View original document

The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.

.

2

.

3

;

0

.

1 3 4

-

2

.

BLS
Handbook
of Methods

Volume II
The Consumer Price Index

4Wj=STm ic s o y ^ n
3 RAR'

^ P O S lT D m H ^



APR 1 8 1984

Volume II
The Consumer Price Index

Method:
U.S. Department of Labor
Raymond J. Donovan, Secretary
Bureau of Labor Statistics
Janet L. Norwood, Commissioner
April 1984
Bulletin 2134-2




For sale by the Superintendent of Documents, U.S. Government Printing Office
W ashington, D.C. 20402

.




The BLS Handbook o f Methods presents in two
volumes detailed explanations of how the Bureau of
Labor Statistics obtains and prepares the economic data
it publishes. Volume I, published in December 1982,
contains the information for all bls programs except
consumer prices. Volume II contains this information
for the Consumer Price Index and also describes in
detail how the current index was constructed.
Volume II was written by the members of the staff of
the Office of Prices and Living Conditions under the




direction of Kenneth V. Dalton, Associate Commis­
sioner. Major contributors include Kenneth P. Archer,
John F. Early, Patrick C. Jackman, Curtis A. Jacobs,
Herbert C. Morton, John P. Sommers, and William L.
Weber. It was prepared for publication by Rosalind
Springsteen and Rosalie Epstein in the Division of
Special Publications, Office of Publications.
Material in this publication is in the public domain
and, with appropriate credit, may be reproduced
without permission.

Comftdimts

Page
Volume II: The Consumer Price Index
Introduction...........................................................................................................................................

1

Part I. The index in b rie f.......................................................................................................................
H istory.............................................................................................................................................
Concepts..........................................................................................................................................
Prices and living costs.............................................................................................................
Sam pling..................................................................................................................................
Weights and relative im portance..........................................................................................
Scope and calculation.....................................................................................................................
U ses..................................................................................................................................................
Analysis and presentation..............................................................................................................
Limitations of the index.................................................................................................................

3
3
4
4
5
5
5
6
6
6

Part II. Construction of the index........................................................................................................
Definition of the index....................................................................................................................
The Consumer Expenditure Survey..............................................................................................
Sample design.........................................................................................................................
Sample sizes and response ra te s.............................................................................................
Weighting the questionnaires................................................................................................
CPI sample areas and publication a reas..................................................................................
CPI item structure and item sam pling..........................................................................................
CPI outlet and price surveys..........................................................................................................
Point-of-Purchase Survey.....................................................................................................
CPI outlet sampling procedures.............................................................................................
Non-pops sampling procedures ............................................................................................
Selection procedures within outlets......................................................................................
Rent survey.....................................................................................................................................
Owner surveys.................................................................................................................................
Estimation of price change............................................................................................................
Food, commodities, and services..........................................................................................
R e n t..........................................................................................................................................
Property t a x .............................................................................................................................
House prices and mortgage interest......................................................................................
Quality adjustments, linking, and im putation.....................................................................
Timing of data collection, seasonal items, and average prices...........................................
Estimation of the CPI cost-population weights............................................................................
Preliminary mean expenditures......................................................................................
Composite estimation process..............................................................................................
Raking .....................................................................................................................................
Special cost weight procedures for housing.........................................................................
Rental equivalence..........................................................................................................................
Sample selection.....................................................................................................................
Data collection........................................................................................................................

7
7
8
8
8
9
10
10
10
11
12
13
14
14
15
15
15
16
16
17
17
19
20
20
20
21
21
21
21
22




IV

Contents — Continued
Page
Calculation of cost weights.................................................................................................... 22
Calculation of monthly index and relative change............................................................... 22
Precision of estim ates..................................................................................................................... 22
Basic approach to cpi variance estim ation........................................................................... 22
Estimation of relatives by half-sam ple................................................................................. 23
Replicate cost weights............................................................................................................. 23
Calculation of variances for the index and price change.................................................... 23
Technical references................................................................................................................................

25

Tables:
1.
2.

13

3.

Response rates at initiation for cpi commoditiesand services in outlet surveys..............
Designated outlet sample sizes for each half-sample within a market
basket for cpi commodities and services......................................................................
eli’s priced monthly everywhere.......................................................................................

Appendixes:
A.
Chronology of changes in the Consumer Price Index, 1980 to d a te ...............................
B.
Relative importance of components in the Consumer Price Indexes: U.S. city
average December 1977 .................................................................................................
C.
Non-pops non-household sample designs.........................................................................
D.
Pricing cycles for sample a re a s...........................................................................................
E.
Seasonal adjustment methodology at BLS........................................................................




v

14
19

26
28
31
34
36

Introduetion

and other cooperating organizations. An important
aspect of the work of the regional staffs has been ex­
plaining the concepts and techniques which the Bureau
uses in compiling the statistics.

When U.S. Commissioner of Labor Carroll Wright
issued his first annual report in March 1886, he
established the policy of explaining his statistical
methods to his readers and of seeking to avoid misinter­
pretation of the figures presented. During the 98 years
which have followed that initial report, the definitions,
methods, and limitations of the data published by the
Bureau of Labor and its successor, the Bureau of Labor
Statistics, have been explained again and again. The
reason for this is not merely to make the readers aware
of the known limitations of the statistics, but also to in­
struct them in the proper use of the information and to
assure them that proper standards have been observed.
This volume continues that tradition by providing
detailed descriptions of the Bureau’s statistical series.
The Bureau’s role, organization, and staff, and its ap­
proach to its data collection activities also are discussed
briefly.

Staff
The Bureau’s work extends beyond the initial collec­
tion and processing of data. Its findings frequently in­
fluence, and sometimes are crucial to, the determining
and shaping of public policy. Over the years, it has
developed a staff of professional analysts, trained to
search out the implications of survey findings for the
welfare of workers and to present them as cogently and
as promptly as possible in written and oral form. How
successfully this can be accomplished depends greatly
upon the competence of the analysts and their suppor­
ting personnel.
In bls, analytical and statistical work is performed by
economists, statisticians, and mathematical statisticians
with the aid of an experienced corps of programmers,
systems analysts, and other professionals. For analytical
work, economists at even the lowest grade level must
meet requirements roughly equivalent to a college major
in economics. There are comparable requirements for
other professionals. Great efforts are made to hire the
best qualified persons available.
The Bureau provides training needed for on-the-job
skills, as background to special assignments, to keep
professionals abreast of changes in their fields, and to
aid higher level and executive professionals in obtaining
the best results from their staffs. In training staff, a
special effort is made to impart detailed knowledge of
the techniques used in collecting and compiling the
statistics, so that maximum application of the data to
current problems can be made without a risk of ex­
ceeding the limits of their significance.

BIS role
Among Federal agencies collecting and issuing
statistics, the Bureau of Labor Statistics has been term­
ed a general-purpose statistical collection agency. The
Bureau’s figures are prepared to serve the needs of
business, labor, Congress, the general public, and the
administrative and executive agencies for information
on economic and social trends, bls statistics are often
quite specialized, yet they meet general economic and
social data requirements. As the needs of users are likely
to differ, no statistic is ideal for all. This makes it im­
portant that the characteristics of the measures and their
limitations be well understood.
Organization
The statistical program s of the Bureau were
developed, for the most part, independently of each
other, taking on characteristics suited to the re­
quirements of the subject under observation. As a
result, the Bureau was organized according to subjectmatter areas, an arrangement which has proved efficient
and has been continued over the years. Expertise in
techniques, economic analysis, and other staff activities
across subject-matter lines was added to provide better
use of the Bureau’s resources.
As the Bureau’s collection activities increased,
regional offices were established to administer the field
programs, to disseminate data to local users, and to fur­
nish technical advice and assistance to State agencies




Consultation and advice
A statistical program too much detached from the
users of its data may fail in its principal mission. To
avoid sterility, the Bureau continuously invites advice
and ideas from users and experts in business, labor, and
academic organizations and from members of the
public. Over the years, the advice the Commissioner of
Labor Statistics has received on policy and technical mat­
ters from responsible parties, relating to the collection
and analysis of Bureau statistics, has been very helpful.
Of course, decisions on statistical policy have always
been the final responsibility of the Commissioner.
1

tion closely related to their daily affairs and their per­
sonal lives. To some who have supplied the desired in­
formation, the Bureau has gone back often for later in­
formation on the same subject or for new types of infor­
mation. The response has been remarkable in its generosi­
ty. In no small measure, the cooperation received is due
to the Bureau’s pledge of confidentiality to the firm or
the person supplying the information. The policy of not
identifying respondents is implemented by combining
the data reported by the different sources and issuing
the findings in summary form or nonidentified micro
tabulations.
Another assurance given respondents is that their
reports will be used for statistical purposes only. At­
tempts to “ break” this policy, by organizations or in­
dividuals who wanted access to the data and were will­
ing to go to the courts to secure it, have been successful­
ly resisted.1 A similar problem occurs when an ad­
ministrative agency of government seeks court action to
compel a company to release its file copy of information
provided in confidence to a statistical agency.2
While it cannot be proved that these policies result in
more reliable statistics, Bureau Commissioners and
their staffs over the years have been convinced from ex­
perience that it is so. It is notable that some other
Federal agencies (especially the Bureau of the Census),
well equipped with authority to compel the submittal of
certain reports, rarely if ever invoke this power. Rather,
they also choose to rely on persuasion. The Bureau of
Labor Statistics, while its functions as a statistical agen­
cy are prescribed by law,3 has always relied upon volun­
tary cooperation of respondents in collecting informa­
tion.

In order to keep in touch with the current and ancitipated needs of business and labor groups and to seek
advice on technical problems, the Commissioner
established standing research advisory committees in
1947. These groups, now called the Business Research
Advisory Council and the Labor Research Advisory
Council, advise on technical problems and provide
perspectives on Bureau programs in relation to needs of
their members. The councils accomplish their work in
general sessions and also through committees on
specialized subject-matter fields. Committees are
augmented by persons in industry or labor who,
although not council members, have special com­
petence. The councils may take formal action through
resolutions or recommendations on appropriate mat­
ters, but such resolutions are merely advisory. Members
of the councils and the subcommittees serve in their in­
dividual capacities, not as representatives of their
organizations.
The members of the Business Research Advisory
Council are designated by the Commissioner under
authorization of the Secretary of Labor, after nomina­
tion by the National Association of Manufacturers, the
U.S. Chamber of Commerce, the Business Round
Table, and the National Federation of Independent
Business. The members of the Labor Research Advisory
Council are designated by the Commissioner of Labor
Statistics under authorization of the Secretary of Labor,
from nominations by the Director of Research, afl-cio,
on behalf of affiliated unions, and by other unions upon
invitation of the Commissioner.
The Bureau often seeks the advice of professional
economists, statisticians, social scientists, educators,
and others, either in their individual capacities or as
members of professional organizations. This is most
likely to occur when a conceptual or theoretical question
arises which is considered fundamental to the work of
the Bureau in a specialized field, and where professional
acceptance of the Bureau’s work in that field may be
reinforced by the findings of an independent analyst.

1 For example, see Norwegian Nitrogen Company v. United States,
288 U.S. 294; United States v. Kohler, 13 Fed. Rules Serv. 33.333
(E.D., Pa., 1949); Hawes v. Walsh, 277 Fed. 569, the Court of Ap­
peals for the District of Columbia. In all of these cases, the courts sus­
tained the policy of protecting the confidentiality of information given
voluntarily and in confidence to an agency of the Federal Govern­
ment.
2 See Supreme Court of the United States, St. Regis Paper Com­
pany, Petitioner, v. United States, No. 47, October term, 1961.
3 Excerpts from 29 U.S.C. 1, acts of June 27, 1884, ch. 127, 23 Stat.
60; June 13, 1888, ch. 389, 1,25 Stat. 182; Feb. 14, 1903, ch. 552, 4, 32
Stat. 826; Mar. 18, 1904, ch. 716, 33 Stat. 136; Mar. 4, 1913, ch. 141,
3, 37 Stat. 737.

Voluntary reporting and confidentiality
Voluntary reporting and the preserving of the con­
fidential nature of reported data are important
characteristics of bls programs. Over the course of
almost a century, the Bureau has asked hundreds of
thousands of firms and individuals to provide informa­




2

The Consumer Price Index ( c p i ) is a measure of the
average change in the prices paid by urban consumers
for a fixed market basket of goods and services. It is
calculated monthly for two population groups, one con­
sisting only of wage earners and clerical workers and the
other consisting of all urban families.1The wage earner
index ( c p i - w ) is a continuation of the historic index that
was introduced well over a half-century ago for use in
wage negotiations. As new uses were developed for the
c p i in recent years, the need for a broader, and more
representative index became apparent. The all urban in­
dex ( c p i -u ) introduced in 1978 is representative of the
buying habits of about 80 percent of the noninstitutional population of the United States, compared with
40 percent represented in the older index. The
methodology for producing the index is the same for
both populations and is described in detail in part II of
this publication.

The Consumer Price Index 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 pro­
vide 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
1917-19. Periodic collection of prices was started, and,
in 1919, the Bureau of Labor Statistics began publica­
tion of separate indexes for 32 cities. Regular publica­
tion of a national index, the U.S. city average, began in
1921, and indexes were estimated back to 1913.2
Because people’s buying habits had changed substan­
tially, a new study was made covering expenditures in
the years 1934-36 which provided the basis for a com­
prehensively revised index introduced in 1940, with
1 The all-urban-consumer population consists of all urban
households in the Standard Metropolitan Statistical Area (smsa).
Some of these include rural areas as well as cities and suburbs. Non­
farm families living in these rural areas within smsa’s are included, but
the index excludes other rural families and the military and institu­
tional population. The urban wage earner and clerical worker popula­
tion consists of families with clerical workers, sales workers, craft
workers, operatives, service workers, or laborers. These workers must
be the member of the family who earns more than half of the family
income and has worked for at least 37 weeks during the survey year.




retroactive calculations back to 1935.
During World War II, when many commodities were
scarce and goods were rationed, the index weights were
adjusted temporarily to reflect these shortages. In 1951,
the Bureau again made interim adjustments, based on
surveys of consumer expenditures in seven cities be­
tween 1947 and 1949, to reflect the most important ef­
fects of immediate postwar changes in buying patterns.3
The first comprehensive postwar revision of the index
was completed in January 1953, using weights from the
1950 expenditure survey.4 At that time, not only were
the weighting factors, list of items, and sources of price
data updated (appendix A), but many improvements in
pricing and calculation methods were introduced.
Medium-size and small cities were added to the city sam­
ple to make the index representative of prices paid by all
urban wage-earner and clerical-worker families.
Another revision, completed in 1964, introduced new
expenditure weights based on spending patterns in
1960-61 of single persons as well as families, and up­
dated samples of cities, goods and services, and retail
stores and service establishments.
The most recent major revision was completed in
1978. It incorporated new expenditure weights from the
1972-73 Consumer Expenditure Survey, new retail
outlet samples from the 1974 Point of Purchase Survey,
and population data from the 1970 census. It also in­
troduced a second index, the more broadly based c p i for
All Urban Consumers ( c p i - u ), which took into account
the buying patterns of professional and salaried workers,
part-time workers, the self-em ployed, the unemployed,

and retired people, in addition to wage earners and
clerical workers. The two indexes differ chiefly in the
weighting used, as shown in appendix B.
In January 1983, the Bureau changed the way in
which homeownership costs are measured.5 A rental
2 Collection of food prices back to 1890 had been initiated in 1903.
During the course of the 1917-19 expenditure survey, retail prices for
other articles 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.
3 Interim Adjustm ent o f Consumers’ Price Index, Bulletin 1039
(Bureau of Labor Statistics, 1951).
4 Consumer Prices in the United States, 1953-58, Bulletin 1256
(Bureau of Labor Statistics, 1959).
5 “ Changing the Homeownership Component of the Consumer
Price Index to Rental Equivalence,” cpi Detailed Report, January
1983, pp. 7-13.

repricing essentially the same market basket of goods
and services at regular time intervals and comparing ag­
gregate costs with the costs of the same market basket in
a selected base period.
A unifying conceptual framework for dealing with
practical questions that arise in construction of the cpi is
provided by the concept of the cost-of-living ( col) in­
dex.1*As it pertains to the cpi, the col index for the cur­
0
rent month is based on the answer to the following ques­
tion: “ 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 expen­
diture—the lowest expenditure level necessary at this
month’s prices to achieve the base period’s living stan­
dard. The ratio of this hypothetical cost to the actual
cost of the base-period consumption basket in the base
period is the col index. 1
1
The col index is a measure of price change (it com­
pares current-period and base-period prices). However,
the concept is difficult to implement operationally
because it holds the standard of living constant, and the
living standard must be estimated in some way.
The cpi uses a fixed market basket to hold the baseperiod living standard constant. The cpi equals the ratio
of the cost of the base-period basket at this m onth’s
prices to the actual cost of the base-period basket in the
base period. The formula used for calculating the cpi is
the one known in price index literature as the Laspeyres
index. (See part II.) The cpi provides an approximation
to a col index as a measure of consumpton costs. It is
sometimes said that the cpi ’s Laspeyres formula pro­
vides an “ upper bound’’ on the col index.
Note that both the cpi and the col index that was
defined above measure changes in expenditures. Neither
one measures the change in income required to maintain
the base-period living standard. For this reason, neither
the col index nor the cpi is affected by changes in in­
come taxes, but both will include changes in a sales tax
and other indirect taxes.
For certain purposes, one might want to define price
indexes to include, rather than exclude, income taxes.
One could develop either a col index or a Laspeyres in­
dex along these lines. Such indexes would provide an
answer to a different question from the one for which

equivalence method replaced the asset-price approach to
homeownership costs for the cpi- u . In January 1985,
the same change will be made in the cpi-w . The central
purpose of the change was to separate shelter costs from
the investment component of homeownership so that
the index would reflect only the cost of shelter services
provided by owner-occupied homes.
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. A major study was conducted during World
War II by the President’s Committee on the Cost of Liv­
ing.6 The House Committee on Education and Labor
conducted a detailed examination of the index in 1951.7
A decade later, a study was made by the Price Statistics
Review Committee, which was appointed by the Na­
tional Bureau of Economic Research, at the request of
the Office of Statistical Standards of the Bureau of the
Budget, to review all Government price statistics.8 A
continuing flow of articles in professional journals and
books has also contributed to the assessment of the cpi’s
quality and of the ways in which it might be improved.9

Oomiespts
Several key concepts indicate the nature of the Con­
sumer Price Index and the way in which it is calculated.
Prices and living costs
The cpi is based on a sample of prices of food,
clothing, shelter and fuels, transportation, medical ser­
vices, and the other goods and services that people buy
for day-to-day living. Price change is measured by
6 Report o f The President’s Committee on the Cost o f Living
(Washington, Office of Economic Stabilization, 1945).
7 Consumers’ Price Index—Report of a Special Subcommittee of
the Committee on Education and Labor, U.S. Congress, House of
Representatives, 82/1, Subcommittee Report No. 2 (Washington,
U.S. Government Printing Office, 1951).
8 Government Price Statistics—Hearings before the Subcommittee
on Economic Statistics, U.S. Congress, Joint Economic Committee,
87/1, Part 1 (Washington, U.S. Government Printing Office, Jan. 24,
1961).
9 Laurits R. Christensen and Marilyn E. Manser, “ Cost of Living
Indexes and Price Indexes for U.S. Meat and Produce, 1947-71,” in
Household Production and Consumption, Nestor E. Terleckjy, ed.,
Studies in Income and Wealth, 40 (National Bureau of Economic
Research, 1975); Marilyn E. Manser, “ A Note on Cost of Living In­
dexes and Price Indexes for U.S. Food Consumption, 1948-1973,”
bls Working Paper 57, January 1976. Also see Robert A. Poliak,
“ The Theory of the Cost of Living Index,” bls Working Paper 11,
1971; Jack E. Triplett, “ Automobiles and Hedonic Quality Measure­
ment,” Journal o f Political Economy, 11 M ay/June 1969; Jack E.
Triplett, “ Quality Bias in Price Indexes and New Methods of Quality
Measurement,” in Price Indexes and Quality Change, Zvi Griliches,
ed., ch. 6 (Cambridge, Mass., Harvard University Press, 1971); Jack
E. Triplett and Richard J. McDonald, “ Assessing the Quality Error in
Output Measures: The Case of Refrigerators,” Review o f Income and
Wealth, 23, June 1977.



