essex, Hudson,
Hunterdon, Mercer,
Middlesex, Monmouth,
Morris, Ocean, Passaic,
Somerset, Sussex, Union,
Warren
PA Portion:
Pike

X

A103

A104

Pricing cycle
(months)
Odd
Even

Northeast Region

New York City
Bronx, Kings, New York,
Queens, Richmond

A102

Sample areas and counties

New York–Northern New Jersey–
Long Island, NY–NJ–CT–PA
CMSA:
A109

A111

PSU

Midwest Region

Philadelphia–Wilmington–
Atlantic City, PA–NJ–DE–MD
CMSA
NJ Portion:
Atlantic, Burlington, Camden,
Cape May, Cumberland,
Gloucester, Salem
DE Portion:
New Castle
MD Portion:
Cecil
PA Portion:
Bucks, Chester, Delaware,
Montgomery, Philadelphia
Boston–Brockton–Nashua, MA–
NH–ME–CT CMSA
CT Portion:
Windham (part)
MA Portion:
Bristol (part), Essex, Hampden
(part), Middlesex, Norfolk,
Plymouth, Suffolk, Worcester
(part)
ME Portion:
York (part)
NH Portion:
Hillsborough (part), Merrimack
(part), Rockingham (part),
Strafford (part)
Pittsburgh, PA MSA
Allegheny, Beaver, Butler,
Fayette, Washington,
Westmoreland

X

X

A207

Chicago–Gary–Kenosha, IL–IN–
WI CMSA
IL Portion:
Cook, DeKalb, DuPage,
Grundy, Kane, Kankakee,
Kendall, Lake, McHenry, Will
IN Portion:
Lake, Porter
WI Portion:
Kenosha

A208

Detroit–Ann Arbor–Flint, MI
CMSA
Genesee, Lapeer, Lenawee,
Livingston, Macomb, Monroe,
Oakland, St. Clair, Washtenaw,
Wayne

X

A209

St. Louis, MO–IL MSA
IL Portion:
Clinton, Jersey, Madison,
Monroe, St. Clair
MO Portion:
Crawford (part), Franklin,
Jefferson, Lincoln, St. Charles,
St. Louis, Warren, St. Louis
city

X

A210

Cleveland–Akron, OH CMSA
Ashtabula, Cuyahoga, Geauga,
Lake, Lorain, Medina, Portage,
Summit

X

X

64

X

X

X

CPI Appendix 4. Sample areas and pricing cycles—continued
Pricing cycle
(months)
Odd
Even

PSU

Sample areas and counties

A211

Minneapolis–St. Paul, MN–WI
MSA
MN Portion:
Anoka, Carver, Chisago,
Dakota, Hennepin, Isanti,
Ramsey, Scott, Sherburne,
Washington, Wright
WI Portion:
Pierce, St. Croix

X

A212

Milwaukee–Racine, WI CMSA
Milwaukee, Ozaukee, Racine,
Washington, Waukesha

X

A213

Cincinnati–Hamilton, OH–KY–IN
CMSA
IN Portion:
Dearborn, Ohio
KY Portion:
Boone, Campbell, Gallatin,
Grant, Kenton, Pendleton
OH Portion:
Brown, Butler, Clermont,
Hamilton, Warren

X

A214

PSU

Sample areas and counties

C212

Faribault–Northfield, MN
Urban parts of:
Rice

C216

Chanute–Iola, KS
Urban parts of:
Allen, Neosho
Housing only in Urban parts
of:
Bourbon, Crawford

C218

Brookings–Madison, SD
Urban parts of:
Brookings, Lake, Moody

C222

Mount Vernon, IL
Urban parts of:
Jefferson
Housing only in Urban parts
of:
Marion

Pricing cycle
(months)
Odd
Even
X

X

X

X

South Region

Kansas City, MO–KS MSA
KS Portion:
Johnson, Leavenworth, Miami,
Wyandotte
MO Portion:
Cass, Clay, Clinton, Jackson,
Lafayette, Platte, Ray

X

B218

Wausau, WI MSA
Marathon

X

B220

Dayton–Springfield, OH MSA
Clark, Greene, Miami,
Montgomery

X

B222

Evansville–Henderson, IN–KY
MSA
IN Portion:
Posey, Vanderburgh, Warrick
KY Portion:
Henderson

X

B224

Columbus, OH MSA
Delaware, Fairfield, Franklin,
Licking, Madison, Pickaway

X

B226

Saginaw–Bay City–Midland, MI
MSA
Bay, Midland, Saginaw

X

B228

Elkhart–Goshen, IN MSA
Elkhart

X

B230

Decatur, IL MSA
Macon

X

B232

Youngstown–Warren, OH MSA
Columbiana, Mahoning,
Trumbull

X

B234

Madison, WI MSA
Dane

X

B236

Lincoln, NE MSA
Lancaster

Washington–Baltimore, DC–MD–
VA–WV CMSA:

X

65

A312

Washington, DC–MD–VA–WV
DC Portion:
District of Columbia
MD Portion:
Calvert, Charles, Frederick,
Montgomery, Prince
George's, Washington
VA Portion:
Arlington, Clarke, Culpeper,
Fairfax, Fauquier, King
George, Loudoun, Prince
William, Spotsylvania,
Stafford, Warren, Alexandria
city, Fairfax city, Falls
Church city, Fredericksburg
city, Manassas city,
Manassas Park city
WV Portion:
Berkeley, Jefferson

X

A313

Baltimore, MD PMSA
Anne Arundel, Baltimore,
Carroll, Harford, Howard,
Queen Anne's, Baltimore city

X

A316

Dallas–Fort Worth, TX CMSA
Collin, Dallas, Denton, Ellis,
Henderson, Hood, Hunt,
Johnson, Kaufman, Parker,
Rockwall, Tarrant

A318

Houston–Galveston–Brazoria, TX
CMSA
Brazoria, Chambers, Fort Bend,
Galveston, Harris, Liberty,
Montgomery, Waller

X

X

CPI Appendix 4. Sample areas and pricing cycles—continued
PSU
A319

A320

A321

B338

Sample areas and counties

Pricing cycle
(months)
Odd
Even

Atlanta, GA MSA
Barrow, Bartow, Carroll,
Cherokee, Clayton, Cobb,
Coweta, DeKalb, Douglas,
Fayette, Forsyth, Fulton,
Gwinnett, Henry, Newton,
Paulding, Pickens, Rockdale,
Spalding, Walton

X

Miami–Fort Lauderdale, FL
CMSA
Broward, Dade

X

Tampa–St. Petersburg–Clearwater,
FL MSA
Hernando, Hillsborough,
Pasco, Pinellas

X

Chattanooga, TN–GA MSA
GA Portion:
Catoosa, Dade, Walker
TN Portion:
Hamilton

X

B340

Florence, SC MSA
Florence

X

B342

Albany, GA MSA
Dougherty, Lee

X

B344

Norfolk–Virginia Beach–Newport
News, VA–NC MSA
NC Portion:
Currituck
VA Portion:
Gloucester, Isle of Wight,
James City, Mathews, York,
Chesapeake city Hampton city,
Newport News city, Norfolk
city, Poquoson city,
Portsmouth city, Suffolk city,
Virginia Beach city,
Williamsburg city

X

B346

Pine Bluff, AR MSA
Jefferson

X

B348

Raleigh–Durham–Chapel Hill, NC
MSA
Chatham, Durham, Franklin,
Johnstown, Orange, Wake

X

B350

Richmond–Petersburg, VA MSA
Charles City, Chesterfield,
Dinwiddie, Goochland,
Hanover, Henrico, New Kent,
Powhatan, Prince George,
Colonial Heights city,
Hopewell city, Petersburg city,
Richmond city

X

B352

Beaumont–Port Arthur, TX MSA
Hardin, Jefferson, Orange

X

B354

Brownsville–Harlingen–San
Benito, TX MSA
Cameron

X

B356

Florence, AL MSA
Colbert, Lauderdale

X

PSU

66

Sample areas and counties

Pricing cycle
(months)
Odd
Even

B358

Greenville–Spartanburg–
Anderson, SC MSA
Anderson, Cherokee,
Greenville, Pickens,
Spartanburg

X

B360

Fort Myers–Cape Coral, FL MSA
Lee

X

B362

Birmingham, AL MSA
Blount, Jefferson, St. Clair,
Shelby

X

B364

Melbourne–Titusville–Palm Bay,
FL MSA
Brevard

X

B366

Lafayette, LA MSA
Acadia, Lafayette, St. Landry,
St. Martin

X

B368

Ocala, FL MSA
Marion

X

B370

Gainesville, FL MSA
Alachua

X

B372

Amarillo, TX MSA
Potter, Randall

X

B374

San Antonio, TX MSA
Bexar, Comal, Guadalupe,
Wilson

B376

Oklahoma City, OK MSA
Canadian, Cleveland, Logan,
McClain, Oklahoma,
Pottawattamie

B378

Baton Rouge, LA MSA
East Baton Rouge, Livingston,
West Baton Rouge

B380

Odessa–Midland, TX MSA
Ector, Midland

X

C328

Arcadia–Wauchula, FL
Urban parts of:
De Soto, Hardee
Housing only in Urban parts
of:
Hendry, Highlands

X

C332

Morristown–Jefferson City, TN
Urban parts of:
Hamblen, Jefferson

C334

Picayune–Poplarville, MS
Urban parts of:
Pearl River
Housing only in Urban parts
of:
Lamar, Marion

C344

Statesboro–Waynesboro–Millen–
Sylvania, GA
Urban parts of:
Bulloch, Burke, Jenkins,
Screven

X

X

X

X

X

X

CPI Appendix 4. Sample areas and pricing cycles—continued
PSU

Sample areas and counties

Pricing cycle
(months)
Odd
Even

PSU

West Region
Los Angeles–Riverside–Orange
County, CA CMSA:
A419

Los Angeles–Long Beach, CA
PMSA
Los Angeles

X

X

A420

Los Angeles suburbs, CA
Orange, Riverside, San
Bernardino, Ventura

X

X

A422

A423

San Francisco–Oakland–San Jose,
CA CMSA
Alameda, Contra Costa, Marin,
Napa, Santa Clara, Santa Cruz,
San Francisco, San Mateo,
Solano, Sonoma

X

Seattle–Tacoma–Bremerton, WA
CMSA
Island, King, Kitsap, Pierce,
Snohomish, Thurston

X

A424

San Diego, CA MSA
San Diego

X

A425

Portland–Salem, OR–WA CMSA
OR Portion:
Clackamas, Columbia, Marion,
Multnomah, Polk, Washington,
Yamhill
WA Portion:
Clark

X

Honolulu, HI MSA
Honolulu

X

A426

67

Sample areas and counties

Pricing cycle
(months)
Odd
Even

A427

Anchorage, AK MSA
Anchorage

X

A429

Phoenix–Mesa, AZ MSA
Maricopa, Pinal

X

A433

Denver–Boulder–Greeley, CO
CMSA
Adams, Arapahoe, Boulder,
Broomfield, Denver, Douglas,
Jefferson, Weld

X

B482

Chico–Paradise, CA MSA
Butte

X

B484

Provo–Orem, UT MSA
Utah

X

B486

Modesto, CA MSA
Stanislaus

X

B488

Boise City, ID MSA
Ada, Canyon

X

B490

Las Vegas, NV–AZ MSA
AZ Portion:
Mohave
NV Portion:
Clark, Nye

B492

Yuma, AZ MSA
Yuma

C450

Bend–Redmond, OR
Urban parts of:
Deschutes

C456

Pullman–Colfax, WA
Urban parts of:
Whitman

X

X
X

X

Appendix 5. Major groups, expenditure classes, item stratum, and entry-level items