1
0
On the use of cost-of-living index concepts as a conceptual
framework for practical decisionmaking in putting together a price in­
dex, see Robert F. Gillingham, “ A Conceptual Framework for the
Revised Consumer Price Index,” Proceedings, Business and
Economic Statistics Section, American Statistical Association, 1974,
pp. 46-52.
" For more information on the cost-of-living index concept, see
Robert A. Poliak, “ The Theory of the Cost-of-Living Index,” bls
Working Paper 11, June 1971; and Paul A. Samuelson and Subrimanian Swamy, “ Invariant Economic Index Numbers and Canonical
Duality: Survey and Synthesis,” American Economic Review,
September 1974, pp. 566-93.

4

the present cpi is relevant, and would be appropriate for
different uses. For a research measure of a consumption
index inclusive of income taxes and social security con­
tributions, see Gillingham and Greenlees.1
2

Scope and Calculation
Prices for most goods and services used in calculating
the index are collected in 85 urban areas across the
country from about 24,000 establishments—grocery
and department stores, hospitals, filling stations, and
other types of stores and service establishments. Prices
of food, fuels, and a few other items are obtained every
month in all 85 locations. Prices of most other goods
and services are collected every month in the five largest
urban areas and every other month in other areas.
Prices of most goods and services are obtained through
visits by the Bureau’s trained representatives. Mail ques­
tionnaires are used to obtain public utility rates, some
fuel prices, and certain other items. Rent and property
taxes are collected from about 18,000 tenants and
18,000 housing units, respectively.
In calculating the index, price changes for the various
items in each area are averaged together with weights
which represent their importance in the spending of the
appropriate population group. Local data are then com­
bined to obtain U.S. city average. Separate indexes are
also compiled by size of city, by region of the country,
for cross-classifications of regions and population-size
classes, and for 28 local areas.
Movements of the indexes from one month to another
are usually expressed as percent changes rather than
changes in index points because index point changes are
affected by the level of the index in relation to its base
period while percent changes are not. The example
below illustrates the computation of index point and
percent changes:

Sampling
Since it is impossible to obtain prices for all expen­
ditures by all consumers, the cpi is estimated from a set
of samples. It is constructed in accord with statistical
methods that make it representative of the prices of all
goods and services purchased by consumers in urban
areas of the United States:
• A sample of cities selected from all U.S. urban areas,
• A sample of families within each sample area,
A sample of outlets from which these families purchased
• goods and services,
A sample of items—goods and services—purchased by
these families.

It is from these samples that weights are developed
and data are obtained for the monthly calculation of the
index, as explained later in this publication.
Weights and relative importance
The weight of an item in the cpi is derived from the
expenditure on that item as estimated by the Consumer
Expenditure Survey. This survey provides data on the
average amount of white bread, gasoline, and so on,
that was consumed by the index population during the
survey period. In a fixed-weight index, such as the cpi,
the quantity of any item used in calculating the index re­
mains the same from month to month.
A related concept is the relative importance of an
item. The relative importance shows the share of total
expenditure that would occur if quantities consumed ac­
tually remained constant while prices faced by con­
sumers followed their historical path. Although the
quantity weights remain fixed, the relative importance
changes over time, reflecting the effect of price changes.
Items whose prices rise faster than the average become
relatively more important. Thus, the relative impor­
tance of transportation in the index of all urban con­
sumers, which was 18 percent in December 1977, was
nearly 22 percent in December 1982. During the same
period, the relative importance of apparel fell from 5.8
percent to 5.2 percent, even though the same amount of
clothing of the same quality figured in the calculation.
The published data on relative importance are often
used to answer such questions as: What was the effect
on the overall cpi of a particular price change (e.g.,
gasoline prices) for a particular period?1
3

Index point change
c p i .............................................................................................

299.3
Less previous index............................................................... 298.1
Equals index point change...................................................... 1.2

Percent change
Index point difference.............................................................. 1.2
Divided by the previous in d ex ........................................... 298.1
Equals ...............
0.004
Results multiplied by 100 ......................................... 0.004 x 100
Equals percent ch an ge............................................................ 0.4

Percent changes for 3-month and 6-month periods are
expressed as annual rates and are computed according
to the standard formula for compound growth rates.
These data indicate what the percent change would be if
the current rate were maintained for a 12-month period.

Uses

1 Robert F. Gillingham and John Greenlees, “ The Incorporation
2
of Direct Taxes in a Consumer Price Index,” Statistics Canada Con­
ference on the Measurement of Prices, 1982.

^ The cpi is used most widely as a measure of inflation,
and serves as an indicator of the effectiveness of
Government economic policy. It is also a major series
used for economic analysis. Components of the cpi,
such as the cpi series for food at home, are also impor­

1 The procedures used to calculate special indexes and their effect
3
on overall price change are described in Chester V. McKenzie,
“ Relative Importance of cpi Components,” Monthly Labor Review,
November 1961.



5

that is, they cannot be used to determine “ high living
cost” or “ low living cost” cities or regions.

tant as measures of price change for segments of the
consumer’s budget, and are used for analysis (such as
changing patterns of consumer demand).
The cpi is used also as a deflator of other economic
series, that is, to adjust other series for price changes
and to translate these series into inflation-free dollars.
Examples include retail sales and hourly and weekly earn­
ings. cpi components are used as deflators for most
personal consumption expenditures ( pce) in the calcula­
tion of the gross national product ( gnp ).
A third major use of the cpi is to adjust income
payments. More than 8.5 million workers are covered
by collective bargaining contracts which provide for in­
creases in wage rates based on increases in the cpi. In
addition to private sector workers whose wages or pen­
sions are adjusted according to changes in the cpi, the
index now affects the income of more than 50 million
persons, largely as a result of statutory action: Almost
38 m illion social security beneficiaries, a b o u t 3 'A
million retired military and Federal Civil Service
employees and survivors, and about 20 million food
stamp recipients. Changes in the cpi also affect the 23
million children who eat lunch at school. Under the Na­
tional School Lunch Act and the Child Nutrition Act,
national average payments for those lunches and
breakfasts are adjusted annually by the Secretary of
Agriculture on the basis of the change in the cpi series,
,3‘Food away from home.”
In addition, the Economic Recovery Tax Act of 1981
provides for adjustments to the income tax structure
based on the change in the cpi- u in order to prevent
inflation-induced tax rate increases. These adjustments,
designed to offset the phenomenon called “ bracket
creep,” are to be calculated initially in 1984 and
reflected in the 1985 tax schedules.
The cpi may not be directly applicable to questions
about price movements for all specific groups. For ex­
ample, the cpi- u is designed to represent the average
movement of prices for the urban population and is thus
not directly applicable to nonurban residents. The cpi
does not provide information on the rate of inflation ex­
perienced by any particular demographic subgroup of
the population, such as the aged.1
4
The area indexes measure the average changes in price
for each area since the base period. They do not
measure differences in the level of prices among cities;

Analyses and Presentation
The monthly cpi is first published in a news release
between the 20th and 25th of the month following the
month in which the data are collected. (The index for
January is published in late February.) The release in­
cludes a narrative summary and analysis of major price
changes, short tables showing seasonally adjusted and
unadjusted percentage changes in major expenditure
categories, and several detailed tables. Summary tables
are also published in the M onthly Labor Review the
following month; shortly thereafter, a great deal of ad­
ditional information appears in the monthly c p i Detailed
Report.
Seasonally adjusted data are presented in addition to
unadjusted data. Seasonal adjustment removes the
estimated effect of changes that normally occur at the
same time and in about the same magnitude every year,
such as price movments resulting from changing
climatic conditions, production cycles, model changeovers, holidays, and sales. Seasonal factors used in com­
puting the seasonally adjusted indexes are derived by the
X -l 1 Variant of the Census Method II Seasonal Adjust­
ment Program and are reevaluated annually. (See ap­
pendix E for an explanation of bls seasonal adjustment
methods.)

limitations of the index
The cpi is not an exact measure of price change. It is
subject to sampling errors which may cause it to deviate
somewhat from the results which would be obtained if
actual records of all purchases by consumers could be
used to compile the index. These estimating or sampling
errors are statistical limitations of the index rather than
mistakes in the index calculation.
Another kind of error occurs because people who give
information do not always report accurately. The
Bureau makes every effort to keep these errors to a
minimum, obtaining prices wherever possible by per­
sonal observation and correcting errors whenever they
are discovered. Precautions are taken to guard against
errors in pricing, which would affect the index most
seriously. The field representatives who collect the price
data and the commodity specialists and clerks who pro­
cess them are well trained to watch for unusual devia­
tions in prices which might be due to errors in reporting.

1 For recent research on the inflation experience of subgroups of
4
the population, see Robert P. Hagemann, “ The Variability of Infla­
tion Rates Across Household Types,” Journal o f Money, Credit and
Banking, 4, November 1982.




6

0 P [r[L (g © i © the Index
®iS O iiD [n if

The construction of the Consumer Price Index is bas­
ed on a series of samples and on estimation procedures
described below.

strata are mutually exclusive and exhaustive of all ex­
penditures, and they are defined identically for both
populations.
If Czt is the total expenditure on base-period (r) quan­
tities of items in the zth stratum at prices observed at
time t, then:

Definition ©f the Index
The cpi is defined as a fixed-quantity price index, and
is a ratio of the costs of purchasing a set of items of con­
stant quality and constant quantity in two different time
periods. We denote the index by It>» where t is the com­
0
parison period for which a new index number is to be
calculated and 0, the reference period:
SPitQirIt.0 =
x 100
2 PioQir.
X
where:
Pit
Pi0
Qir

C zt =

where i ranges over all items in the zth stratum. Then the
estimator can be written as:
2 C2
i

I.«_

z

v 1 na

z
The cpi is computed by a chaining process in which an
estimate of expenditure for the previous month in each
stratum is multiplied by^ an estimate of the relative
change in price from the previous month to the current
month to provide an estimate of the current month’s ex­
penditure for that stratum:

is the price for the ith item in comparison period t
is the price for the ith item in reference period 0
is the quantity of the ith item consumed in the ex­
penditure base period r

c =c

When the expenditure base and the reference periods
coincide, this becomes the Laspeyres price index for­
mula. For the 1978 revision of the cpi, however, they
did not coincide, and the formula was modified. The ex­
penditure data? P ir Qir from the 1972-73 Consumer Ex­
penditure Survey (described later in this chapter and in
volume 1, chapter 6) were updated for relative price
changes (Pi0 P ir) to December 1977, the reference period
/
when they were introduced into the cpi. This was done
in two stages. First, from the expenditure survey col­
lection period r to a month during the period from
January through June 1977 (a function of the item and
area priced), they were moved by price relatives from
the ongoing cpi. Second, from that point to their in­
troduction, they were moved by collected, but un­
published, price relatives for the revised cpi- u and cpiw. Continuity with the pre-1978 version of the cpi was
maintained in the published version by modifying the
above formula to:
2 Pit Qir
X I d EC77, 1967
2 PioQir

r7

where RztjM is an estimate of the one-period price
change in the zth item stratum.
The item stratum expenditure values—or cost
weights—are then aggregated and compared to the totalperiod expenditure to form the overall cpi . (Indexes for
individual strata are simply computed as:
IztO

x 100.) The chaining of one-period (gene­

rally) price relatives based on identical item specifica­
tions in adjacent periods allows the requisite flexibility
to update the sample of outlets and specifications to
reflect an updated distribution of purchases.
The form of the estimator for a one-period price
change, RztjM, depends on how the samples of outlets
and items are drawn. When the samples are selected
with probability proportional to quantity, the form of
the estimator of Rzt> is:
t„,
.2
Rzt,t-l

WjPit

1€ Z
2

W j P it_,

ie z

This is the ratio of the summation of weighted prices,
where the weights (W-) reflect the probability of selec­
tion of the item being priced and a noninterview adjust­
m ent.1 This estimator is used in the rent, property tax,
5
and house price strata.

where ID C , i967 is the 1967-based value of the cpi .
E 77
(Note that the base period for expenditure weights,
1972-73, should not be confused with the base period
for the index, 1967 = 100.)
For sampling and computational purposes, the set of
all retail consumer expenditures by the target popula­
tion for a given index area has been subdivided into 265
classes of similar items called item strata. The item



2 P jt Q ;r
ie z

1 Noninterview adjustment is a statistical procedure designed to ad­
5
just for sample nonresponse. (See the section on the Consumer Expen­
diture Survey-Weighting procedure.)

7

which are normally difficult to recall. These items in­
clude food and beverages, natural gas and electricity,
gasoline, housekeeping supplies, nonprescription drugs,
medical supplies, and personal care products and ser­
vices. Consumer units were asked to list all expenses
during the survey period. Data on income and family
characteristics also were collected. The sample of hous­
ing units was balanced across areas and time of year.
The Interview Survey was a panel survey in which
each consumer unit in the sample was interviewed every
3 months over a 15-month period. This survey was
designed to collect information on major items of ex­
pense as well as on income and family characteristics.
These items included housing, household equipment,
housefurnishings, vehicles, subscriptions, insurance,
educational expenses, clothing, repair and maintenance
of property, utilities, fuels, vehicle operating expenses,
and expenses for out-of-town trips. The final interview
provided the regularly recorded expenses plus informa­
tion on homeownership costs, work experience, changes
in assets and liabilities, estimates of consumer unit in­
come, and other selected financial information.
For the samples of consumer units, in both the Diary
and Interview surveys, the Nation was stratified into 216
geographic strata using stratification variables defined
for the Current Population Survey of the Bureau of the
Census. Thirty of these areas were designated as selfrepresenting. Half of the housing units in each selfrepresenting area were covered in the first survey year
and half in the second survey year. The other 186 areas,
all equal in size, were divided into two 93-area groups.
One sample area from each of the 93 groups was in the
sample in each of the 2 survey years. Each sampling area
was randomly selected proportional to population from
each of the 186 strata.

When the samples of outlets and prices are selected
with probability proportional to expenditure, the form
of the estimator of Rzt)tl is:
Rzt.t-l

2 W —
*
i e z 1 p iO
2 Wi
iez

This is the ratio of the summation of weighted price
ratios, where the weights reflect the probability of selec­
tion of the item being priced and noninterview ad­
justments. It is used for all expenditure categories other
than rent, property tax, and house prices.
Thus, construction of the cpi is a twofold estimation
process. First, the reference-period cost weights, Cz0,
must be estimated. These estimates are derived from the
Consumer Expenditure Survey as explained in the next
section. Second, the one-period price changes, Rztt.,,
must be estimated for each pricing period. The
methodology for estimating price changes is explained
in a later section.

The Consumer Expenditure Surwey
The base-period cost weights currently employed in
the cpi are derived from the 1972-73 Consumer Expen­
diture Survey (ce) conducted by the Bureau of the Cen­
sus for BLS.
Sampl® design
The ce survey consisted of two separate com­
ponents—a Diary Survey and an Interview Panel
Survey—each with its own questionnaire and sample.
The sampling unit for these surveys was a housing unit.
The reporting unit was a consumer unit which was
defined as (1) a group of two or more persons, usually
living together, who pooled their income and drew from
a common fund for their major items of expense; or (2)
a person living alone or sharing a household with
others, or living as a roomer in a private home, lodging
house, or h o tel, but who was financially in ­
dependent—that is, income and expenditures were not
pooled with other residents. Children living with their
parents were considered members of the consumer unit
if they had never been married. The eligible population
included the civilian noninstitutional population of the
United States and doctors’ and nurses’ quarters of
general hospitals. Armed Forces personnel living out­
side military installations were included while Armed
Forces personnel living on a post were excluded. Also
excluded were persons living in college dormitories,
fraternity or sorority houses, prisons, monasteries,
aboard ships, or in other quarters containing five or
more unrelated persons.
The Diary Survey was completed by respondents for
two consecutive 1-week periods in order to obtain ex­
penditure data on small, frequently purchased items



Sample sizes amid response rates
For the Interview Survey, the universe for sample
selection was the 1970 census 20-percent-sample data
file. A sample of housing units was selected by com­
puter from this census file for each year of the Interview
Survey. A sample of 12,613 housing units was
designated for the 1972 Interview Survey component,
and 13,014 housing units for the 1973 component. Of
the 12,613 housing units designated for the first year of
the survey, 1,401 units were determined to be type B or
C nonresponse (vacant, nonexistent, or ineligible),
yielding a nonresponse rate of 11.1 percent. Of the re­
maining 11,212 eligible housing units, interviews were
completed in 9,914 units. The remaining 1,298 housing
units were classed as type A nonresponses (refusals,
temporary absences, or no contacts). The 1972 type A
nonresponse rate for designated sample units was 10.3
percent. Of the 13,014 housing units designated for the
1973 Interview Survey, 1,679 units were determined to
be type B or C nonresponses, yielding a nonresponse
8

as type A nonresponse, yielding a type A nonresponse
rate for designated units of 16.9 percent.
For the 1973 component of the survey, 2,256 of the
15,207 housing units designated were determined to be
type B or C nonresponses for all first weeks, and 2,208
for all second weeks. The type B or C nonresponse rate
was 14.8 percent the first week and 14.5 percent the se­
cond week. During all first weeks, interviews were com­
pleted in 1,673 of the 12,951 eligible units. The remain­
ing 1,278 units were type A nonresponses—a
nonresponse rate of 8.4 percent for designated sample
units. Eligible housing units during all second weeks
totaled 12,999. Interviews were completed in 11,682
units; the remaining 1,317 units were type A
nonresponses, yielding a nonresponse rate for
designated sample units of 8.7 percent.

rate of 12.9 percent. Interviews were completed in
10,158 units. The remaining 1,177 units were classed as
type A nonresponses. The type A nonresponse rate of
designated sample units for the 1973 Interview Survey
was 9.0 percent.
At the time of selection, housing units for the Inter­
view Survey within a primary sampling unit ( p su ) were
distributed within the quarter by month to allow for
data collection throughout each quarter. Each sample
unit was visited at approximately the same time each
quarter, and each consumer unit within the household
was interviewed. Data from previous quarters were
available for the interviewer to use in bounding'6 expen­
diture reporting. The type of expenditures reported dur­
ing each interview varied since the recall periods varied
from 3 months to 1 year. Housing, major equipment,
automobles, subscriptions, and insurance were annual
recall items. A semiannual recall period was used for
minor equipment, housefurnishings, renting and leasing
of vehicles, and education. The following items were
covered each quarter: Repair, alterations, and
maintenance of owned property; utilities, fuel, and
household help; clothing and household textiles; equip­
ment repairs; vehicle operating expenses; and out-oftown trips. Interviews were conducted with any person
available in the consumer unit; no attempt was made to
interview all persons in the unit.
For the Diary Survey, the universe for sample selec­
tion was the 1970 census 20-percent-sample data file. A
sample of housing units was selected by computer from
this file for each year of the survey. Approximately
14,590 housing units were designated for the 1972
survey component, and about 15,210 for the 1973 com­
ponent. These numbers included an augmented sample
of households which were to be visited during the
4-week period preceding the end-of-the-year holidays.
Each housing unit was visited twice, at the end of each
week of the 2-week survey period.
Of the 14,586 housing units designated for the 1972
Diary Survey, 1,925 were determined to be type B or C
nonresponse during the first week of all periods; an ad­
ditional 1,869 for the second week of all periods. The
type B or C nonresponse rate was 13.2 percent for the
first week and 12.8 percent for the second. Of the
12,661 housing units eligible for interview during all
first weeks, interviews were completed in 10,144 units.
The remaining 2,517 units were type A nonresponses.
The first-week type A nonresponse rate for designated
sample units was 17.3 percent. During all second weeks
of the 1972 Diary Survey, 12,717 housing units were
determined to be eligible. Interviews were completed in
10,248 units, and the remaining 2,469 units were classed1
6
1 Bounding is an interviewing technique which separates expen­
6
ditures reported in the previous interview from the current interview.




9

Weighting the questionnaires
In order to use the sample unit responses to estimate
totals, a weight was determined for each usable con­
sumer unit (cu) diary or interview questionnaire receiv­
ed. Each c u ’s weight was the product of several factors.
(1) The first factor was the inverse of the probability of
selection of the housing unit, and was equal to two times
the initial sampling rate, reflecting the splitting of the
sample into two 1-year components. (2) A duplication
factor was assigned to each person in the household
when the diary sample was augmented during the preChristmas holiday week. (3) Since some diaries or inter­
views could not be conducted in occupied sample
households, a complex noninterview adjustment was
made to correct for these deficiencies. (4) For persons in
non-self-representing p s u ’s , there was a ratio adjust­
ment factor for the race-residence cell to which the
household was assigned. This factor adjusts for some of
the differences between the sample areas and the
geographic stratum they represent. There were two race
cells, white and nonwhite, for each of five residence
categories within each of the four census regions—
Northeast, North Central, South, and West. The
residence categories were: Central city of a Standard
Metropolitan Statistical Area ( smsa ) over 400,000,
noncentral city of smsa ’s over 400,000, central city of
other smsa ’s , noncentral city of other smsa ’s , and nonsmsa ’s . (5) A second ratio estimate was computed for
age-sex-race cells to known control totals of population
for each quarter of the year and applied to the civilian
noninstitutional population. No age-sex-race ratio ad­
justment was made for persons under 14 years of age.
The final weight of an individual cu in either survey
was determined by the type of family composition of the
cu in order to insure consistency between the number of
married men and women and determine a single weight
for the cu. Every cu was 1 of the 3 types listed below
and was assigned the weight of a person in the cu.

1.
2.
3.

Husband-wife c u ’s were assigned the wife’s
weight.
c u ’s with a female householder were assigned
the weight of the householder.
c u ’s with a male householder (excluding
husband-wife c u ’s) were assigned an adjusted
weight of the male householder. This weight
was adjusted by the ratio of the number of
male householders with spouse present by agerace class using the husband’s weight.

the country. Each of the 12 region, city-size publication
areas contained four, six, or eight strata. This resulted in
39 areas for which data were published. In addition,
special supplementation was made to support publica­
tion for Denver.

CPI Item Structure and Item Sampling
The cpi item structure has four levels of classifica­
tion. The seven major product groups are disaggregated
into 68 expenditure classes (ec ’s) which, in turn, are
broken into 265 item strata. Within each item stratum,
one or more substrata, called entry level items ( eli’s),
were defined, for a total of 382 eli’s . These are the
ultimate sampling units for items as selected by the bls
Washington office. (See exhibit 1.) They are used by
data collectors as their initial level of item defintion
within an outlet.
Four regional market-basket universes were tabulated
from the Diary and Interview surveys to reflect regional
differences. Within each of the four regions, eight in­
dependent samples of eli’s were selected for each item
stratum. Thus, eight samples of eli’s were selected for
each region and for each population—32 sample selec­
tions nationally for each population. Each selfrepresenting published area was assigned 2 of the 8 item
samples and each non-self-representing area was
assigned 1 of the 8 item samples. These item samples
were designed to accommodate variance estimation for
the cpi.
Because of the requirement to publish indexes for two
populations and the need to minimize cost, the follow­
ing technique was used to select the eli sample from
each item stratum. The eli sample was selected first for
the urban wage earner and clerical population (cpi- w ).
Each probability selection of an eli sample within an
item stratum was made proportional to the relative ex­
penditures for each eli within the item stratum for the
cpi- w population in a region. The eli sample for the allurban population (cpi- u ) was then selected using a con­
ditional probability technique to maximize the overlap
probabilities of selection.

CPI Sample Areas and Pybiicatlon Areas
Pricing for the cpi is conducted in 87 primary sampl­
ing units in 85 geographic areas. (The New York area
has three psu ’s.) The area design and sampling are sum­
marized as follows: The entire country was divided into
1,166 psu ’s . A psu is a county or group of contiguous
counties with similar demographic and economic
characteristics. The basis of the psu definitions was the
geographic areas defined by the Bureau of the Census
for the Current Population Survey in 1960 with popula­
tion estimates from the 1970 census. Each smsa as
defined by the Office of Management and Budget in 1973
is a psu . When metropolitan statistical areas of over 1
million population are defined, the largest area of which
they are component parts is designated as a Standard
C onsolidated A rea ( sca ). The rem aining n o n - smsa
psu ’s containing any urban population were grouped in­
to contiguous areas to form bls psu ’s . Rural areas of the
non- smsa counties were excluded.
Eighty-five geographic strata were defined by com­
bining similar psu ’s according to the following
characteristics in order of importance:
a.

Region, population size, smsa vs. non-SMSA;

b.

Percent population increase from 1960 to
1970;
Major industry;
Percent nonwhite;
Percent urban.

c.
d.
e.

This area design resulted in 27 strata with one pricing
area per stratum (self-representing psu ’s) and 58 non­
self-representing strata. (The three New York psu ’s are
self-representing.) One sample psu was selected from
each non-self-representing stratum. A controlled selec­
tion program1 was used to insure the sample areas were
7
distributed geographically across the United States.
Twelve publication areas consisting of three city sizes
(non-self-representing smsa urban areas) crossed by the
four census regions were defined along with the 27 local
areas to provide estimated indexes for all urban areas of

1 George Werking,
7
Statistics, 1981).



osd

CPI Oytflst and Pries Syrweys

Statistical Notes, No. 6 (Bureau of Labor

10

bls field representatives collect prices monthly for the
food, rent, property tax, and some other commodity
and service components of the cpi in all 87 psu ’s. Each
of these components has a separate survey with its own
sample design. Price data for a few items such as m ort­
gage rates and house prices are obtained from sources
outside bls. The sample design for each component is
described in this section.
The Point of Purchase Survey (pops) is the source of
the outlet-sampling frames for about 60 percent of the
cpi items by expenditure weight. The items not covered

Exhibit 1.

Outline of the CPI Classification Structure, 1977

I. c p i expenditure categories are divided into seven II. Each major group is subdivided into expenditure
classes ( e c ’s), item strata, and entry level items ( e l i ’s),
major groups:
as illustrated below in the reproduction of part of the
Food and beverages
food and beverages group. Almost every expenditure
Housing
class has more than one item stratum; an exception is
Apparel and upkeep
eggs. Most strata have only a single entry level item, but
Transportation
some have two or more. In total, there are 382 entry
Medical care
level items that are priced for the c p i .
Entertainment
Other goods and services
Group I. Food and Beverages
Stratum
Expenditure class Number
EC 1. Cereal and
cereal pro­
ducts

EC 2. Bakery pro­
ducts

1

Flour

EC 4. Pork

2
3

Cereal
Rice
Pasta and cornmeal

1
2
3
4
5
6

White bread
Other bread
Biscuits, rolls, muffins
Fresh cakes and cupcakes
Cookies
Crackers
Bread crumbs, stuffing, cracker crumbs,
meal
Fresh sweetrolls, coffeecake, donuts
Frozen and refrigerated bakery products
Fresh pies, tarts, turnovers

1
2
3
4
5
6
7
8

Entry level item

Bacon
Pork chops
Ham other than canned
3
Pork other than bacon, chops, ham
4
sausage
5
Sausage
Canned ham
6
1

2

7
8

EC 3. Beef and
veal

Stratum
Expenditure class Number

Entry level item

1
2
3
4

Frankfurters
Bologna, liverwurst, salami
Other lunchmeats
Lamb and organ meats

EC 6. Poultry

1
2
3

Whole fresh chicken
Fresh or frozen chicken parts
Poultry other than fresh chicken and
frozen chicken parts

EC 7. Fish and
seafood

1
2

Canned fish and seafood
Fresh and frozen shellfish
Fresh and frozen fish

EC 8. Eggs
O

Ground beef other than canned
Chuck roast
Round roast
Other roasts
Round steak
Sirloin steak
Other steak
Other beef and veal

EC 5. Other meats

1

Eggs

G
©

by the p o ps are grouped together under the heading nompo ps and include rent, property tax, mortgage interest,
house prices, utilities, transportation, insurance, and
several miscellaneous categories. Frame development
and sampling methodologies for various non-pops items
are also described in this section.

demographic data for classification of the households
reporting an expenditure for an outlet. The survey was
conducted in the 87 p s u ’s defined for the c p i . The com­
modities and services for which sampling frames were
developed in each p su included food, apparel, drugs,
personal care items, household furnishings and
housekeeping supplies, beverages, most medical ser­
vices, sports equipment, gasoline and automobiles, and
automotive parts and services. Expenditures, name, and
location of the place of purchase were collected for ap­
proximately 100 relatively broad categories of expen­
ditures with reference periods of 1 week to 2 years
depending on the expected frequency of reporting. To

Poinf-of-Purchase Survey
In the spring and summer of 1974, a household survey,
the Point of Purchase Survey, was conducted by the
Bureau of the Census for bls to provide the sampling
frame of outlets for food and most commodities and
services to be priced in the c p i and to provide




11

control the expected number of responses received from
a household and minimize respondent burden, two
groups of categories were defined; one set was given to
one-fourth of the sample households and the second set
to three-fourths of the sample households. The com­
bination of sample size and reference period for a given
pops category was designed to generate 6 to 12 not
necessarily unique outlets reported for a given psu / pops
category.
For pops, the national sample size was 23,000
designated housing units. Since separate frames of
outlets were required for individual cpi pricing areas
(psu ’s), the sample is not self-weighting across psu ’s,
but, within a psu , the households are selected with a
uniform probability.
In the 1974 version of pops, a highly clustered sample
of households was selected for 60 of the 87 psu ’s on the
assumption that, if families tended to buy in the areas
' where they live, the outlets given as responses to the
survey would also be clustered. In order to increase the
expected chance of clustering outlets, the household
clusters were formed (where possible) around known
shopping complexes. These large clusters are called
secondary sampling units (ssu’s). Within a cluster of
tracts, a sample of Census enumeration districts ( ed ’s)
was selected, and, within the selected ed ’s, the sampled
households were dispersed evenly. Five housing units
were selected in each ed , and since the desired sample
size per cluster was 40 housing units, about eight ed ’s
were sampled from each cluster.
In the initial formation of the ssu’s, all known shop­
ping complexes were spotted on tract maps. These shop­
ping complexes were the central business districts, major
retail centers (as defined by the 1967 Census o f
Business), and other shopping centers as defined by the
1970 Directory o f Shopping Centers. Census tracts were
then grouped into ssu’s around each of these shopping
complexes. The guidelines for forming these ssu’s were
as follows:
a.
b.
c.
d.

omitted, and sampling o f ed ’s was the first stage o f
selection.

The sample of housing units was self-weighting. Each
sample housing unit was assigned an internal weight
which was the inverse of the probability of selection
times a noninterview factor.
Since 1977, the bes has sponsored a Continuing Point
of Purchase Survey (cpops), conducted by the Bureau of
the Census, to acquire current data on outlets. The
cpops has been expanded from 100 categories of expen­
ditures to 134 categories, of which 102 are used for each
of two panels of equal size. This survey is conducted
each year in one-fifth of the 87 psu ’s on a rotating basis.
In 1979 and later years, unclustered samples of
households were selected for cpops. From the results of
these household surveys, new samples of outlets are
selected, bls then initiates these new outlets, selects item
specifications for pricing from each, and replaces the
former set of items in the cpi from each surveyed city
with the new outlets and specifications.
CPI outlet sampling procedures
When a sampled eli was selected, a specific pops
category was identified for outlet selection. In selfrepresenting areas, sample households were divided into
two independent groups by the first-stage order of selec­
tion which defined two frames of outlets for outlet
selection to support variance estimation. For a given
psu/population/cpops category/half-sample, the total
expenditures of a given outlet were edited and insured a
minimum chance for selection if no expenditures were
reported for an outlet or were restricted to 20 percent of
the total cpi- u population expenditure if a very large
number of expenditures were reported.
A systematic selection of outlets reported for a given
pops category for the cpi- w population was made where
the measure of size for each outlet was proportional to
the average daily expenditure reported for the outlet by
all consumer units in the cpi- w population. Before
January 1982, the outlets for the cpi- u population were
then adjusted using a conditional probability techni­
que1 to recompute the measure of size for every outlet
8
in the universe. The sample outlets for the cpi- u
population were then selected by a repeat of the
systematic selection process using the new measures of
size. After January 1982, the independent collection of
prices for the cpi- w population was discontinued and
the use of the conditional probability technique was no
longer applicable. The sample outlets are now selected
systematically with probability proportional to average
daily expenditure of the cpi- u population.
Response rates were calculated as follows. Samples
of 24,278, 7,374, and 7,197 outlets were designated for
the original pops survey, the 1977 cpops survey, and the
1978 cpops survey, respectively. Of the designated

All tracts containing part of the shopping complex
should be in the same ssu.
Each ssu should have only one shopping complex.
The ssu should have a minimum of 2,500 housing
units.
The ssu should be no more than 2 miles square.

Tracts not covered by this operation were put into con­
tiguous groups according to guidelines c. and d. above,
as well as the following:
e.

f.

When possible, all tracts in an incorporated or unin­
corporated place outside the central city of the smsa
and with more than 2,500 population were an ssu.
ssu’s did not cross the central city or urbanized boun­
daries.

The Bureau of the Census sampled new construction
units in each psu , without regard to the ssu’s. For the re­
maining 25 smallest psu ’s , the definition of ssu’s was



1
8
Harry Marks,
Statistics, 1981).