MAJOR GROUP: FOOD AND BEVERAGES
CEREALS AND BAKERY PRODUCTS
FA Cereals and cereal products
FA01
Flour and prepared flour mixes
FA011
FLOUR AND PREPARED FLOUR MIXES
FA02
Breakfast cereal
FA021
BREAKFAST CEREAL
FA03
Rice, pasta, cornmeal
FA031
RICE, PASTA, CORNMEAL
FB Bakery products
FB01
Bread
FB011
BREAD
FB02
Fresh biscuits, rolls, muffins
FB021
FRESH BISCUITS, ROLLS, AND MUFFINS
FB03
Cakes, cupcakes, and cookies
FB031
CAKES AND CUPCAKES (EXCLUDING FROZEN)
FB032
COOKIES
FB04
Other bakery products
FB041
CRACKERS AND BREAD, AND CRACKER PRODUCTS
FB042
SWEETROLLS, COFFEE CAKE, AND DOUGHNUTS (EXCLUDING FROZEN)
FB043
FROZEN BAKERY PRODUCTS, AND FROZEN/REFRIGERATED DOUGHS AND BATTERS
FB044
PIES, TARTS, TURNOVERS (EXCLUDING FROZEN)
MEATS, POULTRY, FISH, AND EGGS
FC Beef and veal
FC01
Uncooked ground beef
FC011
UNCOOKED GROUND BEEF
FC02
Uncooked beef roasts
FC021
UNCOOKED BEEF ROASTS
FC03
Uncooked beef steaks
FC031
UNCOOKED BEEF STEAKS
FC04
Uncooked other beef and veal
FC041
UNCOOKED OTHER BEEF AND VEAL
FD Pork
FD01
Bacon, breakfast sausage, and related products
FD011
BACON, BREAKFAST SAUSAGE, AND RELATED PRODUCTS
FD02
Ham
FD021
HAM
FD03
Pork chops
FD031
PORK CHOPS
FD04
Other pork, including roasts and picnics
FD041
OTHER PORK INCLUDING ROASTS AND PICNICS
FE Other meats
FE01
Other meats
FE011
FRANKFURTERS
FE012
LUNCHMEATS
FE013
LAMB, ORGAN MEATS, AND GAME

68

Appendix 5. Major groups, expenditure classes, item stratum, and entry-level items—continued

FF Poultry
FF01
Chicken
FF011
CHICKEN
FF02
Other poultry including turkey
FF021
OTHER POULTRY INCLUDING TURKEY
FG Fish and seafood
FG01
Fresh fish and seafood
FG011
FRESH FISH AND SEAFOOD
FG02
Processed fish and seafood
FG021
PROCESSED FISH AND SEAFOOD
FH Eggs
FH01
Eggs
FH011
EGGS
DAIRY AND RELATED PRODUCTS
FJ Dairy and related products
FJ01
Milk
FJ011
MILK
FJ02
Cheese and related products
FJ021
CHEESE AND RELATED PRODUCTS
FJ03
Ice cream and related products
FJ031
ICE CREAM AND RELATED PRODUCTS
FJ04
Other dairy and related products
FJ041
OTHER DAIRY AND RELATED PRODUCTS
FRUITS AND VEGETABLES
FK Fresh fruits
FK01
Apples
FK011
APPLES
FK02
Bananas
FK021
BANANAS
FK03
Citrus fruits
FK031
CITRUS FRUITS
FK04
Other fresh fruits
FK041
OTHER FRESH FRUITS
FL Fresh vegetables
FL01
Potatoes
FL011
POTATOES
FL02
Lettuce
FL021
LETTUCE
FL03
Tomatoes
FL031
TOMATOES
FL04
Other fresh vegetables
FL041
OTHER FRESH VEGETABLES, INCLUDING FRESH HERBS
FM Processed fruits and vegetables
FM01
Canned fruits and vegetables and other shelf-stable fruits and vegetables
FM011
CANNED FRUITS AND VEGETABLES AND OTHER SHELF-STABLE FRUITS AND VEGETABLES
FM02
Frozen fruits and vegetables

69

Appendix 5. Major groups, expenditure classes, item stratum, and entry-level items—continued

FM021
FM03
FM031

FROZEN FRUITS AND VEGETABLES
Other processed fruits and vegetables, including dried
OTHER PROCESSED FRUITS AND VEGETABLES, INCLUDING DRIED
NONALCOHOLIC BEVERAGES AND BEVERAGE MATERIALS

FN Juices and nonalcoholic drinks
FN01
Carbonated drinks
FN011
CARBONATED DRINKS
FN02
Frozen noncarbonated juices and drinks
FN021
FROZEN NONCARBONATED JUICES AND DRINKS
FN03
Nonfrozen noncarbonated juices and drinks
NONFROZEN NONCARBONATED JUICES AND DRINKS
FN031
FP Beverage materials, including coffee and tea
FP01
Coffee
FP011
COFFEE
FP02
Other beverage materials, including tea
FP021
TEA
FP022
OTHER BEVERAGE MATERIALS
OTHER FOOD AT HOME
FR Sugar and sweets
FR01
Sugar and artificial sweeteners
FR011
SUGAR AND ARTIFICIAL SWEETENERS
FR02
Candy and chewing gum
FR021
CANDY AND CHEWING GUM
FR03
Other sweets
FR031
OTHER SWEETS
FS Fats and oils
FS01
Butter and margarine
FS011
BUTTER AND MARGARINE
FS02
Salad dressing
FS021
MAYONNAISE, SALAD DRESSING, AND SANDWICH SPREADS
FS03
Other fats and oils, including peanut butter
FS031
PEANUT BUTTER AND OTHER NUT BUTTERS
FS032
OTHER FATS AND OILS
FT Other foods
FT01
Soups
FT011
SOUPS
FT02
Frozen and freeze-dried prepared foods
FT021
FROZEN AND FREEZE-DRIED PREPARED FOODS
FT03
Snacks
FT031
SNACKS
FT04
Spices, seasonings, condiments, sauces
FT041
SALT AND OTHER SEASONINGS AND SPICES
FT042
OLIVES, PICKLES, RELISHES
FT043
SAUCES AND GRAVIES
FT044
OTHER CONDIMENTS (EXCLUDING OLIVES, PICKLES, AND RELISHES)
FT05
Baby food

70

Appendix 5. Major groups, expenditure classes, item stratum, and entry-level items—continued

FT051
FT06
FT061
FT062

BABY FOOD
Other miscellaneous foods
PREPARED SALADS
OTHER MISCELLANEOUS FOODS
FOOD AWAY FROM HOME

FV Food away from home
FV01
Full-service meals and snacks
FV011
FULL-SERVICE MEALS AND SNACKS
FV02
Limited-service meals and snacks
FV021
LIMITED-SERVICE MEALS AND SNACKS
FV03
Food at employee sites and schools
FV031
FOOD AT EMPLOYEE SITES AND SCHOOLS
FV04
Food from vending machines and mobile vendors
FV041
FOOD FROM VENDING MACHINES AND MOBILE VENDORS
FV05
Other food away from home
FV051
BOARD, CATERED EVENTS, AND OTHER FOOD AWAY FROM HOME
ALCOHOLIC BEVERAGES
FW Alcoholic beverages at home
FW01
Beer, ale, and other malt beverages at home
FW011
BEER, ALE, AND OTHER MALT BEVERAGES AT HOME
FW02
Distilled spirits at home
FW021
DISTILLED SPIRITS AT HOME
FW03
Wine at home
FW031
WINE AT HOME
FX Alcoholic beverages away from home
FX01
Alcoholic beverages away from home
FX011
ALCOHOLIC BEVERAGES AWAY FROM HOME

H

MAJOR GROUP: HOUSING
SHELTER

HA Rent of primary residence
HA01
Rent of primary residence
HA011
RENT OF PRIMARY RESIDENCE
HB Lodging away from home
HB01
Housing at school, excluding board
HB011
HOUSING AT SCHOOL, EXCLUDING BOARD
HB02
Other lodging away from home, including hotels and motels
HB021
RENTAL OF LODGING AWAY FROM HOME
HC Owners' equivalent rent of primary residence
HC01
Owners' equivalent rent of primary residence
HC011
OWNERS' EQUIVALENT RENT OF PRIMARY RESIDENCE
HC09
Unsampled owners' equivalent rent of secondary residence

71

Appendix 5. Major groups, expenditure classes, item stratum, and entry-level items—continued

HD Tenants' and household insurance
HD01
Tenants' and household insurance
HD011
TENANTS' AND HOUSEHOLD INSURANCE
FUELS AND UTILITIES
HE Fuel oil and other fuels
HE01
Fuel oil
HE011
FUEL OIL
HE02
Propane, kerosene, and firewood
HE021
OTHER HOUSEHOLD FUELS
HF Gas (piped) and electricity
HF01
Electricity
HF011
ELECTRICITY
HF02
Utility (piped) gas service
HF021
UTILITY (PIPED) GAS SERVICE
HG Water and sewer and trash collection services
HG01
Water and sewerage maintenance
HG011
RESIDENTIAL WATER AND SEWERAGE SERVICE
HG02
Garbage and trash collection
HG021
GARBAGE AND TRASH COLLECTION
HOUSEHOLD FURNISHINGS AND OPERATIONS
HH Window and floor coverings and other linens
HH01
Floor coverings
HH011
FLOOR COVERINGS
HH02
Window coverings
HH021
CURTAINS AND DRAPES
HH022
WINDOW COVERINGS
HH03
Other linens
HH031
BATHROOM LINENS
HH032
BEDROOM LINENS
HH033
KITCHEN AND DINING ROOM LINENS
HJ Furniture and bedding
HJ01
Bedroom furniture
HJ011
MATTRESS AND FOUNDATIONS
HJ012
BEDROOM FURNITURE OTHER THAN MATTRESS AND SPRINGS
HJ02
Living room, kitchen, and dining room furniture
HJ021
SOFAS, FURNITURE COVERS, AND DECORATIVE PILLOWS
HJ022
LIVING ROOM CHAIRS
HJ023
LIVING ROOM TABLES
HJ024
KITCHEN AND DINING ROOM FURNITURE
HJ03
Other furniture
HJ031
INFANTS' FURNITURE
HJ032
OUTDOOR FURNITURE
HJ033
OCCASIONAL FURNITURE
HJ09
Unsampled furniture
HJ090
RENTAL OF FURNITURE

72

Appendix 5. Major groups, expenditure classes, item stratum, and entry-level items—continued

HK Appliances
HK01
Major appliances
HK011
REFRIGERATORS AND HOME FREEZERS
HK012
LAUNDRY EQUIPMENT
HK013
RANGES AND COOKTOPS
HK014
MICROWAVE OVENS
HK02
Other appliances
HK021
FLOOR CLEANING EQUIPMENT
HK022
SMALL ELECTRIC KITCHEN APPLIANCES
HK023
OTHER ELECTRIC APPLIANCES
HK09
Unsampled appliances
HK090
PORTABLE DISHWASHERS
HL Other household equipment and furnishings
HL01
Clocks, lamps, and decorator items
HL011
LAMPS AND LIGHTING FIXTURES
HL012
HOUSEHOLD DECORATIVE ITEMS AND CLOCKS
HL02
Indoor plants and flowers
HL021
INDOOR PLANTS AND FRESH CUT FLOWERS
HL03
Dishes and flatware
HL031
DISHES
HL032
FLATWARE
HL04
Nonelectric cookware and tableware
NONELECTRIC COOKINGWARE
HL041
HL042
TABLEWARE AND NONELECTRIC KITCHENWARE
HM Tools, hardware, outdoor equipment, and supplies
HM01
Tools, hardware, and supplies
HM011
PAINT, WALLPAPER TOOLS, AND SUPPLIES
HM012
POWER TOOLS
HM013
MISCELLANEOUS HARDWARE, SUPPLIES, AND EQUIPMENT
NONPOWERED HANDTOOLS
HM014
HM02
Outdoor equipment and supplies
HM021
POWERED LAWN AND GARDEN EQUIPMENT, AND OTHER OUTDOOR ITEMS
HM022
LAWN AND GARDEN MATERIALS, OTHER DECORATIVE ITEMS, AND PESTICIDES
HM09
Unsampled tools, hardware, outdoor equipment, and supplies
HM090
UNSAMPLED ITEMS
HN Housekeeping supplies
HN01
Household cleaning products
HN011
LAUNDRY AND CLEANING PRODUCTS
HN012
LAUNDRY AND CLEANING EQUIPMENT
HN02
Household paper products
HN021
HOUSEHOLD PAPER PRODUCTS
HN03
Miscellaneous household products
HN031
MISCELLANEOUS HOUSEHOLD PRODUCTS
HP Household operations
HP01
Domestic services
HP011
DOMESTIC SERVICES
HP02
Gardening and lawn care services
HP021
GARDENING AND LAWN CARE SERVICES
HP03
Moving, storage, freight expense
HP031
MOVING, STORAGE, FREIGHT EXPENSE

73

Appendix 5. Major groups, expenditure classes, item stratum, and entry-level items—continued