12

osd

Statistical Notes, No. 2 (Bureau of Labor

outlets, 3,055, 859, and 742, respectively, were deter­
mined to be out of scope of the survey, that is, were out
of business, could not be located, were outside the pric­
ing area, had no items available for eli pricing, or had
moved. The out-of-scope nonresponse rate was 12.6
percent, 11.6 percent, and 10.3 percent, respectively.
There were 1,649, 334, and 305 outlets with type A
nonresponses resulting from no contact, refusals, or
temporary absences. The respective type A nonresponse
rate for designated sample units was 6.8 percent, 4.5
percent, and 4.2 percent. Of the respective 21,223,
6,515, and 6,455 outlets eligible for interview, the
response rate was 93, 95, and 95 percent. On average,
the c p i experiences an additional annual attrition rate
for sample outlets of 2.9 percent.
Table 1 presents the c p i Commodities and Services
response rate experienced at initiation of each of these
surveys and the proportion of all outlets or quotes inter­
viewed to the total outlets or quotes interviewed that
should have responded.
Table 2 presents the designated number of sample
outlets by half-sample for each po ps category for either
population (u) or (w) within the commodities and ser­
vices item groups. (A single selection of e l i ’s and outlets
for all item stata assigned to a psu is called a half­
sample.) The designated sample size is the number of
outlet selections and does not reflect the nonresponse in
table 1, nor determine the number of unique outlets.
Where multiple selections of the same outlet occur, a
comparable increase in quotes priced for the outlet oc­
curs. The New York City sample area consists of six
half-samples. The Chicago, Los Angeles, and other selfrepresenting sample areas each have two half-samples.
Each region-size class consists of four, six, or eight sam­
ple pricing areas with each psu having one half-sample.
The region-size classes are defined in appendix B. Thus,
the designated outlet sample size for a given c p i publica­

Table 1.

Response rates at initiation for

cpi

tion area is a multiple (as indicated above) of the value
provided in table 2.
Table 2 gives designated sample size for a given po ps
category and half-sample. The designated sample sizes
allow for nonresponses. If these allowances were not in­
cluded, the desired sample size would be as follows: For
food, the desired outlet sample size of outlets per halfsample/population is five, six, four, and three, respec­
tively, for each city column of table 2. For all other
items,except gasoline, the desired outlet sample size is two,
three, two, and two, respectively.
Mon-POPS sampling procedures
The non-pops categories were excluded from the po ps
either because existing sampling frames were adequate,
or it was felt the po ps would not yield an adequate
sampling frame. For each non-pops component, the
sampling frame was either acquired from another agen­
cy or constructed by bls . Since 1977, some of these
categories have been added to the c p o p s .
Each non-pops commodities and services item has its
own sample design. For each item, the frame consisted
of all outlets providing the commodity or service in each
sample area. A measure of size was associated with each
outlet on the sampling frame. Ideally, this measure of
size was the amount of revenue generated by the outlet
by providing the item to the c p i - u population in the
sample area.
Whenever revenue was not available, an alternate
measure of size, such as employment, number of
customers, or quantity of sales, was substituted. Since
no measures of size could be determined strictly for the
c p i - w population, a single sample of outlets and quotes
was selected for estimating the relative for both popula­
tions. All samples were selected using the systematic
sampling techniques with probability proportional to
the measure of size defined in appendix C.

commodities and services in outlet surveys

Original 1975

19771

19781

Type of interview

Number
of
outlets

Number
of
quotes

Number
of
outlets

Number
of
quotes

Number
of
outlets

Number
of
quotes

Total designated outlets or quotes
At least one quote obtained
Unable to price during pricing period
Out of season
Refusal....................................................................................
Total of outlets or quotes interviewed
Proportion of interviewed outlets or quotes responding
No items available for E I pricing.............................................
L
Out of business or out of scope
Unable to locate......................................................................
Outside pricing area
Outlet has moved
Total of outlets or quotes not interviewed...............................

24,278
19,574
485
71
1,093
21,223
.93
1,583
1,334
91
9
38
3,055

159,305
138,449
2,807
179
5,149
146,584
.95
5,557
6,486
464
60
154
12,721

7,347
6,181
87
12
235
6,515
.95
467
293
46
2
51
859

34,997
31,615
388
41
740
32,784
.97
1,170
809
118
8
108
2,213

7,197
6,150
97
4
204
6,455
.95
409
282
42
3
6
742

32,362
29,241
497
14
632
30,384
.96
977
813
152
11
25
1,978

1
One-fifth of p s u ’s for sample rotation initiation. These samples replace the sample of outlets and quotes the year following the date of the survey.




13

percentage of dollar sales or approximations to those
sales. The procedures developed to obtain the propor­
tion of sales were:

Table 2. Designated outlet sample sizes for each half-sample
within a market basket for cpi commodities and services

Chicago
and
Los
Angeles

Re
Other gionself-repre­ size
senting class

Group

New York

Food:
At home............... ..........
Away from home

6
3

7
4

5
3

4
3

Fuels:
Fuel oil2 ____________
Natural gas
Electricity

12
2
4

15
3
5

12
2
4

12
2
3

Household furnishings and
operations

3

4

3

3

Apparel:
Men’s and boys’
Women’s and girls’

3
3

4
4

3
3

3
3

Transportation:
New cars
Motor fuels3 ...................
Public transportation
Vehicle repair.................

8
20
3
3

11
30
4
4

8
30
3
3

8
8
3
3

Medical care:
Professional services2 ....
Commodities

8
4

12
6

8
4

8
4

Entertainment:2

5

7

5

5

Other goods and services_

4

6

4

4

a.
b.

c.
d.

To define the categories, direct responses from the
respondent as to what he sells or an inventory technique
was used.
These procedures make possible an objective prob­
ability sampling of items throughout the c p i . They also
allow broad definitions of e l i ’s so that the same tight
specification need not be priced everywhere. The wide
variety of specific items greatly reduces the within-item
component of variance, reduces the correlation of price
movement between areas, and allows a substantial
reduction in the number of quotes required. A second
important benefit from the broader e l i ’s is a significant­
ly higher probability of finding a priceable item within
the definition of the eli within the sample outlet.
Each eli is contained in only one p o ps category. Since
the c p i outlet sampling is done at the p o ps category
level, the item sample must be matched to the cor­
responding p o ps category outlet samples. All e l i ’s cor­
responding to a given p o ps category are assigned for
pricing in each sample outlet selected for that category.
The matching process is controlled by a concordance
between the e l i ’s and the p o ps categories.

1 The New York sample area consists of three self-representing PSU’s.
(See appendix D.)
2Sample size is larger to offset large nonresponse experience.
3 Sample size is designed to support average prices for 3 gasoline
types.

Appendix C lists the non-pops commodities and ser­
vices items, the source of the sampling frame, the defini­
tion of the sampling unit, the measure of size employed,
the final pricing unit, and the number of designated
outlets and quotes.

Rent Survey
The current c p i rent index is based on a sample of ap­
proximately 23,000 rental units, allocated among the 87
p s u ’s . The units were selected from two universes, a
stratified multistage, systematic, self-weighting area
sample of housing units built before 1970 and a con­
tinuously updated sample of newly constructed units.
The Bureau of the Census provides the sample of new
construction units from building permits. Approximate­
ly 2,000 units have been obtained from this source as of
1982. The following steps were used to select a sample
of pre-1970 rental units:

Selection] procedures within outlets
For each e l i , whether in a po ps or non-pops category,
a bls field representative selects a specific store item us­
ing multistage probability selection techniques with
measures of size proportional to percentages of dollar
sales, usually provided by the respondent for the outlet.
The field representative uses a checklist that includes the
descriptive characteristics necessary to identify the item
and to determine or explain price differences for all
items defined within the e l i . In addition, the field
representative has the definition of the e l i , suggested
stages of groupings of items for quick selection, and
worksheets on which to define the categories of items,
post the probabilities, and identify the next category
within which to select the specific store item by use of
the random number table on the worksheet.
In developing this procedure, it was necessary to pro­
vide the field representative with several alternative
methods for defining the categories and obtaining the




Obtaining the proportions directly from a respondent;
Ranking the categories by importance o f sales and then
obtaining the proportions directly or using preassigned
proportions;
Using shelf space to estimate the proportions where
applicable;
Using equal probability if all else fails.

a.

b.

c.

14

A minimum number of units was allocated to each
local area self-representing psu , if needed, and a pro­
portional sample was allocated to the remainder o f the
psu ’s .
For each psu , Census Enumeration Districts were
stratified according to percent o f rental units, income,
and types o f units.
Each ed was assigned a measure o f size according to
the number of year-round renter housing units within
the ed as defined by the 1970 census ed summary files.

d.
e.

f.

select the segments for rental units, the property tax
sample could not have a uniform probability of selec­
tion within a p s u , unless significant clustering of owner
housing units were to be accepted. This was undesirable
because of the relatively high intraclass correlation of
property tax rates of change. Thus, to control the average
segment size of sample owner units, the probabilities of
selection were different for 18 different strata within a psu .
The difference in probabilities for each owner unit was
determined such that the ratio of the probability of any
two given units within a psu was not greater than four.
This was used to control the variance of the measured price
level. The response rate for the property tax survey was
100 percent since pricing is at the local tax jurisdictions.
For the mortgage interest rate sample each month, bls
obtains information from the Federal Home Loan Bank
Board ( fhlbb ) on a national sample of about 1,600
mutual service and commercial banks, savings and loan
associations, and mortgage companies. Three lendersize strata were defined with a differential sampling rate
used in each strata across the Nation. All transactions
which occur the first week of the previous month within
the sample units are used to estimate the average con­
ventional mortgage interest rate. For the Federal Hous­
ing Administration/Veterans Administration ( fh a / va )
component of mortgage interest, bls uses the f h a / va
mortgage interest ceiling as of the 15th of the reference
month as the interest rate.
For the house price sample, bls obtains a tape each
month of the universe of all sales of FHA-insured loans
which are processed by fh a during the previous month.
The sales may have been concluded several months
before bls receives the information from the f h a . There
is no sampling process involved in the collection of this
data. All sales of housing units not insured by fh a are
excluded from measurement of price movement for the
c p i . On a national level, the fh a currently accounts for
about 7 percent of all housing sales and is subject to
legal constraints on the value ceiling of mortgages they
offer. When the fh a mortgage ceiling is raised, this
change is not shown as a price change in the c p i . The
price relative obtained from these data is used directly
or indirectly (as a multiplier) for the house price, mort­
gage interest cost, and property insurance components
of homeownership.

ed ’s were systematically selected within each stratum
with probability proportional to size.
Each selected ed or group of ed ’s was then divided in­
to area segments where the number of area segments
corresponded to the size parameter for the ed . A sam­
ple segment was selected at random with equal pro­
bability from the set of segments.
Using sampling rates that yield equal final pro­
babilities of selection for all rental units in a psu , a
final set of rental units was selected from each seg­
ment after the segment had been completely listed and
a sample of units screened for tenure and eligibility by
bls field representatives.

Using this method, 19,000 rental units were selected
from 6,422 area segments. There was attrition of about
2,000 units due to conversions to owner housing. The
sample was augmented with approximately 1,500 new
segments and 4,000 rental units to minimally support
the rental equivalence concept of homeownership. This
augmentation followed a process similar to the original
sample with differential sampling for renter units in
areas of high owner concentration.
In order to collect the monthly information necessary
to calculate the rent index, the sample is divided into six
panels of approximately 3,800 units each. The units in
each panel are visited by bls field staff twice a year on a
6-month cycle. The information collected includes the
rents paid for the current month and the previous
month, information on extra charges and reductions, a
description of the unit, and the facilities included in the
rent. The latter questions are used to make quality ad­
justments to the calculated rents in order to assure that
the rent charge measured is for a set of units of a consis­
tent quality.
O w n er S u rv e y s
The first three steps of the sample selection pro­
cedures used to select the housing units are common to
both the rent and property tax sample. There are four
basic steps:
a.

b.

c.
d.

A multistage selection of clusters within a psu of yearround housing units (owner and renter) from the 1970
census ed summary data supplemented for new con­
struction.
A listing of all housing units (owner and renter) in the
sample segments; approximately 240,000 housing
units were listed.
Screening a subsample of the housing units (71,000
owner and renter units) to identify tenure.
Initiation of a subsample of owner units.

Estimation of Price Change

From the sample of screening schedules identified as
owner units or “ tenure not available,” a sample of
owner units built before 1970 was selected for the pro­
perty tax component of the c p i such that on the average
about 2.5 owner housing units were selected per seg­
ment. This resulted in a sample of approximately 17,000
designated housing units. However, since the number of
rental units was used to determine the measure of size to



Food, eomm©dities, and services
At the end of each pricing period, the estimate of the
one-period (t-1 to t) price change (price relative) is com­
puted for each item stratum, market basket, and
population. Only price quotes obtained in both the cur­
rent and previous pricing periods for the same or com­
parable items are used in the estimate. Also, where
15

possible, prices are normalized to a fixed quantity basis
(i.e., food prices are converted to a price per ounce).
The same quote weights are used for both the current
and previous period price quotes. The estimate of the
one-period price change for the zth item stratum for a
given market basket is computed as:

R zt,t-l

Rzt.o
Rzt-1,0

tions of weighted averages of 1- and 6-month rent
relatives designed to minimize mean squared error.
A final 1-month estimate of rent price change for the
particular market basket is:

.2^Wj(Pzit/p zi0
)
Property tax
Housing units included in the property tax sample
may be taxed by more than one jurisdiction. Thus, the
property tax data contains a record for each housing
unit/tax jurisdiction. The value of the property tax is
computed separately for the current pricing period (T)
and for the previous pricing period (T-1) for each hous­
ing unit/tax jurisdiction. The current survey period (T)
is defined to include all property tax values obtained in
months t through t-11. The previous survey period (T-1)
is defined to include the property tax values obtained in
months t-1 through t-12. The property tax value (Gx)
for the current period for a given housing unit/tax
jurisdiction is computed as:

le z

where:
Pzit

is the price o f the ith quote in the current pricing
period, t, for item z;

PziM is the price of the ith quote in the previous pricing
period, t-1, for item z;
Pzi0

is the base period price for the ith quote for item z;

Wj

is the quote weight for the ith quote in the current
pricing period for item z.

The quote weight, Wi} consists of the product of the
following factors: An estimate of the total daily expen­
diture ( e ) for the po ps (or non-pops) category for the psu
andthecpi-u orcpi-w populations; a duplication factor
(0 to reflect any special subsampling of outlets or
quotes; the percent of sales ( a. ) of the eli to the total
sales of the p o ps category in the outlet; the proportion
( b ) the number of eli hits selected is of the total in the
item stratum and the number of usable quotes ( m ) for
the ELi/half-sample for the item stratum and p s u :

G j = (At — Bj — ) E t — Lt
Ft

where:
A =
B =

assessment
assessed dollar value of capital change
(may be either positive or negative)
F = assessed dollar value of exemption
L = tax dollar value of exemption
E = tax rate

Wj = aEf/MB
Rent
Estimates of the monthly rent price relatives for each
market basket are calculated using cost weights and 1and 6-month estimates of rates of change.
Let Si be the set of units interviewed in time t in a
market basket which has rent values for time t and t-1,
and S6 be the set of units interviewed in time t in a
market basket which has rent values for times t and t-6.
The rents for the ith unit in a market basket for the
given time period are represented by riT, where T = t, t-1,
or t-6. The 1- and 6-month rates of change, Rtit_ and Rt t_
,
6,
are calculated by:
2

The property tax value (Gx.j) for the previous period is
computed as:
Gx_ —(Ax_ —
i
! Fx_])Ex_ — j
i Lx_
where the variables are defined above.
The 1-month price relative for a given market basket
is defined as:
n
i = l Wit Gxi(t t_ )
U
n-b
b
i = 1 ^ il ^Ti(t-l.t-ll) + i = 1 Wit G(T-l)i (M
2)

2

rit Wj
j
leS i rit W
ieSs
and Rt>
t^
R t.M
2
~
Wj
ht-l Wj
ieSi
ieS6 r it-6
where W; reflects the probability of selection adjusted for
nonresponse. Using Rt t , and Rt> 6 a composite estimate is
t_
made of a current month’s cost weight CWt for the
market basket:

where:
Wit
n

where P = 0.65. The value of P was based on simula-




16

=

n-b

CWt = P Rtit., CWt.j + (1 —
P)Rt> 6 CWt_
t_
6

=

=

The weight of the ith housing unit within the
market basket
the number of matched in-scope housing
unit/tax jurisdiction records in the market
basket
the number of matched in-scope housing
unit/tax jurisdiction records priced in periods
t-1 through t-11

b

tion o f the 40 cpi geographic areas, three downpayment
classes (0 to less than 10 percent down, 10 percent to 20
percent down, 20 percent or more down), and two
classes to distinguish between mortgages on new and ex­
isting houses.

the number o f matched in-scope housing
unit/tax jurisdiction records priced in t and
t-12.

H@yse prices and mortgage interest
The fh a housing units are stratified by age and living
area (in square feet). Each house sale or record to be
used in the cpi is classified into 1 of 15 cells or strata.
The 15 cells consist of five age classifications crossed by
three living area classes. The following boundaries are
used to define age codes for all market baskets:
Age

The average interest rate per loan is computed for each
stratum (1) for each market basket as follows:
Rit

=

2 Wk x y k x Rk / x Wk x Rk
I'

k

where Wk is a weight associated with each loan to repre­
sent the probability of selection of the lender (a basic
weight), a replicate factor, and a nonresponse factor,
Vk = value of the kth loan, and
Rk = interest rate of the kth loan.

Newly constructed
0-8 years
9-15 years
16-29 years
29-76 years

The monthly price relative for each stratum for each
market basket is computed as:

The following living area boundaries are used to
define living area codes for all market baskets:
The monthly price relative for a given market basket is
computed as:

Living area
500-999 square feet
1,000-1,499 square feet
1,500-6,000 square feet

Ru-, = ( S W jt_, Rlt,t_,) / 2 Wj t_
,

i

where Wj Mis a weight reflecting the relative morgage in­
terest cost in stratum 1 at time t-1.
In addition, there are cells for each m arket basket,
for fha and v a , and ceiling rates.