HP04
HP041
HP042
HP043
HP09
HP090
A

Repair of household items
APPLIANCE REPAIR
REUPHOLSTERY OF FURNITURE
INSIDE HOME MAINTENANCE AND REPAIR SERVICES
Unsampled household operations
UNSAMPLED ITEMS
MAJOR GROUP: APPAREL
APPAREL COMMODITIES
MEN'S AND BOYS' APPAREL

AA Men's apparel
AA01
Men's suits, sport coats, and outerwear
AA011
MEN'S SUITS
AA012
MEN'S SPORT COATS AND TAILORED JACKETS
AA013
MEN'S OUTERWEAR
AA02
Men's furnishings
AA021
MEN'S UNDERWEAR, HOSIERY, AND NIGHTWEAR
AA022
MEN'S ACCESSORIES
AA023
MEN'S ACTIVE SPORTSWEAR
AA03
Men's shirts and sweaters
AA033
MEN'S SHIRTS, SWEATERS, AND VESTS
AA04
Men's pants and shorts
AA041
MEN'S PANTS AND SHORTS
AA09
Unsampled men's apparel
AA090
UNSAMPLED ITEMS

AB Boy's apparel
AB01
Boy's apparel
AB011
BOYS' OUTERWEAR
AB012
BOY'S SHIRTS AND SWEATERS
AB013
BOYS' UNDERWEAR, NIGHTWEAR, HOSIERY, AND ACCESSORIES
AB014
BOYS' SUITS, SPORT COATS, AND PANTS
AB015
BOYS' ACTIVE SPORTSWEAR
AB09
Unsampled boy's apparel
AB090
UNSAMPLED ITEMS
WOMEN'S AND GIRLS' APPAREL
AC Women's apparel
AC01
Women's outerwear
AC011
WOMEN'S OUTERWEAR
AC02
Women's dresses
AC021
WOMEN'S DRESSES
AC03
Women's suits and separates
AC031
WOMEN'S TOPS
AC032
WOMEN'S SKIRTS, PANTS, AND SHORTS
AC033
WOMEN'S SUITS AND COORDINATES
AC04
Women's underwear, nightwear, sportswear, and accessories
AC041
WOMEN'S UNDERWEAR AND OTHER UNDERGARMENTS
AC042
WOMEN'S HOSIERY AND ACCESSORIES

74

Appendix 5. Major groups, expenditure classes, item stratum, and entry-level items—continued

AC043
AC09
AC090

WOMEN'S ACTIVE SPORTSWEAR
Unsampled women's apparel
UNSAMPLED ITEMS

AD Girls' apparel
AD01
Girls' apparel
AD011
GIRLS' OUTERWEAR
AD012
GIRLS' DRESSES
AD013
GIRLS' TOPS
AD014
GIRLS' SKIRTS, PANTS, AND SHORTS
AD015
GIRLS' ACTIVE SPORTSWEAR
AD016
GIRLS' UNDERWEAR, SLEEPWEAR, HOSIERY, AND ACCESSORIES
AD09
Unsampled girls' apparel
AD090
UNSAMPLED ITEMS
AE Footwear
AE01
Men's footwear
AE011
MEN'S FOOTWEAR
AE02
Boys' and girls' footwear
AE021
BOYS' FOOTWEAR
AE022
GIRLS' FOOTWEAR
AE03
Women's footwear
AE031
WOMEN'S FOOTWEAR
AF Infants' and toddlers' apparel
AF01
Infants' and toddlers' apparel
AF011
INFANTS' AND TODDLERS' OUTER PLAY, DRESS, SLEEPWEAR, AND ACCESSORIES
AF012
INFANTS' AND TODDLERS' UNDERWEAR AND DIAPERS
AG Jewelry and watches
AG01
Watches
AG011
WATCHES
AG02
Jewelry
AG021
JEWELRY
T

MAJOR GROUP: TRANSPORTATION
PRIVATE TRANSPORTATION

TA New and used motor vehicles
TA01
New vehicles
NEW CARS AND TRUCKS
TA011
NEW MOTORCYCLES
TA012
TA02
Used cars and trucks
TA021
USED CARS AND TRUCKS
TA03
Leased cars and trucks
TA031
VEHICLE LEASING
TA04
Car and truck rental
TA041
AUTOMOBILE AND TRUCK RENTAL
TA09
Unsampled new and used motor vehicles
TA090
UNSAMPLED ITEMS

75

Appendix 5. Major groups, expenditure classes, item stratum, and entry-level items—continued

TB Motor fuel
TB01
Gasoline (all types)
TB011
REGULAR UNLEADED GASOLINE
TB012
MID-GRADE UNLEADED GASOLINE
TB013
PREMIUM UNLEADED GASOLINE
TB02
Other motor fuels
TB021
AUTOMOTIVE DIESEL FUEL
TB022
ALTERNATIVE MOTOR FUELS
TC Motor vehicle parts and equipment
TC01
Tires
TC011
TIRES
TC02
Vehicle accessories other than tires
TC021
VEHICLE PARTS AND EQUIPMENT OTHER THAN TIRES
TC022
MOTOR OIL, COOLANT, AND FLUIDS
TD Motor vehicle maintenance and repair
Motor vehicle body work
TD01
TD011
MOTOR VEHICLE BODY WORK
Motor vehicle maintenance and servicing
TD02
TD021
MOTOR VEHICLE MAINTENANCE AND SERVICING
Motor vehicle repair
TD03
TD031
MOTOR VEHICLE REPAIR
TD09
Unsampled vehicle maintenance and repair
TD090
UNSAMPLED SERVICE POLICIES
TE Motor vehicle insurance
Motor vehicle insurance
TE01
TE011
MOTOR VEHICLE INSURANCE
TF Motor vehicle fees
TF01
State motor vehicle registration and license fees
TF011
STATE MOTOR VEHICLE REGISTRATION AND LICENSE FEES
TF03
Parking and other fees
TF031
PARKING FEES AND TOLLS
TF032
AUTOMOBILE SERVICE CLUBS
TF09
Unsampled motor vehicle fees
TF090
UNSAMPLED ITEMS
PUBLIC TRANSPORTATION
TG Public transportation
Airline fare
TG01
TG011
AIRLINE FARES
TG02
Other intercity transportation
TG021
INTERCITY BUS FARES
TG022
INTERCITY TRAIN FARES
TG023
SHIP FARES
Intracity transportation
TG03
TG031
INTRACITY MASS TRANSIT
TG032
TAXI FARE
TG033
CAR AND VANPOOLS
TG09
Unsampled public transportation
TG090
UNSAMPLED ITEMS

76

Appendix 5. Major groups, expenditure classes, item stratum, and entry-level items—continued

M

MAJOR GROUP: MEDICAL CARE
MEDICAL CARE COMMODITIES

MF
MF01
MF011
MF02
MF021
MG
MG01
MG011
MG012
MG013
MG09

Medicinal drugs
Prescription drugs
PRESCRIPTION DRUGS
Nonprescription drugs
NONPRESCRIPTION DRUGS
Medical equipment and supplies
Medical equipment and supplies
DRESSINGS AND FIRST-AID KITS
MEDICAL EQUIPMENT FOR GENERAL USE
SUPPORTIVE AND CONVALESCENT MEDICAL EQUIPMENT
UNSAMPLED RENT OR REPAIR OF MEDICAL EQUIPMENT
MEDICAL CARE SERVICES

MC Professional services
MC01
Physicians' services
MC011
PHYSICIANS' SERVICES
MC02
Dental services
MC021
DENTAL SERVICES
MC03
Eyeglasses and eye care
MC031
EYEGLASSES AND EYE CARE
MC04
Services by other medical professionals
MC041
SERVICES BY OTHER MEDICAL PROFESSIONALS
MD Hospital and related services
MD01
Hospital services
MD011
HOSPITAL SERVICES
MD02
Nursing homes and adult daycare
NURSING AND CONVALESCENT HOME CARE
MD021
MD022
ADULT DAYCARE
MD03
Care of invalids and elderly at home
MD031
CARE OF INVALIDS, ELDERLY AND CONVALESCENTS IN THE HOME
ME01
Commercial health insurance
ME011
COMMERCIAL HEALTH INSURANCE, RETAINED EARNINGS
ME02
Blue Cross/Blue Shield
ME021
BLUE CROSS/BLUE SHIELD HEALTH INSURANCE, RETAINED EARNINGS
ME03
Health maintenance plans
ME031
HEALTH MAINTENANCE PLANS, RETAINED EARNINGS
ME04
Medicare and other health insurance
ME041
MEDICARE AND COMMERCIAL MEDICARE SUPPLEMENTS, RETAINED EARNINGS
R

MAJOR GROUP: RECREATION

RA Video and audio
RA01
Televisions
RA011
TELEVISIONS
RA02
Cable and satellite television and radio service
RA021
CABLE AND SATELLITE TELEVISION AND RADIO SERVICE

77

Appendix 5. Major groups, expenditure classes, item stratum, and entry-level items—continued

RA03
RA031
RA04
RA041
RA042
RA05
RA051
RA06
RA061
RA09
RA090

Other video equipment
OTHER VIDEO EQUIPMENT
Video discs, and other media, including rental of video and audio
PRERECORDED VIDEO DISCS/DIGITAL FILES/DOWNLOADS AND OTHER MEDIA
RENTAL OF VIDEO OR AUDIO DICS AND OTHER MEDIA
Audio equipment
AUDIO COMPONENTS, RADIOS, TAPE RECORDERS/PLAYERS, AND OTHER EQUIPMENT
Audio discs, tapes, and other media
AUDIO DISC TAPES, DIGITAL FILES, AND DOWNLOADS
Unsampled video and audio
UNSAMPLED ITEMS

RB Pets, pet products, and services
RB01
Pets and pet products
RB011
PET FOOD
RB012
PURCHASE OF PETS, PET SUPPLIES, ACCESSORIES
RB02
Pet services, including veterinary
RB021
PET SERVICES
RB022
VETERINARIAN SERVICES
RC Sporting goods
RC01
Sports vehicles, including bicycles
RC011
OUTBOARD MOTORS AND POWERED SPORTS VEHICLES
RC012
UNPOWERED BOATS AND TRAILERS
RC013
BICYCLES AND ACCESSORIES
RC02
Sports equipment
RC021
GENERAL SPORTS EQUIPMENT, EXCLUDING WATER
RC022
WATER SPORTS EQUIPMENT
RC023
HUNTING, FISHING, AND CAMPING EQUIPMENT
RC09
Unsampled sporting goods
RC090
UNSAMPLED ITEMS
RD Photography
RD01
Photographic equipment and supplies
RD011
FILM AND PHOTOGRAPHIC SUPPLIES
RD012
PHOTOGRAPHIC EQUIPMENT
RD02
Photographers and film processing
RD021
PHOTOGRAPHER'S FEES
RD022
FILM PROCESSING
RD09
Unsampled photography
RD090
UNSAMPLED RENT AND REPAIR OF PHOTOGRAPHIC EQUIPMENT
RE Other recreational goods
RE01
Toys and games
RE011
TOYS, GAMES, HOBBIES, AND PLAYGROUND EQUIPMENT
RE012
VIDEO GAME HARDWARE, SOFTWARE, AND ACCESSORIES
RE02
Sewing machines, fabric, and supplies
RE021
SEWING ITEMS
Music instruments and accessories
RE03
RE031
MUSIC INSTRUMENTS AND ACCESSORIES
RE09
Unsampled recreation services
RE090
UNSAMPLED ITEMS

78

Appendix 5. Major groups, expenditure classes, item stratum, and entry-level items—continued

RF Recreation services
RF01
Club membership dues and fees for participant sport and group exercises
RF011
CLUB DUES AND FEES FOR PARTICIPANT SPORTS AND GROUP EXERCISES
Admissions
RF02
RF021
ADMISSION TO MOVIES, THEATERS, CONCERTS, AND OTHER REOCCURING EVENTS
RF022
ADMISSION TO SPORTING EVENTS
Fees for lessons or instructions
RF03
RF031
FEES FOR LESSONS OR INSTRUCTIONS
RF09
Unsampled recreation services
RF090
UNSAMPLED ITEMS
RG Recreational reading materials
RG01
Newspapers and magazines
RG011
SINGLE-COPY NEWSPAPERS AND MAGAZINES
NEWSPAPER AND MAGAZINE SUBSCRIPTIONS
RG012
RG02
Recreational books
RG021
BOOKS PURCHASED THROUGH BOOK CLUBS
RG022
BOOKS PURCHASED AT RETAIL OUTLETS OTHER THAN BOOK CLUBS
RG09
Unsampled recreational reading materials
RG090
UNSAMPLED ITEMS
E