The average price per square foot is computed for each
stratum for each market basket as:
“jt
PSFjt =

P S F jit

njt i = l
where njt is the number of houses in the jth stratum at
time t; PSFjit is the price per square foot for the ith
house in the jth stratum at time t.
The monthly price relative for each stratum for each
market basket is then computed as:

Quality adjustments, linking, and im putation
One of the more difficult conceptual problems faced in
compiling a price index is the accurate measurement and
treatment of quality change due to constantly changing
product specifications and consumption patterns. The
concept of the c p i requires measures of the cost of pur­
chasing a fixed market basket of goods and services of
constant quality through time. Ideally, estimates would be
obtained for the dollar value of each quality change
resulting from a change in the model or item priced. This
estimate would reflect how much consumers value the
quality change. The direct measurement of the value con­
sumers place on quality change, of course, is rarely possi­
ble. As an approximation, bls uses several methods to ad­
just for quality change and account for the change in item
specifications. These methods may be categorizd as 1)
directly comparable, 2) direct quality adjustment, 3) link­
ing with overlap price, and 4) linking without overlap
price. In all cases, it is necessary to estimate a new base
period price in order to use the new item specification in
future, if not current, periods.

P S F jt

R

P S F jM

The monthly price relative for a given market basket is
computed as:
Ru-i = (SVjt_ Rj^d/SV*.,
i
J

J

where Vjt_ is a weight reflecting the total value of homes
i
in stratum j at time t-1. The weights were derived by
prorating the total dollar amount purchased in each
stratum to the total dollar amount purchased in all 15
strata each month. The purchasing data came from 1974
data supplied by f h a .
The mortgage interest cost component shows the
amount required to finance a given house as the result
of changes in both the interest rate and the price of the
house. The product, at the market basket level, of the
house price relative, discussed above, and the relative
for mortgage interest rates yield the relative to move the
weight for mortgage interest costs.
The mortgage interest rate change is estimated for
conventional loans, using cells from the cross classifica­




i

Directly comparable. If the new and old item specifica­
tions are considered directly comparable, i.e., the
characteristics that define the new specification are
essentially the same as the old item’s characteristics, the
base-period price for the new specification is set equal to
the base-period price for the old specification.
17

Quality adjustments exclude changes in style or ap­
pearance, 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. No satisfactory value can
usually be developed for such a change. In such cases, it
is ignored, and prices are compared directly.
If the new item specification is similar to the previous
one but has changed one or more of its component
parts, a quality adjustment may be made to establish
comparability between their prices. A synthetic
previous-period price for the new item (P*t-i) is
calculated as follows:

Direct quality adjustment. This is the most explicit
measure for dealing with specification changes. Direct
quality adjustments are frequently made for the food,
rent, and automobile components of the c p i . The con­
version of food prices to price per ounce accounts for
some quality adjustment. If the net weight of an item
changes, then the method used in recording food prices
will take into account this type of change in quality.
Quality adjustments are also made to the cost of ren­
tal housing used in the rent and rental equivalence in­
dex. bls collects the rent charged plus a description of
major services and facilities provided by the landlord. If
the services and facilities differ between two collection
periods when rents are compared, the rent for the cur­
rent period is adjusted to reflect the differences in ser­
vices between the time periods. For instance, if the
owner no longer provides a certain utility, bls would
calculate and add an estimate of the value of that utility
to the current rent in order to have an adjusted rent
value. This adjusted rent would be the cost of the same
services provided by the previous rent payment.
The most frequently cited example of direct quality
adjustment is the annual model changeover for new
automobiles. Direct quality adjustments are made for
changes in standard features between model years. This
estimate is based on all costs incurred in manufacturing
plus the established company markup to the selling price
of passenger cars. This estimate of cost applies to all
new features that are installed as standard equipment,
that is, features on cars in the same or comparable
series. Any former optional item that becomes standard
has a market price (i.e., the former option price) which
is the consumer value of that option for those who
bought it. For such items, the value of the quality
change is a weighted average of the former option price
and the producer cost. For all items that replace or
modify some previously existing feature, the estimate is
based on the difference in producer cost between the old
and the new feature, marked up to retail. In other
words, the estimate of total production cost for new
items is computed for both the new and the old feature.
The difference betw een these values is used as the
estimate of quality change.
Adjustments for quality change in the c p i new car in­
dex include structural and engineering changes that af­
fect safety, environment, reliability, performance,
durability, economy, carrying capacity, maneuverabili­
ty, comfort, and convenience. Although antipollution
equipment on automobiles does not directly increase
the quality of the automobile for the buyer, these de­
vices do improve the quality for consumers in general.
Consequently, quality adjustments are made for pollu­
tion controls to automobiles on the assumption that, by
legislative definition, the cost of installing antipollution
devices was at least as great as the value derived from
them.




PJti =

+ QA

where P ijt_ is the previous-period price of the old specifica­
,
tion,
and
QA is the dollar value of the quality change which
may be either positive or negative. After the above
imputation is made, the base-period price for the
new item (P*0) is computed as:

where P i0 is the base-period price for the previous
item.
Linking with overlap price. When a substitution occurs
and the quote has been coded as “ noncomparable link*’
and a price is obtained for both the old (Pit) and new
(P^) specifications in the same period (overlap pricing),
the estimation of the new base-period price is done at
the quote level to maintain the best estimate of continui­
ty for the new quote. The base-period price (P&) is
estimated as follows:
Pfo - Pt(Pi0/Pi,)
The linking of quotes with overlap prices is done before
item relatives are compared.
Linking without overlap price. For quotes which are not
comparable because of a change in specification
(substitution) and no quality adjustment or overlap
price can be obtained, the new specification price is not
used in the current-period estimate of the relative. Im­
plicitly, this means that the price change is assumed to
be the same as the average change of those quotes in the
same item stratum /m arket basket/pricing cycle.
To execute the link, an estimate of the long-term
change for the previous pricing period (Rzt_ for the
lj0)
item stratum and the current 1-month pricing period
relative (Rzt,t-i) is required. A new base-period price ( P |0
)
for the specification is computed as follows:

R zt-1,0 R zt,t-1

18

a market basket, one-half of the psu ’s/ eli’s are priced
every month. The second half of the psu ’s/ eli’s are priced
in the bimonthly alternate months. For a given item
stratum in a non-self-representing market basket, two
cost-population weights exist: One cost-population weight
for the psu ’s priced on cycle 1 and a second costpopulation weight for the psu ’s priced on cycle 2. The
price movement is a 2-month measurement of change for
a given cycle.
Seasonal items—those items that are available only at
certain times of the year—require a special method to
insure that they continue to be represented in the cpi.
There are two methods for identifying seasonal items.
First, eli’s are identified as seasonal on a regional basis;
that is, their seasonality is generally evident throughout
the marketplace within each of the four regions. Second,
some items are declared seasonal in a given sample outlet,
though they are not considered seasonal across the mar­
ketplace. For example, most big stores carry tennis
equipment year round, while smaller stores may carry
this equipment only during the spring and summer
months.
Outlets selected for seasonal eli’s are assigned double
the number of quotes normally allocated for those e l i ’s .
An eli is independently determined to be seasonal for
each region. H alf the quotes are assigned for initiation
to the first 6-month seasonal period (fall-winter), and the
remaining quotes are assigned to the second 6-month
period (spring-summer). An item specification is selected
and priced for the seasonal eli for the first seasonal
period from all items available in the outlet during the
period. The same is done for the second seasonal period.
This insures that there is always something priced for the
eli throughout the year. For items declared seasonal in spe­
cific outlets, either because the item is seasonal or the
outlet itself is seasonal, the item specification is main­
tained during its out-of-season months. In both cases,
this methodology prevents the substitution of regular
items for seasonal items; hence, the continued inclusion
of seasonal items in the cpi is assured.

The value of Rzt-i,0 is estimated by using the ratio of cost
weights for the item stratum /m arket basket/cycle,
where:

CWzt, is the cost weight of the previous pricing
period for the item stratum /m arket basket/population/pricing cycle.
CWz0is the base-period cost weight.
RztM is the one-period price change relative for
the stratum.
The imputation of missing prices takes place at the
end of each pricing period. A price (Pzit(i)) is imputed for
each quote which is missing a current-period price (Pzit)
and which satisfies the following criteria: 1) The quote
was scheduled for pricing during the current period or it
is a seasonal item which is out of season; and 2) it had
an actual or imputed price for the current version (no
substitution had occurred) in the previous period (t-1).
The imputed price is equal to the price for the previous
period (Pzit.,) times the price relative for the item
strata/market basket/pricing cycle:
^ ^ zit(I))

Pzit-1 R zt,t-1

The imputed price is not used in estimating the price
relative for the current period. It will be used to estimate
the relative (Rzt+i;t) if a valid price is obtained for the
item in period t + 1.
Timing ©f data eo!!@eti@n, seasonal items, and
average prices
To ensure representative pricing over time, each
month is divided into three pricing periods, each con­
sisting of 6 business days; an approximately equal pro­
portion of the sample is priced in each period. In addi­
tion, care is taken to assure that pricing of different
types of outlets is spread among the different pricing
periods.
Dividing the pricing period serves to: (1) Improve the
reliability of the price relatives for items having frequent
price changes, and (2) make processing of data more ef­
ficient by spreading out the flow of information from
the field to the central office over the month. All items
in a given outlet are priced during the same period in
each month in order to maintain a constant time interval
between successive price quotations.
Prices for eli’s are collected monthly in the following
market areas: New York, Philadelphia, Detroit, Chicago,
and Los Angeles, eli’s in all other market baskets are
priced bimonthly except the eli’s listed in table 3, which
are priced monthly everywhere. A timing chart of psu ’s
assigned to pricing cycles is found in appendix D. For the
self-representing psu ’s, all bimonthly eli’s are priced in the
months as assigned. For the nomself-representing psu ’s in



Table 3. Entry level items priced monthly everywhere
All food at home items
Alcoholic beverages at home
Fuels and utilities
Postage
Babysitting
Domestic services
Gardening
Used cars
Gasoline
Motor oil, other automotive
products

Tires
State auto registration
Driver’s license
Vehicle inspection
Admission to movies, etc.
Tobacco, smoking accessories
Personal care services
College textbooks
College tuition
Elementary and secondary
school tuition

Average prices, estimated from cpi data, are published
for food items, gasoline, natural gas, electricity, and fuel
oil. Two definitions of average prices can be formulated:
a. Accept only quotes with specifications of quantity
meeting a fixed quantity definition (only prices for
quarts of milk are included in any tabulation); or

19

total mse of the index will be reduced.
The procedure of the composite estimator is: Let Xm
and Xr be respectively the average market-basket expen­
diture and the average regional expenditure for a given
item stratum (or ec level), where the region is the census
region in which the market basket is located. Let Wmbe
the proportion of cu’ in the region for the mth market
s
2
_
basket and o— the variance of Xm
.
Am
The composite estimator (Xc) is a linear combination
of Xmand Xr, chosen to minimize the mse:

b. Convert all prices to price per normalized quantity
and use all prices to estimate a defined, fixed quantity
(for example, regardless of whether the prices were for
gallons, quarts, or pints of milk, convert all prices to
price per ounce and multiply by 32 to estimate a price
per quart).

The cpi average prices are generally of the second type
unless otherwise specified. The estimation of average
prices is:
? W i,(P zit/P z i0 )

p

_ _i_________
SW it/P z
i0
i

Xc = b Xr + (1 —b) Xm

where Wit is an estimate of expenditure. In the equation,
dividing the expenditure weight by the base-period price
for a quote gives an implicit estimate of quantity. Thus,
the average price is conceptually a weighted average of
prices where the weights are quantity amounts.

where:
<4 0 - Wm
)
b = — ----= —
E(Xr - Xm
)2

The estimate of‘b’is obtained by estimating its constitu­
ent parts. Estimates of variance for the local and regional

Estimation ©f the CPI Cost-Population
W eights

2

using the measured relvariances of items reported from
the ce survey and generalizing the relvariances using
average expenditure, proportion of units reporting posi­
tive expenditure, and sample size as dependent variables.
Also, estimates of the average bias (A(r) for each major
group are made, where i represents the item stratum
under consideration in the rth region. The estimate of the
denominator of b is:

For each of the market-basket areas for which data are
published, for each item stratum and U or W population,
an estimate of expenditure is needed to define the market
basket of goods and services for which the index is com­
puted. This expenditure estimate is called a cost-popula­
tion weight. The estimation of cost-population weights is
the product of estimates of mean expenditures per con­
sumer unit derived from the 1972-73 ce survey and esti­
mates of consumer units obtained from a special tabula­
tion of the number of families, subfamilies, and unrelated
individuals by population definition and market basket
from the 1970 census sample detail files. Calculation of
the mean expenditures requires three steps: (1) Estima­
tion of preliminary cost weights, (2) estimation of mean
expenditures using a composite estimation procedure,
and (3) estimation of final mean expenditures using a
raking process on the mean expenditures determined in
( 2).

E(Xr - Xm = c L (1 - 2Wm + a2 + A l
)2
)
-

A
-m

Ar

The estimate of b is:

axm
(1-Wm)

b = min

’ <4r 0 - 2Wm) + < 4 + A2

Am «
■

Ar

The Air are measures of the variation between market
baskets and can be interpreted as estimates of a p r io r i
variances on the parameters E(Xm A2 was estimated
). r
within a region and major group by:

Preliminary mean etxpendityres
Preliminary mean expenditures are calculated for each
item stratum and expenditure class, for each population,
market basket, and region. These mean expenditures are
estimated using information from the ce survey and are
the simple weighted averages of the expenditures for the
particular item stratum or expenditure class for all con­
sumer units in the market-basket population desired. The
weights used are those weights described in the section on
thecE survey. Also, percentages of units by market basket
and population with positive expenditures for each item
stratum or ec are estimated.

8
1-6

M

)
x 2 WmVar(Xm
m=l
where 5 is a measure of the average intramarket basket
correlation of item stratum within a major group and
Var(Xm are the unit within variances of the market
)
baskets within a region of M market baskets. Thus, the
resulting composite estimator can be viewed as an empir­
ical Bayesian estimator (with the minor adjustments of
(1 - Wm
)).
The above process defines a “shrinkage” estimator. To
limit the shrinkage, provisions are made to adjust the
estimate resulting from Xc as follows:

Composite estiimatiora process
A composite estimator is used to decrease the mean
square error (mse) of an average expenditure for a given
item stratum within a market basket. By doing this, the




2

mean expenditures denoted by 0— and 0— are made
Ar
Am

if |Xc - Xm <ko £7_
|

Xn

Xi=<
Xm- Sgn(Xr - Xc) k0 o
a

20

X,

otherwise

where Sgn (Xm- Xc =
)

proach, measured changes in the cost of five items (home
purchase, contracted mortgage interest cost, property
taxes, property insurance, and maintenance and repairs).
It did not isolate the consumption aspects of homeowner
cost from the investment aspects.
The flow-of-services concept isolates the consumption
aspect by focusing on the shelter service of owneroccupied homes. This is the homeowner cost item whose
price change should be included in an index of consumer
prices. The cost of the shelter services of owner-occupied
homes is called implicit rent (just as the cost of the shelter
services of renter-occupied homes is called rent). Implicit
rent covers both direct and indirect expense items of
homeowner cost. Direct items include mortgage interest,
property tax, and maintenance and repair costs. Indirect
items are those such as physical depreciation and the cost
of keeping wealth tied up in the equity in one’s home,
which homeowners do not pay for on a regular basis.
Price appreciation is a negative component of implicit
rent.
Rental equivalence attempts to measure the change in
the cost of obtaining in the rental market housing services
equivalent to those provided by owner-occupied homes.
For the 1983 rental equivalence implementation, a new
homeowners’ cost component was derived to replace the
old-method homeownership component. The new com­
ponent contains two items: (1) Owners’ equivalent rent,
which has most of the weight and represents most of
implicit rent; and (2) household insurance, which contains
that part of property insurance which is not in owners’
equivalent rent.
In addition, the previous maintenance and repairs
component was made a new component covering those
owners’ expenses not in homeowners’ cost as well as
renters’ expenses. The weight for household appliances
was reduced to remove that portion of those expenses in
homeowners’ cost.

+1 if average bias defined by (Xm—Xc) > 0
0 if average bias = 0
— if average bias < 0
1
The parameter (k0) was determined for each popula­
tion, each survey, and each major group by graphing
alternative estimates of the mse of Xc and selecting the
point which gave the major portion of gain in reduction
of mse and minimum level o f average change for estimates
in the major group.
Raking
After composite estimation, and in order to reestablish
data consistency between item strata and ec levels, an
iterative ratio estimation procedure (raking) is performed
to marginal mean expenditures of the regional values
from the ce survey. That is, the sum of the expenditures
for all item strata within an ec for a market basket is
forced to equal the total expenditures for the ec in the
market basket, and the sum of the expenditures for a
specific item stratum across all market baskets within a
region is forced to equal the original regional estimate of
expenditures for the item strata.
Special e©st weight procedures for housing
For the asset approach to homeownership, the defini­
tions of cost weights for home purchase and mortgage
interest are of special interest. The resultant values are
subject to the composite estimation defined above.
The cost weight for the home purchase component of
the cpi is determined from the ce survey by asking the total
purchase price of any newly purchased home and the
total realized amount of any home sold over a 5-year
period. A mean annual expenditure per consumer unit in
each market basket for home purchase is calculated, and
a mean annual amount realized from home sales per
consumer unit is calculated. The difference between the
two values is taken as the mean net amount spent on
home purchase.
The mortgage interest cost weight for each market
basket is determined from the ce survey by asking if a
consumer unit has acquired any home mortgages during
the previous 5-year period, the term of the mortgage, and
the interest rates for each mortgage. For each cu taking
out a mortgage, the total interest on that mortgage due
during the first half of its term is computed. From these
totals, the average mortgage interest due per consumer
unit in each market basket is then computed and divided
by 5 (the number of years observed.)

Sample selection

The price change for the new item, owners’ equivalent
rent, is obtained from thecpi rent sample. However, the
sample has been reweighted and augmented for this pur­
pose. In order to provide a sample of rental units to
support rental equivalence calculations, the present sam­
ple of 21,000 rental units was augmented by approxi­
mately 2,000 rental units. These new units were selected
to provide a higher proportion of rental units than was
formerly the case in strata with exceptionally high con­
centrations of owner units. This is to assure that an
additional number of the proper type of rental units are
available for matching with owner units.
In addition to the augmented sample of rental units, a
sample of 40,000 owner units was selected from the same
sample of 6,422 original area segments, 1,500 augmenta­
tion area segments, and new construction segments that
were selected for the rent sample.

Rental Equiwalenee
The new concept of rental equivalence, introduced for
thecpi-u in January 1983, measures changes in the cost of
shelter which owner-occupied homes provide. The old
homeownership method, which is called the asset ap­



21

In the 1983 version of rental equivalence, no adjustment
is made to the rent value rik to account for different
services provided in the contract rent for different renters.
That is, some rents include utilities and some do not.
Later versions may adjust contract rent to “pure” rent at a
macro level.
The weight Wb for the ith rental unit is calculated as:

Data collection
Data collection for rental equivalence is conducted as
part of the rent data collection process. Rental units are
visited on a regular basis every 6 months. Owner units are
visited in a different fashion; they are divided into six
panels and placed on 6-month cycles. Because informa­
tion on tenure and characteristics is generally stable for
owner units, the owner units are further divided into two
subgroups. These subgroups are visited for information
on an alternate 6-month cycle. Thus, each individual
owner unit is visited once a year. The primary purpose of
visiting owner units is to identify current tenure. Infor­
mation collected from owner units includes unit descrip­
tions to be used in matching the unit to rental units in
order to determine equivalence.

Wj' = Li Wi
where:
Wi is the original unit weight (essentially the number of
rent units represented by the ith sample unit times a
noninterview adjustment). Li is the ratio from census data
of the number of owner units to rental units in the segment
from which the ith unit was selected. Under two assump­
tions, (1) if the same sampling rate is used for each rental
unit in a selected area, that is, Li owner units would have
been selected each with weight Wi' and (2) if each owner
unit in the area is well represented by the sample of rental
units selected from the area segment, then ad and ad are
proper estimates of the 1-month and 6-month changes for
a rental equivalence type of measurement.

Calculation of cost weights
The calculation of the preliminary cost weights for
owners’ equivalent rent is based on estimates made by
respondents who own their residences. They are asked to
estimate the rental value of their unit as if it were rented
with a specified set of services for the purpose of calculat­
ing the rental equivalence cost weights. This value is to be
considered as an actual expenditure for the respondent.
Such expenditure estimates are used as a basis of mean
expenditure estimates in the same manner as actual
expenditures are used to estimate expenditures for other
items in the c p i .
Because rent payments tend to include payment for
other expenditures aside from the cost of shelter, adjust­
ments are made to other cost weights to avoid double
counting of expenditures. For this reason, rental equival­
ence is assumed to include all costs for housing, most
maintenance, and portions of the costs for large applian­
ces and certain types of home insurance. The cost weights
for these items were adjusted by subtracting a proportion
determined by comparing full expenditures paid for these
items by homeowners and the expenditures by renters for
the item for the same type of rental units.

Freeiiloin @ Estimates
f
Basie approach to C variance estimation
PS
An important advantage of probability sampling
methods is that a measure of the sampling error can be
computed directly from the sample data. The c p i sample
design accommodates error estimation by making two
selections (half-samples) of items and outlets and, there­
fore, two samples of quotes in each self-representing p s u
and one in each non-self-representing p s u . Given this
structure, which reflects most stages of the sample design,
different variance estimation techniques can be employed.
The method b l s plans to use depends upon the statisti­
cal independence of the estimated indexes for individual
market baskets.1 For each market basket, two independ­
9
ent estimates of the market-basket index can be con­
structed using the half-samples specified in the design.
Squared differences of these indexes (properly scaled)
provide estimates of the variance of the market-basket
index and form the building blocks upon which the
regional and national variance estimates are based. These
variance estimates may be for the index for all items or
for a subset of items. Also, variance estimates for price
change can be computed as well.
The variance estimates will not reflect all elements of
error in the index. In particular, they will not reflect

CaS©ylaSiosn ®f monthly imid©K amid relative ©hang©
The calculation of the rental equivalence relative used
from January 1983 forward follows the calculation
procedures outlined earlier for rent. The only difference
between the two results is that the preliminary 1- and
6-month changes are calculated using weights WY which
are different from the weights for the rent units used in
calculation of the rent index.
Consider
X
,

ieS 1

W,' Tit

X
,

W;' Tit

;eS6

1 The independence is violated somewhat by the fact that the same
9
half-sample of ELI’s is used in more than one PSU within a
region. However, since the selection of specific items to be priced
involves considerable subsampling of the ELI’s within outlets, the
market-basket indexes may be regarded as statistically independent for
variance estimation.

ai = ~r:---- — ------ and a6 = — ---- — ------X Wi Tit-1
X Wi' rit_
6
ieSi

ieS6

where rik, Si and S6 are as defined in the section on price
relative estimation for rent.




22

statistical error arising from the estimation of the base
expenditure weights. The variance estimates may be
thought of as conditional on the values of these estimated
base expenditures, bls expects to include an analytically
derived estimate of this expenditure component of error
in any published sampling errors. Certain other errors,
such as any systematic biases in quality adjustments or
imputation, also may not be reflected.

current relative for the same market basket / half-sample/
item stratum/pricing cycle. The replicate cost weight is
left unchanged if the item stratum is not priced that
month for the market basket/ half-sample/ pricing cycle.
CaSeuSats@ of variances for She indent and
n
price change
The index may be for all items or a subset of items (an
item stratum, an ec, or a major group). It may be at the
national, regional, or market basket level.
Let the quantity Uht,o be a relative importance com­
puted as a ratio as follows:

EsSSmation off relatives by hallt-samp!©
Estimates of price relatives for each market basket are
currently computed for estimation of the index. The
method of variance estimation requires another relative
be computed for each pricing cycle and item stratum
half-sample in each market basket. All computations are
based on the cpi- u population with sales and excise taxes
included.
For commodities and services, relatives are computed
at the item stratum/market basket/half-sample level for
each pricing cycle. If necessary, because of an insufficient
number of sample quotes, item strata are collapsed
separately and independently within each market basket/
half-sample pricing cycle.
For the rent survey, for each self-representing psu , each
rental unit is assigned to half-sample A or half-sample B.
For non-self-representing market baskets, the index psu
determines the half-sample for a given rental unit.
For the purpose of relative computation for rent, arti­
ficial replicate cost weights for each market basket/
half-sample are constructed to provide a basis for weight­
ing the 1-month and 6-month relatives together.

a. The numerator is the sum of the total cost weights over
the item strata and pricing cycles being considered for
market basket h at time t.
b. The denominator is the summation of all base-period
cost weights for the item strata and market baskets
being considered. The base period is December 1977
for variance purposes.

The corresponding quantities based on the replicate
cost weights for each half-sample (instead of the total cost
weights) are denoted, respectively:
n r
Uht,o or TUht,o

T

A variance formula for the index It,0 is the sum over all
market baskets h being considered:
2

o ,( I ,o ) =

*

“

Uht’°)2 +

( U S

-

U h t,o )2 ]

The variance is also estimated by a second procedure
summed over all market baskets being considered:

Fet:

2 t-i X_ 2 100 /T (A tt(BK2
t )
o 2(1,o) — , — (Uht,0 — Uht/o)

rju denote the corresponding 1-month replicate relative
Ah‘ denote an artificial replicate cost weight for market
t
basket h at month t
rht-6 denote the corresponding 6-month replicate relative.

n

4

The price change for period t relative to a period m
months earlier will be denoted by:
It,0

Then Aht is computed by:

I t-m
,0

Am = 0.65 A ll, r( t + 0.35 A(U rKL
h
h

In order to compute the estimate of the variance of
the covariance of the numerator and denominator must
be estimated. The estimated covariance of It,o and It-m is
,o
the sum over all market baskets h being considered:

The final relatives [Rm] will be treated as monthly items for
each market basket. They are computed by:

C (I,,0 It-m = ^
l
,o)

RepSiesifie cost weights
The term T-cost weight (tcw ) refers to the costpopulation weights (for the cpi- u population with sales
and excise taxes included) that are used for estimation of
the index. The term R-cost weight ( rcw) refers to the
cost-population weights that will be used for estimation
of the variance of the index. In general, the methods used
to compute the rcw and tcw are very similar. A data base
of rcw’s is created, maintained, and updated monthly. A
replicate cost weight is updated by multiplying it by the



[(U( to - Uht,o) (UltU - U .o)
h
ht-m

+ (U - U
'hu'o
h,.o) (U'ht’m - Uht-m
.0
.o)]
An alternative estimate of the covariance is denoted by
C (It,o, It-m.o), and is the sum over all market baskets h
2
being considered:
C (1,0 It-m
l
,o)

2 1002
h

4

[(U|$ - Ul^o) (U .0 -ULB
hft-m
tU,o)]

Then one variance o f I , t-m is estimated by the Taylor
series approximation:

23

2

a l(It,t-m ) =

—------- “ 2

( V l( It ,o ) +

2It,t-m

C l ( I t>0 , It - m ,o )J

An alternative estimate of o2 of




obtained by substituting a 22 and C for a2i and C in the
2
i
formula above.
The foregoing discussion has treated two half-samples.
The method can be extended for those market baskets
with four and six half-samples.

2

a l(It-m ,o )

is o12

24

T@©tol<ss)l] ^©ferem©®®

1. Braithwait, Steven D. “The Substitution Bias of the
Laspeyres Price Index: An Analysis Using Estimated
Cost-of-Living Indexes,” American Economic Re­
view, March 1980.
2. Bounpane, Peter, and Curtis A. Jacobs. “Point of
Purchase Survey and Its Uses in the Consumer Price
Index Revision,” Proceedings, Social Statistics Sec­
tion, American Statistical Association, 1975.
3. Gillingham, Robert F. “A Conceptual Framework
for the Revised Consumer Price Index,” Proceedings,
Business and Economics Statistics Section, Ameri­
can Statistical Association, 1974.
4. Gillingham, Robert F., and Walter Lane. “Changing
the Treatment of Homeownership in the cpi ,” Monthly
Labor Review, June 1982.
5. Ginsburg, Daniel H. “Medical Care Services in the
Consumer Price Index,” Monthly Labor Review,
August 1978.
6. Jacobs, Curtis A. “Sample Design for the Consumer
Price Index,” Proceedings, Social Statistics Section,
American Statistical Association, 1978.
7. National Bureau of Economic Research. The Price
Statistics o f the Federal Government: Review, A p­
praisal, and Recomm endations. W ashington,
National Bureau of Economic Research, General
Series, Number 73, 1961. Also appears in Govern­
ment Price Statistics: Hearings, Subcommittee on
Economic Statistics of the Joint Economic Commit­
tee, 87th Cong., 1st sess., Part 1, January 24, 1961.
U.S. Government Printing Office, 1961.
8. Norwood, Janet, c p i Issues, Report 593, U.S. Depart­




9.

10.

11.

12.

13.
14.

15.

16.

25

ment of Labor, Bureau of Labor Statistics, February
1980.
Norwood, Janet. Problems in Measuring Consumer
Prices, Report 697, U.S. Department of Labor,
Bureau of Labor Statistics, September 1983.
Poliak, Robert A. “The Theory of the Cost-of-Living
Index,” Bureau of Labor Statistics Working Paper
11, June 1971. Examines the theory of the “cost-ofliving index” and the “preference field quantity
index,” and their relation to price indexes.
Poliak, Robert. “The Treatment of Quality in the
Cost-of-Living Index,” Bureau of Labor Statistics
Working Paper 90, June 1979.
Triplett, Jack E. “The Measurement of Inflation: A
Survey of Research on the Accuracy of Price Indexes,”
in Analysis o f Inflation, Paul H. Earl, ed., ch. 2.
Lexington, Mass., Lexington Books, 1975.
Triplett, Jack E. “Reconciling the cpi and the pce
Deflator,” Monthly Labor Review, September 1981.
Samuelson, Paul A., and S. Swamy. “Invariant Eco­
nomic Index Numbers and Canonical Duality: Sur­
vey and Synthesis,” American Economic Review,
September 1974, Vol. LXIV, No. 4.
Shiskin, Julius. The Consumer Price Index: Con­
cepts and Content Over the Years, Report 517, U.S.
Department of Labor, Bureau of Labor Statistics,
May 1978 (revised).
U.S. Department of Labor, Bureau of Labor Statis­
tics. The Consumer Price Index: History and Tech­
niques, Bulletin 1517, 1966.

Ap>pein)dfa®§:
Appendix A. Chronology of changes in the Consumer Price Index, 1890 to date
Survey providing
expenditure weight
Date

Base
period

Census
providing
population
weights

Number
of areas
included

None

Varied

Group
weights

Item
weights

1890* . . .

None

1901

Varied

1 9 1 9 ___

1917-19

1917-19

Earnings
of
chief earner

Family
composition

Source and
amount of
family income

Length
of
employment

Economic level,
length of
residence,
nativity, and
race

Title(s)

1913

Two or more
persons.

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

No limitation.

No limitation.

232

Feb...........
1921

Minimum of
Salaried worker At least 75 per­
earning $2,000 cent from
husband,
or less. No
principal
wife, and 1
child who
limitation on
earner or
was not a
wage earners.
others who
boarder or
contributed
lodger. No
all earnings to
boarders nor
family fund.
more than 3
lodgers
present.

7
33

Two or more At least $300.
At least $500.
At least 1,008
No relief fami­ Indexes of the
persons. Not
Salaried
Less than one- hours spread
lies, either on
cost of living
worker earn­
more than 2
fourth from
over 36 weeks. direct or work of wage earn­
boarders or
ing less than
interest, divi­
relief; white
ers and lowerlodgers, or
$2,000 during
dends, royal­
only, except
salaried work­
year or less
guests for
ties, specula­
where black
ers in large
more than 26 than $200
tive gains,
population
cities.
guest-weeks.
during any
rents, gifts, or
was significant
income in
month. No
part of total;
kind. No rent
upper limita­
in area 9
in payment of
tion on wage
months or
services. Less
earners.
more.
than 3 months’
free rent. No
subsidiary
clerical work­
er earning
$2,000 or
over.

No slum or
Cost of living.
charity famil­
ies; white only;
in area entire
year and in
the United
States 5 years
or more; no
non-English
speaking
families.

3Average
1920-30

Sept..........
1935
1923-25
19354
Aug..........
1940s

1934-36

61934-36

1935-39

1930

34

May . . . .
194D
91940

July ___
1943

Price Index
for Moderate
Income Fami­
lies in Large
Cities.

1945

Jan...........
1951'°

1947-49

"1934-36

1950

Two or more
persons.

No limitation.
(Family in­
come not in
excess of
$10,000.)

See footnotes at end of table.




26

Family income Family head
under $10,000
must have
after taxes in
been em­
the survey
ployed at
year. No mini­ least 26
weeks.
mum income
limit, except
that families
with no in­
come from
wages or
salaries were
excluded.

No exclusion
for receipt of
relief as such,
but only fami­
lies with wage
or salary earn­
ings included.
No length of
residence,
nativity, or
racial limita­
tions.

Appendix A. Chronology of changes in the Consumer Price index, 1890 to date-Continued
Survey providing
expenditure weight
Date
Group
weights
Jan...........
19531
2

Item
weights

1 1950
3

>31950

Jan...........
1962

Base
period

Census
providing
population
weights

141947 9

46

1960

Jan...........
1966*8

Jan........... 2*1972-73 2'1972-73
19782
0
to date

Earnings
of
chief earner

Family
composition

Length
of
employment

No specific re­
quirement, but
major portion
of income of
family head
must be from
employment
as wage earn­
er or salaried
clerical
worker.

50

Economic level,
length of
residence,
nativity and race

Titlefs)

Short title: Con­
sumer Price
Index
Complete name:
Index of
Change in
Prices of
Goods and
Services Pur­
chased by City
Wage-Earner
and ClericalWorker Fami­
lies to Main­
tain Their
Level of
Living.

Families of 2 No limitation.
or more per­
sons and
single work­
ers; at least 1
full-time
wage earner.

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

Same as above
for earner
and clericalworker in­
dex. No limi­
tation for
urban con­
sumer index.

Same as above Same as above Same as above I) Consumer
for wage-earn­ for wage-earn­ for wage-earn­
Price Index
er and clerical- er and cleri­
er and clericalfor Urban
worker index.
cal-worker in­ worker index.
Wage Earners
No limitation
No limitation
dex. No em­
and Clerical
for urbanployment re­
for urbanWorkers
consumer
quired for ur­
(CPI-W). 2)
consumer
index.
index.2
2
ban-consumer
Consumer
index.
Price Index
for All Urban
Consumers
(CPI-U).

No restriction
Consumer Price
on other than
Index for Ur­
the wage-earn­ ban Wage
er and cleri­
Earners and
cal-worker
Clerical
definition.
Workers.

56
1967
1970

85

1 Item weights were revised for only the 7 cities for which 1947-49 expenditure
2
data were available. I ndex published in February for January 1953. Linked to old
series as of December 1952. Old series also published for a 6-month overlap period.
1 Data were adjusted to 1952 for weight derivation.
3
1 Indexes were also calculated on the base of 1935-39=100 through December
4
1957.
1 Index published in February for January 1962. Indexes were also calculated
5
on bases of 1947-49=100 and 1939=100.
1 Index published March 3 for January 1964. Linked to old series as of
6
December 1963. Old series also published for a 6-month overlap period.
1 Data were adjusted to December 1963 for weight derivation.
7
1 Index published in February for January 1966. Linked to old series as of
8
December 1965.
1 Index published in February for January 1971. Indexes were also calculated
9
on the 1957-59=100 base.
20 Index published in February for January 1978. Linked to old series as of
December 1977. Old series also published for a 6-month overlap period.
2 Data were adjusted to December 1977 for weight derivation.
1
22 Coverage was expanded to include wage earners and clerical workers in the
entire nonfarm parts of the metropolitan areas in addition to those living within
the urbanized areas of the metropolitan areas and urban places of 2,500 or more
inhabitants.