MAJOR GROUP: EDUCATION AND COMMUNICATION
EDUCATION

EA Educational books and supplies
EA01
Educational books and supplies
EA011
COLLEGE TEXTBOOKS
EA012
ELEMENTARY AND HIGH SCHOOL BOOKS AND SUPPLIES
EA013
ENCYCLOPEDIAS AND OTHER SETS OF REFERENCE BOOKS
EA09
Unsampled educational books and supplies
EA090
UNSAMPLED ITEMS
EB Tuition, other school fees, and childcare
EB01
College tuition and fees
EB011
COLLEGE TUITION AND FIXED FEES
EB02
Elementary and high school tuition and fees
EB021
ELEMENTARY AND HIGH SCHOOL TUITION AND FIXED FEES
EB03
Childcare and preschool
EB031
DAYCARE AND PRESCHOOL
EB04
Technical and business school tuition and fees
EB041
TECHNICAL AND BUSINESS SCHOOL TUITION AND FIXED FEES
EB09
Unsampled tuition, other school fees, and childcare
EB090
UNSAMPLED ITEMS
COMMUNICATION
EC Postage and delivery services
EC01
Postage
EC011
POSTAGE
EC02
Delivery services
EC021
DELIVERY SERVICES

79

Appendix 5. Major groups, expenditure classes, item stratum, and entry-level items—continued

Information and information processing
ED Telephone services
ED03
Wireless Telephone services
ED031
WIRELESS TELEPHONE SERVICES
ED04
Telephone services, land-line
ED041
TELEPHONE SERVICES, LANDLINE
EE Information and information processing other than telephone services
EE01
Personal computers and peripheral equipment
EE011
PERSONAL COMPUTERS AND PERIPHERAL EQUIPMENT
EE02
Computer software and accessories
EE021
COMPUTER SOFTWARE AND ACCESSORIES
EE03
Computer information processing services
EE031
INTERNET ACCESS AND OTHER INFORMATION SERVICES
EE04
Other information processing equipment
EE041
TELEPHONE, PERIPHERAL EQUIPMENT, AND ACCESSORIES
EE042
CALCULATORS, TYPEWRITERS, AND OTHER INFORMATION PROCESSING EQUIPMENT
EE09
Unsampled information and information processing
EE090
UNSAMPLED ITEMS
G

MAJOR GROUP: OTHER GOODS AND SERVICES

GA Tobacco and smoking products
GA01
Cigarettes
GA011
CIGARETTES
GA02
Tobacco products other than cigarettes
GA021
TOBACCO PRODUCTS OTHER THAN CIGARETTES
GA09
Unsampled tobacco and smoking products
GA090
UNSAMPLED ITEMS
GB Personal care products
Hair, dental, shaving, and miscellaneous personal care products
GB01
GB011
PRODUCTS AND NONELECTRIC ARTICLES FOR THE HAIR
GB012
DENTAL AND SHAVING PRODUCTS, INCLUDING NONELECTRIC ARTICLES
GB013
DEODORANT/SUNTAN PREPARATIONS, SANITARY/FOOTCARE PRODUCTS
GB014
ELECTRIC PERSONAL CARE APPLIANCES
GB02
Cosmetics, perfume, bath, nail preparations, and implements
GB021
COSMETICS, PERFUME, BATH, NAIL PREPARATIONS, and IMPLEMENTS
GB09
Unsampled personal care products
GB090
WOMEN'S HAIRPIECES/WIGS AND RENT/REPAIR PERS. CARE APPLS
GC Personal care services
Haircuts and other personal care services
GC01
GC011
HAIRCUTS AND OTHER PERSONAL CARE SERVICES
GD Miscellaneous personal services
GD01
Legal services
GD011
LEGAL SERVICES
Funeral expenses
GD02
GD021
FUNERAL EXPENSES
GD03
Laundry and dry cleaning services
GD031
LAUNDRY AND DRY CLEANING SERVICES
Apparel services other than laundry and dry cleaning
GD04

80

Appendix 5. Major groups, expenditure classes, item stratum, and entry-level items—continued

GD041
GD042
GD043
GD05
GD051
GD052
GD09
GD090

SHOE REPAIR AND OTHER SHOE SERVICES
CLOTHING ALTERATIONS, RENTALS, AND REPAIRS
WATCH AND JEWELRY REPAIR
Financial services
CHECKING ACCOUNTS AND OTHER BANK SERVICES
TAX RETURN PREPARATION AND OTHER ACCOUNTING FEES
Unsampled items
UNSAMPLED ITEMS

GE Miscellaneous personal goods
Miscellaneous personal goods
GE01
GE011
STATIONERY, STATIONERY SUPPLIES, AND GIFT WRAP
GE012
LUGGAGE
GE013
INFANTS' EQUIPMENT

81

CPI Appendix 6: Sample Allocation Methodology for Commodities and
Services
The primary objective of the commodities and services sample design is to determine values for all
sample design variables that minimize the sampling variance of 6-month price change for the
commodities and services portion of the CPI. The sample design variables are the number of entrylevel items (ELIs) to select in each item stratum and the number of outlets to select per Telephone
Point of Purchase Survey (TPOPS) category-replicate panel in each Primary Sampling Unit (PSU). To
that end, the variance of price change for the commodities and services portion of the CPI and the total
annual cost of data collection and processing are modeled as functions of the design variables. These
models allow the sample design problem to be expressed as that of minimizing the total variance of
price change, subject to various cost and sample allocation constraints. Within this framework,
nonlinear programming methods are used to solve the problem for optimal values of the sample design
variables.
Certain simplifying assumptions are made to render the problem tractable and operationally more
manageable. The number of PSUs, the number of replicate panels per PSU, and the classification of
ELIs into item strata have been determined in previous work (Williams et al., 1993; Lane, 1996). Item
strata are divided into 13 item groups for the design: four food-at-home groups (nonmeat staples; meat,
poultry, and fish; fruits and vegetables; and other food at home and alcoholic and nonalcoholic
beverages); food away from home; household furnishings and operations; fuels and utilities; apparel;
transportation less motor fuel; motor fuel; medical care; education and communications; and recreation
and other commodities and services. The 87 PSUs are divided into 15 groups according to size and
number of replicate panels. (See the table that follows.) It is assumed that the same item and outlet
sample sizes will apply to all PSUs within the same PSU group. This assumption reduces the allocation
problem to that of determining the number of ELI selections per replicate panel by PSU group and item
group [{Kij, i = 1, ..., 15, j = 1, ..., 13}, i = PSU group, j = item group], and the number of outlet
selections per TPOPS category per replicate by PSU group and item group {Mij, i = 1, ..., 15, j = 1, ...,
13}. These are the design variables.
PSU Groups for Commodities and Services Design
PSU
group
1.
2.

Name
New York City
New York City
suburbs

PSU
group
8.
9.

Name
Smaller self-representing PSUs
Non-self-representing PSUs,
Census Region 1
Non-self-representing PSUs,
Census Region 2
Non-self-representing PSUs,
Census Region 3
Non-self-representing PSUs,
Census Region 4
Smaller non-self-representing
PSUs, Census Regions 2–4
(there isn’t one for Region 1)
Los Angeles City

10.

3.

Los Angeles suburbs

11.

4.

Chicago

12.

5.

Philadelphia and
San Francisco

13.

6.

Detroit and Boston

7.

Other large selfrepresenting
PSUs

14.
.
15

Anchorage, AK, and Honolulu, HI

82

2
Let  Total
be the total price change variance for the commodities and services portion of the CPI, and

let CTotal be the total annual cost of data collection. Then the sample design problem can be
2
expressed as that of minimizing  Total
subject to the following cost and sample allocation
constraints:

CTotal ≤ Total data collection budget for commodities and services,
M ij ≥ 2 , i = 1,..., 15, j = 1 ,..., 13,
K ij ≥ Number of item strata in PSU group i, item group j, i = 1,..., 15, j = 1,..., 13,
K ij ≤Maximum number of item hits in PSU group i, item group j, i = 1,..., 15,

j = 1,..., 13, and
Average number of item hits per stratum–index area in PSU group i, item group  9,
i = 1, ..., 15, j = 1, ..., 13.
A detailed description of BLS sample allocation methods follows.
The sampling variance function
Variance component models attempt to allocate parts of the total sampling variance to different sources
of variation. For the commodities and services item–outlet sample, the following four sources of
variation are modeled: PSU selection, item selection, outlet selection, and a residual component that
includes other sources, such as sampling within the outlet.
The variance function for the commodities and services sample design is modeled for index areas. Each
self-representing PSU is a single index area. Non-self-representing PSUs represent seven index areas,
with the sample for each area represented by 2 to 22 PSUs. In the equations that follow, both k,
representing what we shall call a super-index area, and k', representing an index area, are used. The
only difference between the two is that the three index areas for smaller non-self-representing PSUs by
Census region are combined into one super-index area for all small non-self-representing PSUs. As
mentioned earlier, the variance model assumes that the total variance of price change for item group j
within super-index area k can be expressed as a sum of four components:
2
2
2
2
 2j , k   psu
, j , k   item, j , k   outlet , j , k   error , j , k .

In this equation,
2
 psu
, j ,k

is the component of variance due to sampling of PSUs in non-self-representing areas,

2
 item
, j ,k

is the component of variance due to sampling of ELIs within item strata,

2
 outlet
, j ,k

is the component of variance due to sampling of outlets, and

2
 error
, j,k

is a residual component of variance that includes disaggregation, the final stage of within-outlet
item selection .

Similarly, it is assumed that the variance of price change of an individual sampled unit or quote has the
same structure:
2
2
2
2
2
 unit
, j ,k   unit, psu , j , k   unit,item , j ,k   unit,outlet , j ,k   unit,error , j ,k .

83

In this equation,
2
 unit
, j ,k

is the total variance of price change of an individual sampled unit or quote for item j in
super-index area k,

2
 unit,
psu , j , k

is the component of unit variance due to sampling of PSUs in non-self-representing areas,

2
σ unit,
item , j , k

is the component of unit variance due to sampling of ELIs within item strata,

2
σ unit,
outlet , j , k

is the component of unit variance due to sampling of outlets, and

2
σ unit,
error, j,k

is the corresponding residual component of unit variance.

Thus, the projected sampling variance for a given index area k in PSU group i is

 2 PC k  
13

∑∑RI
j =1 k '∈k

2
j ,k

σ
σ
+
f (K , N ) f (M
2

2

unit, item , j , k

unit, outlet , j , k

ij

1

k

2

ij

, K ij , N k )

+

σ

2

f (M
3

σ
+
) f (N )
2

unit, error , j , k ′
ij

, K ij , N k

unit, psu, j , k ′

4

,

k'

where

f
f
f
f

1

2

3

4

(K , N ) = (N , H , K ) ,
(M , K , N ) = [(N H M ' + N H M NPV ) NRO ],
(M , K , N ) = [(N H M K ) (NRQV )], and
ij

k

k

k

ij

ij

ij

k

k

k

ij

ij

k

k

k

i, j

ij

k

ij

k

i, j

j

j

j

(N ) = the number of non-self -representing PSUs in index area k' ,
k'

in which
PCk

is the price change in super-index area k,

Hk

is the number of replicate panels per PSU in the super-index area k,

NROj

is the outlet initiation response rate for major group j,

NRQVj

is the quote quote-level response rate for major group j for variance projection,

NPVj

is the weighted sum of non-POPS categories in major group ji, with each category
weighted by its probability of selection, for variance projection,

M ij'

is the number of unique in-scope outlets selected per PSU replicate, modeled as the
quadratic function

2

M ij  ( AVij M ij  BVij M ij )
'

of the outlet sample size, where AVij

and BVij are coordinates that are determined by the unique outlet cost predictor
function program, and
Nk

is the number of PSUs in super-index area k.

84

The sampling variance of price change for the All U.S. City Average commodities and services index
is
2
 Total


 RI2j,k wk2 2j,k ,
j

k

where
RI j,k is the relative importance of item group j in super-index area k, scaled to sum to 1.0 over all

commodities and services item groups, and

wk is the 1990 Census population weight of super-index area k.
Relative importances of item groups are obtained from the most recent 2 years of the Consumer
Expenditure Survey and are the proportion of total expenditures in super-index area k that come from
item group j.
The cost function

The modeled costs of the commodities and services portion of the CPI are the costs of initiation data
collection and travel and of pricing data collection (personal visit and telephone) and travel. Each of
these models is developed in terms of outlet- and quote-related costs and as a function of the design
decision variables.
Initiation costs

Outlet related initiation costs. For PSU group i and major group j, outlet-related costs for initiation are
given by

(

)

CIO M ij , Kij = 0.25Ni

(

CIO M ij , Kij

)

Hi

(CO j + COTj ) (ACij M ij + BCij M ij2 + NPCj M ij ) , where

is the outlet-related initiation cost for major group j in PSU group i,

Ni

is the number of PSUs in group i,

Hi

is the number of replicates per PSU in PSU group i,

COj

is the compensation initiation cost per outlet for major group j,

NPCj

is the weighted sum of non-POPS categories in major group j, with each category
weighted by its probability of selection,

COTj

is the per diem and mileage cost per outlet for major group j,

and ACij M ij  BCij M ij2  is a quadratic overlap function used to predict the number of unique sample
outlets and to account for the overlap of elements in the outlet sample within and between major
groups for a replicate panel. Like AVij and BVij, ACij and BCij are coordinates determined by the unique
outlet cost predictor function program. The number 0.25 accounts for the rotation or reinitiation of the
outlet sample in one-fourth of the sample TPOPS categories’ PSUs each year.
Quote related-initiation costs. Quote related initiation costs are given by
CIQ M ij , K ij = 0.25 Ni H i WOD j CQ j M ij K ij NRO j , where

(

)

CI Q ( M ij , K ij )

is the quote-related cost of initiation for major group j in PSU group i,

WODj

is a seasonal items initiation factor for major group j,

CQj

is the initiation cost per quote for major group j, and

NROj

is as before.
85

Repricing costs

The costs of ongoing price data collection and processing are also developed as both outlet- and quoterelated costs.
Outlet-related repricing costs.For PSU group i and major group j, outlet-related costs for ongoing
pricing are given by
CPO (Mij , Kij ) = MBOij Ni Hi NROj ( ACij M ij + BCij M ij2 + NPCj Mij ) (CPVOj + CPOj ) (1 RTOj ) + CTOj RTOj ], where

CPO(Mij,Kij)

is the total outlet-related cost for ongoing pricing for major group j in PSU group I,

CPVOj

is the compensation cost (time spent in travel) for a personal visit for pricing per outlet
for major group j,

CPOj

is the travel cost (per diem and mileage) for a personal visit for pricing per outlet for
major group j (equal to 3.33 for every j),

RTOj

is the proportion of outlets priced by telephone for major group j,

CTOj

is the per-outlet cost for telephone collection, (equal to 3.43 for every j initially),

NPOj

is the weighted sum of non-POPS categories in major group j, with each category
weighted by its probability of selection for cost projections,

MBOij

is a factor used to adjust for the monthly–bimonthly mix of outlets by PSU and major
group, and NROj and NPCj are as before.

Quote-related repricing costs. Quote-related costs for ongoing pricing are given by
CP Q ( M ij , K ij ) = MBQ

ij

Ni

Hi

M ij

K ij

NRQC

j

[ CPVQ

j

(1

RTQ j ) + CTQ

j

RTQ j ] , where

CPQ(Mij,Kij)

is the total quote-related cost for ongoing pricing,

MBQij

is a factor to adjust for the monthly–bimonthly mix of quotes by PSU and major
product group.

CPVQj

is the per-quote cost (compensation not spent in travel) for a personal visit for
pricing,

RTQ,j

is the proportion of telephone collected quotes for major group j,

CTQ,j

is the per quote cost for telephone collection for major group j, and

NRQCj

is the quote-level response rate for projecting costs for major group j.

Total cost function

The total cost function associated with data collection for commodities and services, summed over all
item groups and PSU groups, is then given by

CTotal =

∑[CI
i, j

O

(M ij , K ij ) + CIQ ( M ij , K ij ) + CPO ( M ij , K ij ) + CPQ ( M ij , K ij )],

2
and the sample design problem can be expressed as that of minimizing the total variance,  Total
,
subject to the following constraints:

CTotal ≤Total expenditure limit,

M ij ≥2 , i = 1, ..., 15, j = 1, ...,13,
K ij  Number of item strata in PSU group i, item group j,i = 1, ..., 15, j = 1, ..., 13,
86

K ij ≤Maximum number of item hits in PSU group i, item group j, i = 1 ,..., 15, j = 1, ..., 13,

and
Average number of item hits per stratum-index area in PSU group i, item group  9,
i = 1, ..., 15, j = 1, ..., 13.
We note here that the last set of constraints is added to address concerns regarding small-sample bias at
the elementary index level by ensuring a minimum average sample allocation of nine expected quotes
in total per index area–item stratum combination.
Model coefficients

The parameters of the cost function are estimated from agency administrative records dating from
fiscal year 1996 forward and a Time and Travel Study conducted by the BLS Office of Field
Operations (OFO). Distinctions between personal visits and telephone collection of data are made on
the basis of information from OFO and from an analysis of commodities and services microdata
conducted within the BLS Prices Statistical Methods Division. Response rates for each item group are
derived from field initiation records and ongoing pricing experience.
Because outlet samples are selected independently for each TPOPS category and outlets may be listed
in the sample frames for more than one TPOPS category, an individual outlet may be selected more
than once. For example, a grocery store could be selected for both bakery products and dairy products.
Thus, the number of unique outlets realized by the sampling process is needed to project outlet-related
costs. Quadratic regressions are used to predict the number of unique outlets realized in sample
selection as a function of designated sample size. The regressions are developed and reevaluated with
each rotation by using the most current sampling frames available for each item to model the number
of unique outlets obtained in simulations of sampling procedures for each PSU and item group as a
function of designated sample sizes.
Components of price change variance are computed with the use of restricted maximum-likelihood
estimation methods with commodities and services price microdata, the most recent estimates being
based on price data collected in 2009–2011. Component estimates are developed for 6-month price
changes for the 13 item groups for each index area and month. Mean unit components of variance
estimates are then computed by averaging the unit components of variance across months.
Solutions

Solutions are found with the SAS procedure PROC OPTMODEL. For each item group, the number of
item selections is bounded below by the number of strata in the item group.
ELI selections are then distributed among item strata within each item group, with consideration given
to differences in relative importance, estimates of variances for production stratum-level price change
and response rates among the item strata within each item group, as well as special problems identified
by commodity analysts and field staff. Similarly, designated outlet sample sizes are distributed among
the various TPOPS categories in item groups in order to manage variations in expected response rates
and respondent burden.
In general, recent sample designs have shifted resources in many item groups from sampling many
outlets to sampling fewer outlets, with more item selections per outlet. This shift is due primarily to the
large residual component of price change sampling variance estimated for most item groups, coupled
with a trend of an increasing number of unique outlets realized in TPOPS sampling.

87

Appendix 7. Point-of-Purchase Survey (POPS) categories1
The following tabulation lists retail entry-level items (ELIs) and numbers, by category, in the BLS point-ofPurchase Survey (POPS):
POPS Category and title, ELI number and title
F14

Fresh fish or fresh seafood
FG011 Fresh fish and seafood

F15

Processed fish or processed seafood, including frozen, canned, or cooked
FG021 Processed fish and seafood

F20

Miscellaneous dairy products, including yogurt, powdered milk, or coffee creamers
FJ041 Other dairy and related products

F29

Canned fruits or vegetables
FM011 Canned and other shelf-stable fruits and vegetables

F30

Frozen fruits or vegetables
FM021 Frozen fruits and vegetables

F31

Dried or other processed fruits and vegetables
FM031 Other processed fruits and vegetables, including dried

F32

Carbonated Drinks
FN011 Carbonated Drinks

F33

Noncarbonated juices or drinks, frozen and non frozen
FN021 Frozen noncarbonated juices and drinks
FN031 Non frozen noncarbonated juices and drinks

F34

Roasted, instant, or freeze-dried coffee
FP011 Coffee

F35

Powdered drinks, tea, cocktail mixes, or ice
FP021 Tea
FP022 Other beverage materials

F36

Sugar or artificial sweeteners
FR011 Sugar and artificial sweeteners

F37

Candy or chewing gum
FR021 Candy and chewing gum

1
88

Appendix 7. Point-of-Purchase Survey (POPS) categories—continued
F38

Jellies, syrups, honey, molasses, marshmallows, icings, or fudge mixes
FR031 Other sweets

F39

Butter or margarine
FS011 Butter and margarine

F40

Salad dressing or mayonnaise
FS021 Mayonnaise, salad dressing, and sandwich spreads

F41

Peanut butter, or cooking fats and oils
FS031 Peanut butter and other nut butters
FS032 Other fats and oils

F47

Baby food
FT051 Baby food

F48

Prepared salads or salad bars, excluding restaurants
FT061 Prepared salads

F49

Easy-to-prepare canned or packaged foods, excluding fruits, vegetables, and soups
FT062 Other miscellaneous foods

F50

Full-service meals or snacks, such as meals at sit-down restaurants
FV011 Full-service meals and snacks

F51

Limited-service meals or snacks, such as meals from fast food restaurants or delivered meals
FV021 Limited-service meals and snacks

F52

Meals or snacks at schools or employer-provided cafeterias, dining rooms, or snack bars
FV031 Food at employee sites and schools

F53

Food or beverages from vending machines or mobile vendors
FV041 Food from vending machines and mobile vendors

F54

Catered events or board
FV051 Board, catered events, and other food away from home

F55

Beer, ale, sake, or other malt beverages for home use
FW011 Beer, ale, and other malt beverages at home

F56

Hard liquor for home use
FW021 Distilled spirits at home

F57

Wine for home use
FW031 Wine at home

F58

Alcoholic beverages served in bars, restaurants, clubs, or similar places
FX011 Alcoholic beverages away from home

F59

Cereal, rice, pasta, cornmeal or flour
FA011 Flour and prepared flour mixes
FA021 Breakfast cereal
FA031 Rice, pasta, and cornmeal

2
89

Appendix 7. Point-of-Purchase Survey (POPS) categories—continued
F60

Bread, cookies, or other baked goods, including frozen
FB011 Bread
FB021 Fresh biscuits, rolls, and muffins
FB031 Cakes and cupcakes (excluding frozen)
FB032 Cookies
FB041 Crackers, and bread and cracker products
FB042 Sweet rolls, coffeecakes, and doughnuts (excluding frozen)
FB043 Frozen bakery products and frozen/refrigerated doughs and batters
FB044 Pies, Tarts, and turnovers (excluding frozen)

F61

Meat, including beef, poultry, pork, or lunch meats
FC011 Uncooked ground beef
FC021 Uncooked beef roasts
FC031 Uncooked beef steaks
FC041 Other uncooked beef and veal
FD011 Bacon, breakfast sausage, and related products
FD021 Ham
FD031 Pork chops
FD041 Other pork, including roasts and picnics
FE011 Frankfurters
FE012 Lunch meats
FE013 Lamb, organ meats, and game
FF011 Chicken
FF021 Other poultry, including turkey

F62

Milk, eggs, cheese, ice cream, or frozen yogurt
FH011 Eggs
FJ011 Milk
FJ021 Cheese and cheese products
FJ031 Ice cream and related products

F63

Fresh Fruits or Vegetables
FK011 Apples
FK021 Bananas
FK031 Citrus fruits
FK041 Other fresh fruits
FL011 Potatoes
FL021 Lettuce
FL031 Tomatoes
FL041 Other fresh vegetables, including fresh herbs

F64

Frozen prepared foods, chips, nuts, or other snacks or soups
FT011 Soups and soup bases
FT021 Frozen and freeze-dried prepared foods
FT031 Snacks

F65

Condiments, spices, sauces, or gravies
FT041 Salt and other seasonings and spices
FT042 Olives, pickles, and relishes
FT043 Sauces and gravies
FT044 Other condiments (excluding olives, pickles, and relishes)

3
90

Appendix 7. Point-of-Purchase Survey (POPS) categories—continued
H02

Lodging away from home, such as hotels, motels, or vacation rentals
HB021 rental of lodging away from home

H03

Fuel oil for household heating
HE011 Fuel oil

H04

Propane, firewood, coal, or charcoal used for household heating or cooking
HE021 Other household fuels

H05

Residential water or sewer service
HG011 Residential water and sewerage service

H06

Garbage or trash collection service
HG021 Garbage and trash collection

H07

Floor coverings, such as hard surface tiling, carpets, or scatter rugs
HH011 Floor coverings