1 Food Price Index 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 1918-29 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 betwen 1925-29 were recomputed retroactively with group weights
based on the average of 1917-19 and 1934-36, indexes between March 15, 1930,
and March f5, 1940, were recomputed retroactively using 1934-36 group weights.
6 During World War II, weights were adjusted to account for rationing and
shortages.
7 51-56 cities included in the food index.
8Index published in May 1941 forM arch 14,1941. Food indexes were based on
51 cities.
9 1940 census data were supplemented by ration book registration data.
1 Index published in March 1951 for January 1951.
0
1 Indexes between January 1950 and January 1951 were revised retroactively
1
for all items and group indexes. Indexes for rent and all items were corrected for
the new unit bias from 1940. Old series also published through 1952.




Source and
amount of
family income

'M957-59

Jan........... ,71960— *
61 71960—
61
19641
6

Jan...........
1971 *
9

Number
of areas
included

27

Appendix 1. Relative importance of components in the Consumer Price Jndexes: U.S. city average, December 1977
(Percent of all items)
U rb a n W a g e

U rb a n W a g e
A ll U r b a n

C le r ic a l

(C P I-U )

A ll U r b a n

E a rn e rs a n d

C o n s u m e rs

G r o u p a n d ite m

W o rk e rs

G r o u p a n d ite m

E a rn e rs a n d

C o n s u m e rs

C le r i c a l

(C P I-U )

W o rk e rs
(C P I-W )

( C P I- W )

E x p e n d it u r e c a te g o r y

E x p e n d it u r e c a te g o r y
1 0 0 .0 0 0

A ll it e m s .................................................................................................

1 0 0 .0 0 0

F o o d a n d b e v e r a g e s — C o n t in u e d
..............................................................

3 .3 4 9

..................................................................

.4 3 5

.4 6 6

.211

.2 2 9

O th e r fo o d s a t h o m e
2 0 .4 8 1
1 9 .2 9 8

C a n d y a n d c h e w in g g u m

...............................................

................................................................................

1 2 .2 3 6

1 3 .4 9 3

S u g a r a n d a r t if ic ia l s w e e te n e r s ....................................

.1 1 6

.1 2 8

C e r e a ls a n d b a k e r y p r o d u c t s .............................................

1 .5 3 0

1 .6 9 2

O t h e r s w e e t s ............................................................................

.1 0 7

.1 0 9

.............................................

.3 8 5

.4 1 9

Food

.......................................................................

......................................................................................................

Food at hom e

C e re a l a n d c e re a l p ro d u c ts

v

3 .7 4 5

1 8 .8 1 4
1 7 .7 1 9

F o o d a n d b e v e ra g e s

S u g a r a n d s w e e ts

F a ts a n d o ils

..............................................................................

.3 6 0

.3 9 0

F lo u r a n d p r e p a r e d f l o u r m i x e s ....................................

.1 0 4

.1 1 8

M a r g a r in e

.................................................................................

.1 0 7

.1 0 9

C e r e a l ..........................................................................................

.1 6 4

.1 7 0

N o n d a ir y s u b s t it u te s a n d p e a n u t b u t t e r .................

.0 7 2

.0 7 8

O t h e r fa ts , o ils , a n d s a la d d r e s s in g s

........................

.181

.2 0 2

....................................................

.............................................

.1 1 7

.1 3 0

.....................................................................

1 .1 4 5

1 .2 7 3

1 .5 1 3

1 .7 2 8

W h it e b r e a d ..............................................................................

.3 2 4

.3 7 2

C o la d r in k s , e x c lu d in g d ie t c o la

.................................

.5 1 8

.6 5 0

O t h e r b r e a d ..............................................................................

.1 1 5

.1 2 0

C a r b o n a te d d r in k s , in c lu d in g d ie t c o l a ...................

.2 8 8

.3 1 6

F re s h b is c u i t s , r o lls , a n d m u f fin s

...............................

.1 1 7

.1 3 2

R o a s te d c o ff e e

.2 3 6

.2 6 3

.............................................

.1 3 5

.1 5 5

F re e z e d r ie d a n d in s t a n t c o f f e e ....................................

.2 1 2

O t h e r n o n c a r b o n a t e d d r in k s ........................................
O t h e r p r e p a r e d f o o d s ...........................................................

.2 5 8

.2 9 2

1 .0 4 1

1 .1 6 1

R ic e , p a s ta , a n d c o r n m e a l
B a k e ry p ro d u c ts

F re s h c a k e s a n d c u p c a k e s

N o n a lc o h o lic b e v e ra g e s

.......................................................................

......................................................................................

.1 4 7

.1 6 2

C r a c k e r s a n d b r e a d a n d c r a c k e r p r o d u c t s ............

.081

.081 i.

............

.1 2 4

.1 3 9

C anned and packaged so up

.....................

.1 0 2

.111

S nacks

C o o k ie s

F re s h s w e e t r o ll s , c o ff e e c a k e , a n d d o n u t s
F r o z e n a n d r e f r ig e r a t e d b a k e r y p r o d u c t s
a n d fr e s h p ie s , t a r ts , a n d t u r n o v e r s

........................................

.1 0 6

....................................................

.1 6 2

.1 7 6

.......................................................................................

.1 8 9

.2 1 6

.1 0 6

.1 1 2

F ro z e n p r e p a r e d f o o d s

S e a s o n in g s , o liv e s , p ic k le s , a n d r e lis h
...........................................

3 .9 4 3

.......................................................

3 .7 2 0

4 .1 5 4

M is c e lla n e o u s p r e p a r e d f o o d s

2 .8 8 7

3 .2 7 4

O th e r c a n n e d a n d p a c k a g e d p re p a re d fo o d s

1 .4 3 6

1 .5 8 4

M e a ts , p o u lt r y , a n d f is h
M e a ts

B eef a nd veal

.......................................................................

4 .3 9 9

...................

..........................................................................................

M e a ts , p o u lt r y , f is h , a n d e g g s

.2 0 7

O t h e r c o n d im e n ts

................................................................

.1 6 6

.11 1

.1 8 3

.1 6 0

.1 8 5

...

.1 5 3

.1 8 0

F o o d a w a y fr o m h o m e ..............................................................

5 .4 8 3

5 .8 0 5

....................................

............................

.3 7 2

.4 1 8

C h u c k ro a s t

........................................................................

.1 6 8

.1 8 5

D i n n e r ...............................................................................................

1 .9 8 9

1 .9 4 6

R o u n d ro a s t

.......................................................................

.1 5 7

.1 7 4

O t h e r m e a ls a n d s n a c k s

1 .0 3 8

1 .3 0 6

R o u n d s te a k

.......................................................................

.0 9 5

.1 1 6

U n p r ic e d it e m s 1 .........................................................................

.6 9 8

.5 6 0

S ir lo in s t e a k

.......................................................................

.101

.1 0 5

A lc o h o lic b e v e ra g e s .....................................................................

1 .0 9 5

1 .1 8 3

.5 4 4

.5 8 7

A lc o h o lic b e v e ra g e s a t h o m e ...............................................

G ro u n d b e e f o th e r th a n c a n n e d

O th e r b e e f a n d ve a l

............................................... ..

P o r k .............................................................................................
B acon

......................................................................................

Chops

.....................................................................................

...............................................................................................

.9 5 5

1 .1 0 4

.1 6 8

.1 8 2

W h is k e y

B e e r a n d a le

.......................................................

.4 9 6

.201

.2 0 0

.................................................................................................

.2 0 3

.2 3 5

W in e

.2 0 7

O t h e r a lc o h o lic b e v e ra g e s a t h o m e

.131

.1 6 2

, .0 9 4
.1 8 2

.1 0 7

..........................................................................

.4 9 6

.5 8 6

H o u s in g

.......................................................................

S h e lt e r

O t h e r p o r k ............................................................................
O t h e r m e a ts

.1 1 0

...........................................................

.1 8 0

.2 2 7

.0 5 4

.041

.................................................................................................

4 3 .9 0 8

4 0 .6 7 9

.................................................................................................

2 9 .1 8 3

2 6 .3 7 4

R e n t, r e s i d e n t i a l ............................................................................
O t h e r re n ta l c o s ts .......................................................................

.2 2 7

...............................................

.101

U n p r ic e d it e m s 1 ....................................k , ........................

.0 0 2

098
.0 0 2

L a m b a n d o r g a n m e a ts

.1 1 0

..........................

.1 9 8

U n p r ic e d it e m s 1 ' . .......................................................................

.1 2 9

O t h e r lu n c h m e a ts

.1 0 8

.1 2 0

5 .6 2 4

5 .3 2 2

A lc o h o lic b e v e ra g e s a w a y fr o m h o m e

.1 3 0

.1 0 4

B o l o g n a , liv e r w u r s t , a n d s a la m i

.1 3 4

...............................

.2 1 0

............................

F r a n k fu rte rs

L o d g in g w h ile o u t o f t o w n

..................................................

..................................................................

T e n a n t ’s in s u r a n c e

.7 1 2

.4 8 8

.4 3 8

.3 1 0

.0 7 4

.4 2 2

.451

.......................................................

.1 6 7

.1 8 3

F re s h a n d f r o z e n c h ic k e n p a r t s .................................

.1 3 9

.1 4 7

H o m e p u rc h a s e

O t h e r p o u lt r y .......................................................................
F is h a n d s e a f o o d ..................................................................

.1 1 6

.121

.410

.429

.1 5 3

.1 6 4

.........................................................................

2 .1 2 7

P o u lt r y

........................................................................................

F re s h w h o l e c h ic k e n

C a n n e d f is h a n d s e a fo o d

.............................................
..........................

.2 5 7
.2 2 4

U n p r ic e d it e m s 1 ..........................................................................

.0 5 6

.1 9 9

.1 2 2

..........................................................................

2 2 .8 4 8

2 0 .5 6 4

........................................ .................................

9 .9 6 8

8 .7 5 3

F in a n c in g , ta x e s , a n d in s u r a n c e ......................................
P r o p e r t y in s u r a n c e ................................................................

9 .2 1 1
.5 7 9

8 .5 0 8
.501

H o m e o w n e r s h ip

P ro p e rty ta x e s

.2 6 5

...............................................................................................

.2 4 5

F re s h a n d f r o z e n f is h a n d s e a fo o d
E ggs

.9 1 5

.3 8 9

.1 7 8

.......................................................................

.8 4 3

1 .9 9 3

.................................................................................

................................................

C anned ham

1 .7 5 8

..........................................................................................

.................................................................................

H a m o th e r th a n c a n n e d
S ausage

Lunch

1 .8 6 2

.............................

6 .5 0 5

6 .1 4 5

M a in t e n a n c e a n d r e p a i r s .......................................................

3 .6 6 8

3 .3 0 3

2 .8 0 0

2 .3 2 2

C o n t r a c t e d m o r t g a g e in t e r e s t c o s t
M a in te n a n c e a n d r e p a ir s e r v ic e s

.................................

............................................................................

1 .6 5 4

1 .821

M a in t e n a n c e a n d r e p a ir c o m m o d itie s

F r e s h m i lk a n d c r e a m ...........................................................

.971

1 .1 0 0

P a in t a n d w a llp a p e r , s u p p lie s , to o ls ,

..................................................................

.7 0 6

.8 3 4

D a ir y p r o d u c t s

F re s h w h o l e m i lk

.8 6 8

.981

a n d e q u i p m e n t ...................................................................

.141

.1 5 6

...................

.0 8 3

.1 0 5

..............................................................

.0 5 2

.061

...........................................

.2 6 5

.2 6 7

L u m b e r , a w n in g s , g la s s , a n d m a s o n r y

..................................................

.6 8 3

.7 2 0

P lu m b in g , e le c t r ic a l, h e a t in g , a n d

..........................................................................; ..............

.0 8 0

.0 8 6

O t h e r fr e s h m i lk a n d c r e a m
P r o c e s s e d d a ir y p r o d u c t s
B u tte r

........................

C heese

c o o li n g s u p p lie s

.3 4 4

.3 5 6

M is c e lla n e o u s s u p p lie s a n d e q u ip m e n t

.................

.061

.0 8 0

.................................

.1 6 5

.1 8 0

U n p r ic e d it e m s 1 .....................................................................

.0 2 5

.0 2 5

.........................................................

.0 9 4

.0 9 8
................................................................

6 .5 1 0

6 .3 9 8

..............................................................

1 .7 5 9

1 .8 3 7

...................................................................................................

.8 9 9

.9 4 5

........................................................................................

Ic e c r e a m a n d re la te d p r o d u c t s
O t h e r d a ir y p r o d u c t s

F u e l a n d o t h e r u t i lit ie s
F r u it s a n d v e g e t a b le s

F re s h f r u i t s a n d v e g e ta b le s

.............................................

F u e ls

4 .2 8 3

4 .2 6 2

...........................................

.8 9 7

.8 9 2

..........................................................................................

.7 4 4

.7 4 2

.1 4 7

.1 4 6

F u e l o il, c o a l, a n d b o t t le d g a s

..............................................................................

.431

.4 4 4

.....................................................................................

.0 8 6

.0 9 6

B a n a n a s ...................................................................................

.0 5 0

.0 5 2

U n p r ic e d it e m s 1 .......................................................................

.0 0 6

.0 9 0

.091

G a s ( p ip e d ) a n d e l e c t r i c i t y ..................................................

3 .3 8 6

3 .3 7 0

.2 0 4

.2 0 6

2 .1 0 6

2 .0 7 9

F re s h f r u i t s
A p p le s
O ra n g e s

...................................................................................

O t h e r f r e s h f r u it s

................................................................

F u e l o il

O t h e r f u e ls

E le c t r ic it y

................................................................................

...................................................................................

.0 0 5

..................................................................

.4 6 8

.5 0 0

1 .2 8 0

1 .2 9 1

P o t a t o e s ...................................................................................

.101

.1 0 8

...................................

2 .2 2 7

2 .1 3 1

L e ttu c e

.....................................................................................

.0 9 2

.0 9 8

T e le p h o n e s e r v i c e s ..................................................................

1 .5 9 3

1 .5 3 1

T o m a t o e s .................................................................................

.071

.0 7 7

L o c a l c h a r g e s ............................................................................

.8 2 7

.7 9 7

.2 0 4

.2 1 7

In t e r s t a t e t o l l c a l l s ..................................................................

.4 1 4

.3 9 9

.8 6 0

.8 9 2

In t r a s t a t e t o l l c a l l s ..................................................................

F re s h v e g e t a b le s

O t h e r fr e s h v e g e t a b l e s ....................................................
P r o c e s s e d f r u i t s a n d v e g e ta b le s

....................................

U t i l i t y ( p ip e d ) g a s

..................................................................

O t h e r u t i lit ie s a n d p u b lic s e r v ic e s

.3 5 2

.3 3 6

P r o c e s s e d f r u i t s .....................................................................

.4 1 9

.4 0 9

W a te r a n d s e w e r a g e m a in t e n a n c e

.................................

.4 4 2

.4 2 2

F r o z e n f r u i t a n d f r u i t j u i c e s ..........................................

.121

.1 1 5

U n p r ic e d it e m s 1 ..........................................................................

.0 0 9

.0 0 7

F r u it ju ic e s o t h e r t h a n f r o z e n ......................................

.1 5 4

.1 5 5

C a n n e d a n d d r ie d f r u i t s ..................................................

.1 4 4

.1 4 0

.......................................................

.441

.4 8 3

..............................................................

.1 1 3

.1 1 6

.1 0 8

.1 2 5

.2 2 0

.2 4 2

P r o c e s s e d v e g e ta b le s
F r o z e n v e g e t a b le s

C u t c o r n a n d c a n n e d b e a n s e x c e p t lim a

............

O t h e r c a n n e d a n d d r ie d v e g e t a b le s ........................

8 .2 1 5

7 .9 1 2

4 .6 0 3

4 .7 3 5

.5 6 6

.5 4 7

.2 9 2

.2 9 6

H o u s e f u r n is h in g s
H o u s e h o ld lin e n s

S ee f o o t n o t e s a t e n d o f t a b le .




...............................

.......................................................................

T e x t i le h o u s e f u r n i s h i n g s .......................................................

H o u s e h o ld f u r n i s h in g s a n d o p e r a t io n
1

2 8

..................................................................

Appendix 0. Relative importance of components in the Consumer Price Indexes: U.S. city average, December 1977—Continued
(Percen t o f all item s)
U rb a n W a g e

U rb a n W age

E a rn e rs a n d

E a rn e rs a n d

A ll U r b a n

C o n s u m e rs

C le r ic a l

C o n s u m e rs

C le r ic a l

( C P I- U )

W o rk e rs

(C P I-U )

W o rk e rs

A ll U r b a n
G r o u p a n d ite m

G r o u p a n d ite m

( C P I- W )

( C P I- W )

E x p e n d it u r e c a te g o r y

E x p e n d it u r e c a t e g o r y

A p p a r e l a n d u p k e e p — C o n t in u e d

H o u s i n g — C o n t in u e d

B o y s ’ a n d g ir ls ’ ............................................................................

C u r t a in s , d r a p e s , s lip c o v e r s , a n d
................................................................

0 .2 7 3

0 .2 5 0

U n p r ic e d it e m s 1 .......................................................................

.0 0 2

.0 0 2

1 .3 2 4

1 .3 3 2

s e w in g m a t e r ia ls

F u r n it u r e a n d b e d d i n g

............................................................

W o m e n ’s

........................................................................................

A p p a r e l s e r v ic e s

..........................................................................

0 .1 8 8
.2 9 6

0 .2 1 7
.2 8 8

.6 6 3

.6 3 6

.421

.3 8 8

L a u n d r y a n d d r y c le a n in g o t h e r th a n

.4 0 3

................................................................

.3 8 4

S o f a s ...............................................................................................

.2 6 8

.2 7 5

O t h e r a p p a r e l s e r v i c e s ............................................................