H08

Window coverings, such as curtains, drapes, or blinds
HH021 Curtains and drapes
HH022 Window coverings

H09

Household linens, such as kitchen or bathroom towels, bedding, or tablecloths
HH031 Bathroom linens
HH032 Bedroom linens
HH033 Kitchen and dining room linens

H10

Bedroom furniture, including mattresses or springs
HJ011 Mattresses and foundations
HJ012 Bedroom furniture other than mattresses and springs

H14

Infants’ furniture
HJ031 infants’ furniture

H15

Outdoor furniture
HJ032 outdoor furniture

H16

Other furniture, including entertainment centers, bookcases, or desks
HJ033 Occasional furniture

H17

Refrigerators or home freezers
HK011 Refrigerators and home freezers

H18

Washers or dryers
HK012 Washers and dryers

H19

Stoves, Ovens, or microwave ovens
HK013 Ranges and cooktops
HK014 Microwave ovens

H20

Vacuums or other electric floor-cleaning equipment
HK021 Floor-cleaning equipment

4
91

Appendix 7. Point-of-Purchase Survey (POPS) categories—continued
H21

Small electric kitchen appliances or clothing irons
HK022 Small electric kitchen appliances

H22

Heating or cooling equipment or home safety devices
HK023 Other electric appliances

H23

Household decorative items, including clocks or lamps
HL011 Lamps and lighting fixtures
HL012 Household decorative items and clocks

H24

Dishes, glassware, or flatware
HL031 Dishes
HL032 Flatware

H25

Nonelectric kitchen utensils, cookware, or bake ware
HL041 Non electric cooking ware
HL042 Table ware and nonelectric kitchenware

H26

Indoor plants or fresh-cut flowers
HL021 Indoor plants and fresh-cut flowers

H27

Paint, Wallpaper tools, or related supplies
HM011 Paint, wallpaper tools, and supplies

H28

Power Tools
HM012 Power tools

H29

Non powered tools or miscellaneous hardware
HM013 Miscellaneous hardware, supplies, and equipment
HM014 Non powered tools

H30

Barbeque grills, powered lawn and garden equipment, or other outdoor items
HM021 Lawn and garden equipment and outdoor equipment and grills

H31

Lawn and garden supplies or insecticides
HM022 Lawn and garden materials, other decorative items, and pesticides

H32

Household laundry and cleaning products or supplie
HN011 Laundry and cleaning products
HN012 Laundry and cleaning equipment

H33

Paper napkins, paper towels, facial tissue, or toilet paper
HN021 Household paper products

H34

Other disposable products, such as plastic or foil wraps, garbage bags, paper plates,
batteries, or light bulbs
HN031Miscellaneous household products

H35

Housekeeping services
HP011 Domestic Services

H36

Gardening or Lawn Care Services
HP021 Gardening and lawn care services

5
92

Appendix 7. Point-of-Purchase Survey (POPS) categories—continued
H37

Moving, storage, or freight services
HP031 Moving, storage, and freight expense

H38

Major appliance repair
HP041 Appliance repair

H39

Reupholstery of household furniture
HP042 Reupholstery of furniture

H40

Inside home maintenance or repair
HP043 Inside home maintenance and repair services

H41

Living room, dining room, or kitchen furniture
HJ021 Sofas, furniture covers, and decorative pillows
HJ022 Living room chairs
HJ023 Living room tables
HJ024 Kitchen and dining room furniture

A01

Men’s suits or blazers
AA011 Men’s suits
AA012 Men’s sport coats and tailored jackets

A02

Men’s coats or jackets
AA013 Men’s outerwear

A03

Men’s socks, underwear, sleepwear, or bathrobes
AA021 Men’s underwear, hosiery, nightwear, and loungewear

A04

Men’s accessories, such as ties, belts, or wallets
AA022 Men’s accessories

A05

Men’s active sportswear, such as exercise apparel or bathing suits
AA023 Men’s active sportswear

A06

Men’s shirts, sweaters, or vests
AA033 Men’s shirts, sweaters, and vests

A07

Men’s pants, jeans, or shorts
AA041 Men’s pants and shorts

A08

Boys’ clothing or accessories
AB011 Boys’ outerwear
AB012 Boys’ shirts and sweaters
AB013 Boys’ underwear, sleepwear, hosiery, and accessories
AB014 Boys’ suits, sport coats, and pants
AB015 Boys’ active sportswear

A09

Women’s Outerwear
AC011 Women’s outerwear

A10

Women’s dresses
AC021 Women’s dresses

6
93

Appendix 7. Point-of-Purchase Survey (POPS) categories—continued
A11

Women’s tops, skirts, pants, or shorts
AC031 Women’s tops
AC032 Women’s skirts, pants, and shorts

A12

Women’s suits or suit components
AC033 Women’s suits and coordinates

A13

Women’s underwear or nightwear
AC041 Women’s underwear, nightwear, and other undergarments

A14

Women’s hosiery or accessories
AC042 Women’s hosiery and accessories

A15

Women’s active sportswear, such as exercise apparel or bathing suits
AC043 Women’s active sportswear

A16

Girls’ clothing or accessories
AD011 Girls’ outerwear
AD012 Girls’ dresses
AD013 Girls’ tops
AD014 Girls’ skirts, pants, and shorts
AD015 Girls’ active sportswear
AD016 Girls’ underwear, sleepwear, hosiery, and accessories

A17

Men’s footwear
AE011 Men’s footwear

A18

Boys’ or girls’ footwear
AE021 Boys’ footwear
AE022 Girls’ footwear

A19

Women’s footwear
AE031 Women’s footwear

A20

Infants’ and toddlers’ clothing or accessories, excluding underwear and diapers
AF011 Infants’ and toddlers’ outer, play, dress, and sleepwear and accessories

A21

Infants’ and toddlers’ underwear or diapers
AF012 Infants’ and toddlers’ underwear and diapers

A22

Watches
AG011 Watches

A23

Jewelry
AG021 Jewelry

T01

New cars, trucks, or vans
TA011 New car and truck purchase

T02

New motorcycles
TA012 New motorcycles

7

94

Appendix 7. Point-of-Purchase Survey (POPS) categories—continued
T03

Leased cars, trucks, or vans
TA031 Vehicle leasing

T04

Car, truck, or van rental
TA041 Automobile and truck rental

T05

Gasoline, diesel, or alternative fuels
TB011 Regular unleaded gasoline
TB012 Midgrade unleaded gasoline
TB013 Premium unleaded gasoline
TB021 Automotive diesel fuel
TB022 Alternative motor fuels

T07

Tires
TC011 Tires

T08

Vehicle parts or accessories
TC021 Vehicle parts and equipment other than tires

T09

Motor oil, coolants, or fluids
TC022 Motor oil, coolants, and fluids

T10

Motor vehicle bodywork
TD011 Motor vehicle bodywork

T11

Motor vehicle maintenance, inspections, or towing
TD021 Motor vehicle maintenance and servicing

T12

Motor vehicle repair
TD031 Motor vehicle repair

T15

Parking fees or tolls
TF031 Parking fees and tolls

T16

Automobile service clubs
TF032 Automobile service clubs

T17

Ship travel or passenger cruises
TG023 Ship fares

T18

Intracity mass transit
TG031 Intracity mass transit

T19

Taxi or cab fare
TG032 Taxi fare

T20

Car or van pools
TG033 Car and van pools

M01

Prescription drugs
MF011 Prescription drugs

M02

Nonprescription drugs
MF021 Nonprescription drug
8
95

Appendix 7. Point-of-Purchase Survey (POPS) categories—continued
M03 Medical equipment, supplies, or dressings
MG011 Dressings and first-aid kits
MG012 Medical equipment for general use

MG013 Supportive and convalescent medical equipment
M04

Physician’s services
MC011 Physician’s services

M05

Dental services
MC021 Dental services

M06

Eyeglasses or eye care
MC031 Eyeglasses and eye care

M07

Services by other medical professionals
MC041 Services by other medical professionals

M08

Hospital services
MD011 Hospital services

M09

Adult daycare
MD022 Adult daycare

R01

Televisions
RA011 Televisions

R02

Cable or satellite television or radio service
RA021 Cable and satellite television and radio service

R03

DVD players, camcorders, or other video equipment
RA031 Other video equipment

R04

Prerecorded video, such as dvds, digital files, downloads, and other media
RA041 Prerecorded video discs, digital files, downloads, and other media

R05

Rental of DVDs or video games
RA042 Rental of video or audio discs, tapes, digital files, or downloads

R06

Audio equipment for automobiles or home
RA051 Audio components, radios, tape recorders/players, and other equipment

R07

Prerecorded or blank audiotapes, CDs, or records
RA061 Audio discs and tapes, prerecorded and blank

R08

Pet food
RB011 Pet food

R09

Pets, pet supplies, or accessories
RB012 Purchase of pets, pet supplies, and accessories

R10

Pet services, such as grooming, boarding, or training
RB021 Pet services

9
96

Appendix 7. Point-of-Purchase Survey (POPS) categories—continued
R11

Veterinarian services
RB022 Veterinarian services

R12

Outboard motors or powered sports vehicles
RC011 Outboard motors and powered sports vehicles

R13

Unpowered boats or trailers
RC012 Unpowered boats and trailers

R14

Bicycles or bicycling accessories
RC013 Bicycles and accessories

R15

General sports equipment
RC021 General sports equipment, excluding water
RC022 Water sports equipment

R16

Hunting, fishing, or camping equipment
RC023 Hunting, fishing, and camping equipment

R17

Film or film development supplies or disposable cameras
RD011 Film and photographic supplies

R18

Cameras or other photographic equipment, excluding film
RD012 Photographic equipment

R19

Photographer’s fees
RD021 Photographer’s fees

R20

Digital photo prints or film development
RD022 Film processing

R21

Toys, games, hobby supplies, or playground equipment
RE011 Toys, games, hobbies, and playground equipment

R22

Video game hardware, games, or accessories
RE012 Video game hardware, software, and accessories

R23

Sewing machines, fabric, or sewing supplies
RE021 Sewing items

R24

Musical instruments or musical accessories
RE031 Musical instruments and accessories

R25

Club membership dues for fraternal or civic organizations or fees for participant sports
RF011 Club dues and fees for participant sports and group exercises

R26

Admissions, such as to movies, concerts, or theme parks
RF021 Admission to movies, theaters, concerts, and other reoccurring events

R27

Admissions to sporting events
RF022 Admissions to sporting events

R28

Recreational lessons or instruction
RF031 Fees for lessons or instruction
10
97

Appendix 7. Point-of-Purchase Survey (POPS) categories—continued
R29

Single-copy newspapers or magazines
RG011 Single-copy newspapers and magazines

R30

Newspaper or magazine subscriptions
RG012 Newspaper and magazine subscriptions

R31
Books purchased through book clubs
RG021 Books purchased through book clubs
R32

Other books, audiobooks, or e books
RG022 Books purchased at other than book clubs

E01

College or university-level textbooks
EA011 College textbooks

E02

Elementary or high school textbooks
EA012 Elementary and high school books and supplies

E03

Encyclopedias or other sets of reference books
EA013 Encyclopedias and other sets of reference books

E04

Tuition or fixed fees for a college or university
EB011 College tuition and fixed fees

E05

Tuition or fixed fees for private elementary or high schools
EB021 Elementary and high school tuition and fixed fees

E06

Daycare providers, including nursery schools
EB031 Daycare and preschool

E07

Tuition or fixed fees for technical or vocational schools
EB041 Technical and business school tuition and fixed fees

E08

Delivery services
EC021 Delivery services

E11

Personal computers or peripheral equipment
EE011 Personal computers and peripheral equipment

E12

Computer software, computer accessories, and blank media, including memory cards,
recordable discs and other forms
EE021 Computer software and accessories

E13

Accessing the Internet at home and away, including separate or bundled charges
EE031 Internet access and other information services

E14

Home or cellular telephones, answering machines, or other phone accessories
EE041Telephones, peripheral equipment, and accessories

E15

Calculators, typewriters, or other information-processing equipment
EE042 Calculators, typewriters, and other information-processing equipment

11
98

Appendix 7. Point-of-Purchase Survey (POPS) categories—continued
E16

Cellular telephone service
ED031 Wireless phone service

E17

Local or long-distance landline telephone service, including
Prepaid phone cards
ED041 Telephone service landline

G01

Cigarettes
GA011 Cigarettes

G02

Cigars, smoking tobacco, or chewing tobacco
GA021 Tobacco products other than cigarettes

G03

Hair Products, such as shampoo, sprays, pins, or combs
GB011 Products and nonelectric articles for the hair

G04

Dental or shaving products
GB012 Dental and Shaving Products, Including Nonelectric Articles

G05

Deodorant, feminine hygiene products, suntan lotions, or foot care products
GB013 Deodorant/suntan preparations and sanitary/foot care products

G06

Electric personal care appliances, such as shavers or hair dryers
GB014 Electric personal care appliances

G07 Cosmetics, perfumes, or bath and nail preparations
GB021 Cosmetics/perfume/bath/nail preparations and implements
G08 Personal care services, such as haircuts, nail services, or tanning
GC011 Haircuts and other personal care services
G09

Legal services
GD011 Legal fees

G10

Funeral services
GD021 Funeral expenses

G11

Laundry or dry cleaning services
GD031 Laundry and dry cleaning services

G12

Shoe repair or other shoe services
GD041 Shoe repair and other shoe services

G13

Clothing rental, alterations, or repairs
GD042 Clothing alterations, rentals, and repairs

G14

Watch or jewelry repair
GD043 Watch and jewelry repair

G15

Checking account fees, credit card fees, or other bank services
GD051 Checking accounts and other bank services

12
99

Appendix 7. Point-of-Purchase Survey (POPS) categories—continued
G16

Tax return preparation or other accounting services
GD052 Tax return preparation and other accounting services

G18

Stationary, school supplies, or gift wrap
GE011 Stationery, stationery supplies, and gift wrap

G19

Luggage, briefcases, or other carrying cases
GE012 Luggage

G20

Infants’ equipment, such as strollers, car seats, bottles, or dishes
GE013 Infants’ equipment

1

Some POPS category numbers have been changed to reflect changes in item composition.

13
100

Appendix 8. Non-Point-of-Purchase Survey (Non-POPS) sample designs
For each non-POPS entry-level item (e.g., electricity), the following information is given:
1.

Source of the universe data

2.

Sampling unit for outlets

3.

Measure of size

4.

Desired final pricing unit

EC011 Postage
1.
2.
3.
4.

The distribution of household mail by type of postal service and postal zone, as determined by the
postal service in the Household Mailstream Study, Final Report, prepared for the U.S. Postal Service.
U.S. Postal Service
Postal revenue for each type of postal service and postal zone
Specific postal service and zones traveled

MD031 Care of invalids, elderly, and convalescents in the home
1.
2.
3.
4.

Center for Medicare and Medicaid Services (CMS) Home Healthcare Care system
Facilities providing adult home care in each CPI sample area
Each facility within a given CPI sample area has an equal chance of being selected.
Specific services provided

HD011/TE011 Insurance—tenants/motor vehicle
1.
2.
3.
4.

National Association of Insurance Commissioners
Insurance companies serving the states in which CPI sample areas are located
Total revenue for noncommercial policies, by type of insurance
Specific policy serving within each CPI sample area

HF011 Electricity
1.
2.
3.
4.

a. Consumer Expenditure (CE) survey
b. Direct information from the regional field offices
Electric utility companies reported in the CE survey or electric utility companies serving each of the
CPI sample areas
Expenditures for electricity, as reported in the CE survey, or an estimate of the number of residential
customers provided by the field
Specific type of service for a specific number of kilowatt hours

HF021 Utility natural gas
1.
2.
3.
4.

a. Consumer Expenditure (CE) survey
b. Direct information supplied by the regional field offices
Gas utility companies reported in the CE survey or gas companies serving each of the CPI sample areas
Expenditures for natural gas, as reported in the CE survey or an estimate of the number of residential
customers provided by the field
Specific type of service and specific number of cubic feet or therms of gas

1
101

Appendix 8. Non-Point-of-Purchase Survey(Non-POPS) sample designs—continued
MD021 Nursing and convalescent home care
1.
2.
3.
4.

Center for Medicare and Medicaid Services (CMS) Nursing Home Care system
Facilities providing nursing home care in each CPI sample area
Number of beds
Specific accommodations and services provided

TA021 Used vehicles
1.
2.
3.
4.

J. D. Power data
Selection of used vehicles, based on vehicle sales data from J. D. Powers
Expenditures reported for used vehicles in the CE survey
Specific used vehicles with specific options; prices collected from the National Automobile Dealers
Association (NADA) Official Used Car Guide

TF011 State vehicle registration and driver’s license
1.
2.
3.
4.

Each state’s Department of Motor Vehicles
State motor vehicle department in each CPI sample area
Revenue generated by each type of fee
Specific class/vehicle registration, type of license, or vehicle property tax

TG011 Airline fares
1.
2.
3.
4.

Department of Transportation data file consisting of a 10-percent sample of all passenger itineraries
originating in the United States
All airlines providing service from any CPI sample area
Total number of nonbusiness passengers per airline, per trip itinerary, per fare class
Specific trip itinerary and fare class for the selected airline

TG021 Intercity bus fares
1.
2.
3.
4.

Scheduled intercity bus trips from each CPI sample area
Bus companies serving each CPI sample area
Number of buses that leave for a given destination
Specific trip itinerary and fare class

TG022 Intercity train service
1.
2.
3.
4.

Data file of intercity train trips provided by Amtrak and the Alaskan Railroad
Amtrak and the Alaskan Railroad
Number of tickets sold
Specific trip and class

2
102

Table 1 (2011-2012 Weights). Relative importance of components in the Consumer Price Indexes: U.S.

city average,9.
December
Appendix
Relative2014
importance of components in the Consumer Price Indexes,
2011−2012
weights: U.S. city average, December 2014
(Percent of all items)
U.S. city average

Item and group

CPI-U

CPI-W

Expenditure category
All items ..........................................................................................................

100.000

100.000

Food and beverages .....................................................................................
Food ...........................................................................................................

15.272
14.257

16.011
15.052

Food at home ...........................................................................................
Cereals and bakery products ..................................................................
Cereals and cereal products .................................................................
Flour and prepared flour mixes ...........................................................
Breakfast cereal ..................................................................................
Rice, pasta, cornmeal .........................................................................
Bakery products ....................................................................................
Bread ..................................................................................................
Fresh biscuits, rolls, muffins ...............................................................
Cakes, cupcakes, and cookies ...........................................................
Other bakery products ........................................................................
Meats, poultry, fish, and eggs .................................................................
Meats, poultry, and fish ........................................................................
Meats ..................................................................................................
Beef and veal ....................................................................................
Uncooked ground beef ...................................................................
Uncooked beef roasts .....................................................................
Uncooked beef steaks ....................................................................
Uncooked other beef and veal ........................................................
Pork ..................................................................................................
Bacon, breakfast sausage, and related products ...........................
Ham ................................................................................................
Pork chops ......................................................................................
Other pork including roasts and picnics ..........................................
Other meats ......................................................................................
Poultry ................................................................................................
Chicken .............................................................................................
Other poultry including turkey ...........................................................
Fish and seafood ................................................................................
Fresh fish and seafood .....................................................................
Processed fish and seafood .............................................................
Eggs .....................................................................................................
Dairy and related products .....................................................................
Milk .......................................................................................................
Cheese and related products ...............................................................
Ice cream and related products ............................................................
Other dairy and related products ..........................................................
Fruits and vegetables .............................................................................
Fresh fruits and vegetables ..................................................................
Fresh fruits ..........................................................................................
Apples ...............................................................................................
Bananas ...........................................................................................
Citrus fruits .......................................................................................
Other fresh fruits ...............................................................................
Fresh vegetables ................................................................................
Potatoes ...........................................................................................
Lettuce ..............................................................................................
Tomatoes ..........................................................................................
Other fresh vegetables .....................................................................
Processed fruits and vegetables ..........................................................
Canned fruits and vegetables .............................................................
Frozen fruits and vegetables ..............................................................
Other processed fruits and vegetables including dried .......................
Nonalcoholic beverages and beverage materials ...................................
Juices and nonalcoholic drinks .............................................................
Carbonated drinks ..............................................................................
Frozen noncarbonated juices and drinks ............................................
Nonfrozen noncarbonated juices and drinks ......................................
Beverage materials including coffee and tea ........................................
Coffee .................................................................................................
Other beverage materials including tea ..............................................

8.427
1.138
.370
.048
.197
.126
.767
.230
.116
.189
.233
2.014
1.880
1.229
.582
.238
.085
.207
.053
.372
.141
.078
.064
.089
.275
.360
.294
.066
.291
.148
.142
.134
.898
.283
.286
.126
.204
1.379
1.076
.575
.083
.087
.146
.259
.500
.075
.072
.102
.251
.303
.157
.088
.057
.955
.699
.285
.014
.400
.256
.158
.099

9.259
1.255
.420
.052
.221
.147
.835
.247
.121
.208
.259
2.376
2.221
1.478
.714
.292
.098
.253
.071
.438
.161
.091
.083
.103
.326
.439
.369
.069
.305
.149
.156
.155
.950
.320
.287
.132
.211
1.419
1.088
.580
.086
.102
.159
.233
.509
.080
.077
.116
.236
.331
.177
.091
.063
1.092
.831
.348
.018
.465
.261
.150
.110

103

Table 1 (2011-2012 Weights). Relative importance of components in the Consumer Price Indexes: U.S.

city average,9.
December
2014-Continued
Appendix
Relative
importance of components in the Consumer Price Indexes,
2011−2012
weights: U.S. city average, December 2014─continued
(Percent of all items)
U.S. city average

Item and group

CPI-U

CPI-W

Expenditure category
Other food at home .................................................................................
Sugar and sweets .................................................................................
Sugar and artificial sweeteners ..........................................................
Candy and chewing gum ....................................................................
Other sweets ......................................................................................
Fats and oils .........................................................................................
Butter and margarine ..........................................................................
Salad dressing ....................................................................................
Other fats and oils including peanut butter .........................................
Other foods ...........................................................................................
Soups .................................................................................................
Frozen and freeze dried prepared foods ............................................
Snacks ................................................................................................
Spices, seasonings, condiments, sauces ...........................................
Baby food ...........................................................................................
Other miscellaneous foods .................................................................

2.043
.299
.054
.185
.060
.245
.077
.062
.107
1.499
.093
.285
.330
.292
.055
.444

2.167
.302
.066
.178
.058
.270
.075
.067
.128
1.595
.093
.315
.345
.302
.084
.456

Food away from home ..............................................................................
Full service meals and snacks ................................................................
Limited service meals and snacks ..........................................................
Food at employee sites and schools ......................................................
Food from vending machines and mobile vendors .................................
Other food away from home ...................................................................
Alcoholic beverages .................................................................................
Alcoholic beverages at home .................................................................
Beer, ale, and other malt beverages at home ......................................
Distilled spirits at home .........................................................................
Wine at home .......................................................................................
Alcoholic beverages away from home ....................................................

5.830
2.823
2.413
.212
.064
.319
1.015
.597
.274
.073
.250
.418

5.793
2.361
2.830
.233
.092
.276
.959
.564
.364
.055
.145
.395

Housing ........................................................................................................
Shelter ........................................................................................................
Rent of primary residence ........................................................................
Lodging away from home .........................................................................
Housing at school, excluding board ........................................................
Other lodging away from home including hotels and motels ..................
Owners’ equivalent rent of residences .....................................................
Owners’ equivalent rent of primary residence ........................................
Unsampled owners’ equivalent rent of secondary residences ...............
Tenants’ and household insurance ...........................................................
Fuels and utilities ........................................................................................
Household energy ....................................................................................
Fuel oil and other fuels ...........................................................................
Fuel oil ..................................................................................................
Propane, kerosene, and firewood .........................................................
Energy services ......................................................................................
Electricity ..............................................................................................
Utility (piped) gas service .....................................................................
Water and sewer and trash collection services ........................................
Water and sewerage maintenance .........................................................
Garbage and trash collection ..................................................................
Household furnishings and operations .......................................................
Window and floor coverings and other linens ...........................................
Floor coverings .......................................................................................
Window coverings ..................................................................................
Other linens ............................................................................................
Furniture and bedding ..............................................................................
Bedroom furniture ...................................................................................
Living room, kitchen, and dining room furniture ......................................
Other furniture ........................................................................................
Unsampled furniture ...............................................................................
Appliances ................................................................................................
Major appliances ....................................................................................
Other appliances ....................................................................................
Unsampled appliances ...........................................................................
Other household equipment and furnishings ............................................
Clocks, lamps, and decorator items .......................................................
Indoor plants and flowers .......................................................................

42.173
32.711
7.159
.839
.172
.666
24.339
22.918
1.421
.375
5.273
4.051
.236
.139
.097
3.815
2.940
.875
1.222
.945
.277
4.189
.266
.047
.053
.166
.769
.268
.363
.128
.009
.271
.147
.120
.004
.479
.257
.107

40.464
31.105
9.800
.461
.080
.380
20.511
19.967
.544
.333
5.903
4.578
.208
.118
.090
4.369
3.436
.933
1.326
1.037
.288
3.455
.208
.028
.040
.140
.666
.233
.318
.096
.019
.276
.152
.118
.006
.443
.267
.083

104

Table 1 (2011-2012 Weights). Relative importance of components in the Consumer Price Indexes: U.S.

city average,9.December
Appendix
Relative2014-Continued
importance of components in the Consumer Price Indexes,
2011−2012
weights: U.S. city average, December 2014─continued
(Percent of all items)
U.S. city average

Item and group

CPI-U

CPI-W

Expenditure category
Dishes and flatware ................................................................................
Nonelectric cookware and tableware ......................................................
Tools, hardware, outdoor equipment and supplies ...................................
Tools, hardware and supplies .................................................................
Outdoor equipment and supplies ............................................................
Unsampled tools, hardware, outdoor equipment and supplies ...............
Housekeeping supplies ............................................................................
Household cleaning products .................................................................
Household paper products .....................................................................
Miscellaneous household products ........................................................
Household operations ...............................................................................
Domestic services ..................................................................................
Gardening and lawncare services ..........................................................
Moving, storage, freight expense ...........................................................
Repair of household items ......................................................................
Unsampled household operations ..........................................................

.041
.074
.710
.189
.367
.154
.847
.337
.247
.263
.848
.279
.279
.116
.066
.107

.030
.063
.574
.198
.247
.129
.910
.391
.273
.246
.378
.073
.117
.081
.055
.052

Apparel .........................................................................................................
Men’s and boys’ apparel .............................................................................
Men’s apparel ...........................................................................................
Men’s suits, sport coats, and outerwear .................................................
Men’s furnishings ....................................................................................
Men’s shirts and sweaters ......................................................................
Men’s pants and shorts ..........................................................................
Unsampled men’s apparel ......................................................................
Boys’ apparel ............................................................................................
Women’s and girls’ apparel ........................................................................
Women’s apparel ......................................................................................
Women’s outerwear ................................................................................
Women’s dresses ...................................................................................
Women’s suits and separates ................................................................
Women’s underwear, nightwear, sportswear and accessories ..............
Unsampled women’s apparel .................................................................
Girls’ apparel ............................................................................................
Footwear .....................................................................................................
Men’s footwear .........................................................................................
Boys’ and girls’ footwear ...........................................................................
Women’s footwear ....................................................................................
Infants’ and toddlers’ apparel .....................................................................
Jewelry and watches ..................................................................................
Watches ....................................................................................................
Jewelry .....................................................................................................

3.343
.834
.653
.104
.185
.196
.160
.007
.181
1.439
1.210
.118
.155
.550
.378
.010
.229
.725
.218
.178
.329
.135
.211
.046
.164

3.595
.973
.737
.095
.198
.223
.207
.014
.236
1.427
1.151
.121
.146
.544
.328
.012
.276
.821
.297
.220
.304
.193
.181
.075
.106

Transportation ..............................................................................................
Private transportation .................................................................................
New and used motor vehicles ..................................................................
New vehicles ..........................................................................................
Used cars and trucks ..............................................................................
Leased cars and trucks ..........................................................................
Car and truck rental ................................................................................
Unsampled new and used motor vehicles ..............................................
Motor fuel ..................................................................................................
Gasoline (all types) .................................................................................
Other motor fuels ....................................................................................
Motor vehicle parts and equipment ..........................................................
Tires ........................................................................................................
Vehicle accessories other than tires .......................................................
Motor vehicle maintenance and repair .....................................................
Motor vehicle body work .........................................................................
Motor vehicle maintenance and servicing ..............................................
Motor vehicle repair ................................................................................
Unsampled service policies ....................................................................
Motor vehicle insurance ............................................................................
Motor vehicle fees ....................................................................................
State motor vehicle registration and license fees ...................................
Parking and other fees ...........................................................................
Unsampled motor vehicle fees ...............................................................
Public transportation ...................................................................................
Airline fare ................................................................................................
Other intercity transportation ....................................................................

15.289
14.167
5.720
3.551
1.591
.397
.073
.109
3.979
3.904
.075
.435
.285
.150
1.168
.057
.492
.587
.032
2.300
.565
.312
.235
.018
1.122
.702
.157

18.015
17.211
6.886
3.527
2.827
.353
.048
.130
5.214
5.100
.114
.532
.320
.213
1.206
.066
.494
.607
.040
2.837
.535
.336
.186
.013
.805
.442
.090

105

Table 1 (2011-2012 Weights). Relative importance of components in the Consumer Price Indexes: U.S.

city average,9.
December
2014-Continued
Appendix
Relative
importance of components in the Consumer Price Indexes,
(Percent
of
all
items)
2011−2012 weights: U.S. city average, December 2014─continued
U.S. city average

Item and group

CPI-U

CPI-W

Expenditure category
Intracity transportation ..............................................................................
Unsampled public transportation ..............................................................

.260
.004

.267
.006

Medical care .................................................................................................
Medical care commodities ..........................................................................
Medicinal drugs ........................................................................................
Prescription drugs ...................................................................................
Nonprescription drugs ............................................................................
Medical equipment and supplies ..............................................................
Medical care services .................................................................................
Professional services ................................................................................
Physicians’ services ...............................................................................
Dental services .......................................................................................
Eyeglasses and eye care .......................................................................
Services by other medical professionals ................................................
Hospital and related services ...................................................................
Hospital services ....................................................................................
Nursing homes and adult day services ...................................................
Care of invalids and elderly at home ......................................................
Health insurance .......................................................................................

7.716
1.772
1.696
1.345
.351
.076
5.944
3.032
1.590
.804
.284
.354
2.159
1.853
.174
.132
.753

6.308
1.423
1.378
1.108
.270
.045
4.885
2.474
1.303
.699
.237
.235
1.738
1.634
.080
.024
.673

Recreation ....................................................................................................
Video and audio ..........................................................................................
Televisions ................................................................................................
Cable and satellite television and radio service ........................................
Other video equipment .............................................................................
Video discs and other media, including rental of video and audio ............
Audio equipment .......................................................................................
Audio discs, tapes and other media .........................................................
Unsampled video and audio .....................................................................
Pets, pet products and services .................................................................
Pets and pet products ...............................................................................
Pet services including veterinary ..............................................................
Sporting goods ...........................................................................................
Sports vehicles including bicycles ............................................................
Sports equipment .....................................................................................
Unsampled sporting goods .......................................................................
Photography ...............................................................................................
Photographic equipment and supplies .....................................................
Photographers and film processing ..........................................................
Unsampled photography ..........................................................................
Other recreational goods ............................................................................
Toys ..........................................................................................................
Sewing machines, fabric and supplies .....................................................
Music instruments and accessories ..........................................................
Unsampled recreation commodities .........................................................
Other recreation services ...........................................................................
Club dues and fees for participant sports and group exercises ................
Admissions ...............................................................................................
Fees for lessons or instructions ................................................................
Unsampled recreation services ................................................................
Recreational reading materials ...................................................................
Newspapers and magazines ....................................................................
Recreational books ...................................................................................
Unsampled recreational reading materials ...............................................

5.750
1.847
.133
1.468
.029
.090
.066
.044
.016
1.058
.659
.399
.400
.181
.214
.005
.120
.058
.062
.001
.381
.277
.050
.042
.011
1.724
.602
.640
.211
.271
.220
.123
.094
.002

5.131
2.054
.148
1.668
.031
.099
.054
.040
.015
.959
.680
.279
.430
.214
.211
.006
.095
.034
.059
.002
.387
.291
.048
.031
.017
1.063
.325
.501
.128
.109
.144
.081
.062
.000

Education and communication .....................................................................
Education ....................................................................................................
Educational books and supplies ...............................................................
Tuition, other school fees, and childcare ..................................................
College tuition and fees ..........................................................................
Elementary and high school tuition and fees ..........................................
Child care and nursery school ................................................................
Technical and business school tuition and fees .....................................
Unsampled tuition, other school fees, and childcare ..............................
Communication ...........................................................................................

7.062
3.325
.203
3.122
1.853
.377
.725
.039
.128
3.737

6.875
2.544
.202
2.342
1.183
.253
.772
.042
.092
4.332

106

Table 1 (2011-2012 Weights). Relative importance of components in the Consumer Price Indexes: U.S.

city average,9.
December
2014-Continued
Appendix
Relative
importance of components in the Consumer Price Indexes,
(Percent of all items)
2011−2012
weights: U.S. city average, December 2014─continued
U.S. city average

Item and group

CPI-U

CPI-W

Expenditure category
Postage and delivery services ..................................................................
Postage ..................................................................................................
Delivery services ....................................................................................
Information and information processing ...................................................
Telephone services ................................................................................
Wireless telephone services .................................................................
Land-line telephone services ................................................................
Information technology, hardware and services .......................................
Personal computers and peripheral equipment ......................................
Computer software and accessories ......................................................
Internet services and electronic information providers ...........................
Telephone hardware, calculators, and other consumer information
items ...............................................................................................
Unsampled information and information processing ..............................

.144
.130
.014
3.593
2.462
1.624
.837
1.132
.272
.068
.711

.104
.095
.009
4.228
3.030
2.199
.831
1.198
.249
.054
.819

.068
.012

.062
.014

Other goods and services .............................................................................
Tobacco and smoking products ..................................................................
Cigarettes .................................................................................................
Tobacco products other than cigarettes ...................................................
Unsampled tobacco and smoking products ..............................................
Personal care .............................................................................................
Personal care products .............................................................................
Hair, dental, shaving, and miscellaneous personal care products .........
Cosmetics, perfume, bath, nail preparations and implements ................
Unsampled personal care products ........................................................
Personal care services .............................................................................
Haircuts and other personal care services .............................................
Miscellaneous personal services ..............................................................
Legal services .........................................................................................
Funeral expenses ...................................................................................
Laundry and dry cleaning services .........................................................
Apparel services other than laundry and dry cleaning ............................
Financial services ...................................................................................
Unsampled items ....................................................................................
Miscellaneous personal goods .................................................................

3.394
.718
.661
.050
.006
2.676
.724
.369
.348
.007
.638
.638
1.122
.316
.173
.276
.034
.228
.095
.192

3.600
1.187
1.111
.066
.009
2.413
.709
.375
.325
.008
.576
.576
.953
.250
.123
.288
.022
.189
.080
.176

100.000
37.880
22.608
13.658
10.315
8.950
62.120
32.336
5.625
11.955
85.743
67.289
92.284
23.623
14.673
11.330
28.930
2.619
29.784
56.176
8.030
91.970
77.713
19.408
4.215
58.305
7.094
10.089

100.000
41.147
25.136
15.131
11.536
10.005
58.853
30.772
5.784
11.005
84.948
68.895
93.692
26.095
16.090
12.495
31.142
2.774
28.081
53.968
9.792
90.208
75.156
20.672
5.423
54.484
7.761
11.197

Special aggregate indexes
All items ..........................................................................................................
Commodities .................................................................................................
Commodities less food and beverages ......................................................
Nondurables less food and beverages .....................................................
Nondurables less food, beverages, and apparel ....................................
Durables ...................................................................................................
Services ........................................................................................................
Rent of shelter ..............................................................................................
Transportation services ................................................................................
Other services ..............................................................................................
All items less food .........................................................................................
All items less shelter .....................................................................................
All items less medical care ...........................................................................
Commodities less food .................................................................................
Nondurables less food ..................................................................................
Nondurables less food and apparel ..............................................................
Nondurables .................................................................................................
Apparel less footwear ...................................................................................
Services less rent of shelter .........................................................................
Services less medical care services .............................................................
Energy ..........................................................................................................
All items less energy .....................................................................................
All items less food and energy ....................................................................
Commodities less food and energy commodities .....................................
Energy commodities ...............................................................................
Services less energy services ..................................................................
Domestically produced farm food .................................................................
Utilities and public transportation ..................................................................

107