L iv in g r o o m c h a ir s a n d t a b le s

.........................................

.2 8 8

.2 9 5

U n p r ic e d it e m s 1 ..........................................................................

O t h e r f u r n i t u r e ..........................................................................
A p p lia n c e s i n c lu d in g T V a n d s o u n d e q u ip m e n t . . .

.3 8 3

.3 5 8

1 .7 2 4

1 .8 9 4

.................................

.8 4 7

.9 1 9

................................................................................

.3 9 7

.4 1 9

N e w c a rs

.4 5 0

B e d ro o m fu r n itu r e

S o u n d e q u ip m e n t

..........................................................................

.2 4 2

.2 4 7

(2 )

(2 )

.................................................................................

1 8 .0 2 8

2 0 .2 3 4

...............................................................................................

1 6.9 31

1 9 .2 5 0

........................................................................................

4 .0 4 0

4 .2 7 5

.5 0 0

U s e d c a r s ........................................................................................

3 .0 2 0

3 .8 5 5

.8 7 7

.9 7 5

G a s o l i n e ...........................................................................................

4 .2 0 5

4 .7 8 6

....................................

.2 2 7

.2 5 4

A u t o m o b ile m a in t e n a n c e a n d r e p a ir .............................

1 .5 1 6

1 .6 6 4

...........................................................

.2 0 0

.231

B o d y w o rk

.2 0 6

.2 1 2

.4 4 0

.4 8 9

A u t o m o b ile d r iv e tr a in , b r a k e , a n d
m is c e lla n e o u s m e c h a n ic a l r e p a ir .............................

.3 3 2

.3 8 2

.2 3 5

.2 6 0

M a in t e n a n c e a n d s e r v i c i n g ................................................

.5 7 3

.621

P o w e r p la n t r e p a i r ...................................................................

.4 0 5

.4 4 9
4 .6 6 8

T e le v is io n a n d s o u n d e q u ip m e n t
T e le v is io n

c o in o p e r a te d

................................................................

H o u s e h o ld a p p lia n c e s

.........................................................

R e f r ig e r a t o r a n d h o m e f r e e z e r
L a u n d r y e q u ip m e n t

O t h e r h o u s e h o ld a p p lia n c e s

........................................

T r a n s p o r t a tio n
P riv a te

S to v e s , d is h w a s h e r s , v a c u u m s , a n d
s e w in g m a c h i n e s ...........................................................
O f f i c e m a c h in e s , s m a ll e le c t r ic

; .................................................................................

a p p lia n c e s , a n d a ir c o n d i t i o n e r s ..........................

.2 0 5

.2 2 9

O t h e r p r iv a t e t r a n s p o r t a t i o n ................................................

4 .1 5 0

U n p r ic e d it e m s 1 .....................................................................

.0 0 9

.001

O t h e r p r iv a t e t r a n s p o r t a t io n c o m m o d i t i e s ...............

.7 3 3

.8 1 5

...................

.091

.1 0 6

A u t o m o b ile p a r ts a n d e q u ip m e n t ...............................

.6 4 2

.7 0 8

.4 5 6

.5 0 4

O t h e r h o u s e h o ld e q u ip m e n t

.............................................

.9 8 8

.9 6 2

M o t o r o il, c o o la n t, a n d o t h e r p r o d u c t s

F lo o r a n d w in d o w c o v e r in g s , in f a n t s ’, la u n d r y ,
..........................

.1 9 8

.1 6 3

....................................

.1 7 2

.1 3 8

.3 0 5

.3 1 8

.2 0 7

.2 3 4

c le a n in g , a n d o u t d o o r e q u ip m e n t
C lo c k s , la m p s , a n d d e c o r it e m s

T ir e s

.1 8 6

.2 0 4

........................

3 .4 1 6

3 .8 5 4

A u t o m o b ile i n s u r a n c e ........................................... ...........

1 .9 5 5

2 .1 6 2

.7 6 7

.9 9 3
.6 9 9

O t h e r p a r ts a n d e q u ip m e n t

.........................................

O t h e r p r iv a te t r a n s p o r t a t io n s e r v ic e s

T a b le w a r e , s e r v in g p ie c e s , a n d
n o n e l e c t r i c k it c h e n w a r e

..........................................................................................

................................................

A u t o m o b il e f in a n c e c h a r g e s

L a w n e q u ip m e n t , p o w e r t o o ls , a n d
o t h e r h a r d w a r e .....................................................................

.........................................

A u t o m o b ile r e n ta l, r e g is t r a t io n , a n d

.1 0 7

.1 1 0

o t h e r f e e s ...............................................................................

.6 9 4

H o u s e k e e p in g s u p p lie s

............................................................

1 .5 5 9

1 .6 1 6

S ta te r e g i s t r a t i o n ................................................................

.3 1 9

.3 5 6

S o a p s a n d d e te rg e n ts

............................................................

.3 2 2

.3 7 0

.0 2 9

.0 3 3

.2 5 3

.2 8 0

D r iv e r s ’ l i c e n s e .....................................................................
A u t o m o b ile in s p e c t i o n .....................................................

.0 2 0

.0 2 3

.2 4 9

.261

................................................................

1 .0 9 7

.9 8 5

.2 3 2

.2 1 7

A ir li n e f a r e .....................................................................................

.4 8 3

.3 3 0

.0 4 7

.0 3 5

U n p r ic e d it e m s 1 .......................................................................

O t h e r la u n d r y a n d c le a n in g p r o d u c t s

..........................

O t h e r a u t o m o b ile r e la te d fe e s ......................................

t o w e ls a n d n a p k in s ..............................................................

.2 5 3

.2 1 7

U n p r ic e d it e m s 1 .....................................................................

C le a n s in g a n d t o i l e t t is s u e , p a p e r

.0 4 4

.0 4 0

P u b lic t r a n s p o r t a t io n

S t a t io n e r y , s t a t io n e r y s u p p lie s , a n d
g i f t w r a p .....................................................................................
...............................

.271

.2 9 8

I n t e r c it y b u s fa r e

..................................................

.2 3 2

.191

I n t r a c it y m a s s t r a n s it

............................................................

2 .0 5 3

1 .5 6 0

..........................................................................................

.1 8 9

.1 6 8

M is c e lla n e o u s h o u s e h o ld p r o d u c t s
L a w n a n d g a r d e n s u p p lie s
H o u s e k e e p in g s e r v ic e s
P o s ta g e

.......................................................................
..............................................................

.4 3 5

.5 0 3

T a x i f a r e ..........................................................................................

.0 9 5

.0 8 2

.4 3 2

.........................................

.351

.3 1 3

.0 6 0

.011

.0 0 9

.0 2 6

.0 2 6

4 .9 6 9

4 .4 9 2

.8 5 9

.7 8 0

I n t e r c it y t r a in fa r e

.3 7 5

U n p r ic e d it e m s 1 ..........................................................................

.....................................................................

U n p r ic e d it e m s 1 ..........................................................................

M o v in g , s t o r a g e , f r e ig h t , h o u s e h o ld

.041

la u n d r y , a n d d r y c le a n in g s e r v i c e s .............................
A p p lia n c e a n d f u r n i t u r e r e p a ir

M e d ic a l c a r e

.....................................................................................

M e d ic a l c a r e c o m m o d i ti e s

.....................................................

..................................................................

P r e s c r ip t io n d r u g s

.391

.3 2 2

..............................................................

.081

.0 6 8

.0 6 5

.0 5 3

5 .8 0 0

5 .8 3 6

A n t i- in f e c t iv e d r u g s

..................................................................

5 .1 3 8

5 .2 0 1

T r a n q u iliz e r s a n d s e d a tiv e s

.............................................

A p p a r e l c o m m o d i t i e s le s s f o o t w e a r .................................

4 .4 2 2

4 .4 4 3

C ir c u la t o r ie s a n d d iu r e t ic s

................................................

M e n ’s a n d b o y s ’ ..........................................................................

1 .6 4 6

1 .6 4 4

H o r m o n e s , d ia b e tic d r u g s , b io lo g ic a ls ,

A p p a re l a n d u p k e e p

.................................................... ...................

A p p a r e l c o m m o d i ti e s

.............................................................................................

1 .3 1 7
.3 7 8
.131
.2 4 6

..........................................................................................

.2 2 8

.2 2 3

D u n g a r e e s , je a n s , a n d t r o u s e r s ....................................

.3 1 7

U n p r ic e d it e m s 1 .....................................................................

.0 1 8

.3 4 5
.0 2 4

B o y s ’ ...............................................................................................
C o a t s , ja c k e t s , s w e a te r s , a n d s h i r t s ..........................

.3 3 0

.3 8 7

.1 1 4

.1 3 3

.0 5 8

.0 6 6

.1 4 3

.1 7 3

F u r n is h i n g s

..............................................................................

S u its , t r o u s e r s , s p o r t c o a t s , a n d ja c k e t s

.................

.0 4 8

.0 6 6

.0 5 5

..........

.4 6 8

.4 5 8

E y e g la s s e s .................................................................................
I n te r n a l a n d r e s p ir a to r y o v e r - t h e - c o u n t e r

.1 0 5

.1 0 3

d r u g s ..........................................................................................

.2 6 0

.2 6 3

N o n p r e s c r ip t io n d r u g s a n d m e d ic a l s u p p lie s

N o n p r e s c r ip t io n m e d ic a l e q u ip m e n t a n d
...................................................................................

.1 0 3

.0 9 2

M e d ic a l c a r e s e r v i c e s ................................................................

4 .1 1 1

3 .7 1 2
1 .9 1 6
.9 7 4

s u p p lie s

.0 1 5

.0 1 4

P r o fe s s io n a l s e r v ic e s

..............................................................

2 .0 0 8

...................................................................

2 .0 4 4

2 .0 8 1

P h y s ic ia n s ’ s e r v ic e s

..............................................................

.9 9 0

.....................................................................................

1 .6 9 2

1 .6 8 3

D e n ta l s e r v i c e s ..........................................................................

.7 4 6

.6 9 9

.1 9 5

.2 0 9

O t h e r p r o fe s s io n a l s e r v ic e s

.............................................

.2 3 5

.2 2 0

U n p r ic e d it e m s 1 .......................................................................

.0 3 6

.0 2 2

O t h e r m e d ic a l c a r e s e r v i c e s ................................................

2 .1 0 3

1 .7 9 7
.3 0 7

U n p r ic e d it e m s 1 .....................................................................
W o m e n 's a n d g i r l s '
W o m e n ’s

.0 5 9

a n d p r e s c r ip tio n m e d ic a l s u p p lie s

.1 3 8
.2 4 8

S h irts

..........................

a n d r e s p ir a to r y a g e n t s ....................................................

.2 7 8

C o a t s a n d ja c k e t s ................................................................
F u r n is h i n g s a n d s p e c ia l c l o t h i n g ...............................

.0 4 5
.0 5 4

P a in a n d s y m p t o m c o n t r o l d r u g s ..................................
S u p p le m e n ts , c o u g h a n d c o ld p r e p a r a tio n s ,

1 .2 5 7

S u its , s p o r t c o a t s , a n d j a c k e t s ......................................

M e n ’s

.0 5 4
.0 6 5

C o a t s a n d ja c k e t s

................................................................

.3 7 8

.3 2 4

.............................................

.3 9 6

.4 1 0

U n d e r w e a r , n ig h t w e a r , a n d h o s i e r y ..........................

.4 0 6

.4 3 4

..........................

.3 5 5

S u its

.............................................................................................

.1 9 3

.1 8 5

H o s p ita l r o o m ..........................................................................

.1 6 2

.141

U n p r ic e d it e m s 1 .....................................................................

.1 2 4

.121

O t h e r h o s p it a l a n d m e d ic a l c a r e s e r v ic e s

............

.1 9 2

.1 6 4

G ir l s ’ ...............................................................................................

.3 5 2

.3 9 7

U n p r ic e d it e m s 1 .....................................................................

.0 0 2

.001

C o a ts , ja c k e t s , d r e s s e s , a n d s u i t s ...............................

.1 2 6

.1 4 6

S e p a r a te s a n d s p o r t s w e a r

.1 4 2

.1 5 6

E n t e r t a i n m e n t ...................................................................................

4 .0 8 6

3 .9 1 0

E n t e r t a in m e n t c o m m o d i t i e s ..................................................

2 .4 2 3

2 .4 9 7

D re s s e s

.....................................................................................

S e p a r a te s a n d s p o r t s w e a r

.............................................

H o s p ita l a n d o t h e r m e d ic a l s e r v ic e s

U n d e r w e a r , n ig h t w e a r , h o s ie r y ,
a n d a c c e s s o r i e s ................................................................
U n p r ic e d it e m s 1 .....................................................................
I n f a n t s ’ a n d t o d d le r s '

..............................................................

O th e r a p p a r e l c o m m o d itie s

................................................

S e w in g m a t e r ia l s a n d n o t io n s

.6 3 0

.5 6 3

.3 3 4

.3 2 8

.0 9 0

.0 0 6

.0 0 6

.1 2 7

.1 4 4

.............................

.2 9 5

.2 3 5

.6 0 4

.5 7 5

S p o r t in g g o o d s a n d e q u i p m e n t .........................................

.6 9 0

.7 1 5

.1 7 9

S p o r t v e h i c l e s ............................................................................

.411

.4 3 6

.1 7 9

R e a d in g m a te r ia ls
N e w s p a p e rs

M a g a z in e s , p e r io d ic a ls , a n d b o o k s

............................................................

.4 2 5

.3 9 6

I n d o o r a n d w a r m w e a t h e r s p o r t e q u i p m e n t ............

.0 8 4

.081

...........................................................................................

.7 1 6

.7 5 7

B ic y c le s

........................................................................................

.0 9 3

.0 9 7

...............................................................................................

.2 3 2

.2 5 2

O t h e r s p o r tin g g o o d s a n d e q u ip m e n t ........................

.0 8 6

.0 8 5

J e w e l r y a n d lu g g a g e
F o o tw e a r
M e n ’s

.........................................

.....................................................................

..............................................................................

.0 7 8

S e e f o o t n o t e s a t e n d o f t a b le .




29

Appendix B. Relative importance of components in the Consumer Price Indexes: U.S. city average, December 1977— Continued
(Percent of all items)
U rb a n W age
A ll U r b a n
C o n s u m e rs

C le r i c a l

(C P I-U )

G r o u p a n d ite m

U rb a n W a g e
A ll U r b a n

E a rn e rs a n d
W o rk e rs

G r o u p a n d it e m

E a rn e rs a n d

C o n s u m e rs

C le r i c a l

(C P I-U )

W o rk e rs

(C P I-W )

(C P I-W )

E x p e n d it u r e c a te g o r y
C o m m o d i t i e s — C o n t in u e d

E n t e r t a i n m e n t — C o n t in u e d
U n p r ic e d it e m s 1 .................................................

F o o d a n d b e v e r a g e s ...................................................................

' 1 8 .8 1 4

2 0 .4 8 1

C o m m o d it ie s le s s f o o d a n d b e v e r a g e s

..........................

4 0 .4 9 5

4 1 .6 7 9

N o n d u r a b le s le s s f o o d a n d b e v e r a g e s

........................

1 7 .2 3 1

1 8 .2 0 2

............................................................

5 .1 3 8

5 .2 0 1

1 2 .0 9 3

1 3 .0 0 1

........................................................................................

2 3 .2 6 4

2 3 .4 7 7

...............................................................................................

4 0 .6 9 0

3 7 .8 4 0

.0 1 5

.0 1 7

1 ,1 0 4

1 ,2 1 9

.

.5 4 9

.6 2 4

P h o t o g r a p h ic s u p p lie s a n d e q u i p m e n t '

.2 2 0

.211

A p p a r e l c o m m o d itie s

P e t s u p p lie s a n d e x p e n s e

..........................

.3 0 5

.3 4 9

N o n d u r a b le s le s s f o o d , b e v e ra g e s , a n d

U n p r ic e d it e m s 1 ................................................

.0 3 0

.0 3 4

1 .6 6 2

1 .4 1 3

.5 0 3

.4 1 6

T o y s , h o b b ie s , a n d o t h e r e n t e r t a in m e n t
T o y s , h o b b ie s , a n d m u s ic e q u ip m e n t

E n t e r t a i n m e n t s e r v ic e s

....................................

F e e s f o r p a r t i c ip a n t s p o r ts
A d m is s io n s

..........................

...........................................................

.2 8 5

.271

O t h e r e n t e r t a i n m e n t s e r v i c e s .....................

.2 0 8

.1 7 8

a p p a re l
D u r a b le s
S e r v ic e s

......................................................................................

R e n t, r e s i d e n t i a l ............................................................................

5 .6 2 4

5 .3 2 2

H o u s e h o ld s e r v ic e s le s s r e n t ................................................

2 0 .3 8 9

1 8 .3 7 9

O t h e r g o o d s a n d s e r v i c e s ................................................

4 .3 9 5

4 .3 6 7

T r a n s p o r t a tio n s e r v i c e s ............................................................

6 )0 2 9

T o b a c c o p r o d u c t s ..............................................................

1 .2 0 2

1 .4 5 4

M e d ic a l c a r e s e r v i c e s ................................................................

4 .1 1 1

3 .7 1 2

1 .0 8 9

1 .341

O t h e r s e r v ic e s

4 .5 3 7

3 .9 2 4

C ig a r e t t e s

............ ...............................................................

...............................................................................

6 .5 0 3

O t h e r t o b a c c o p r o d u c t s a n d s m o k in g
a c c e s s o r ie s

.1 1 2

.1 1 3

1 .7 5 2

1 .8 1 3

...

.791

.871

A ll ite m s le s s s h e l t e r .....................................................................

7 0 .8 1 7

7 3 .6 2 6

..

.211

.2 4 3

A ll it e m s le s s m o r t g a g e in t e r e s t c o s t s ...............................

9 3 .4 9 5

9 3 .8 5 5

D e n t a l a n d s h a v in g p r o d u c t s .................................

.1 5 6

.1 6 2

A ll ite m s le s s m e d ic a l c a r e

.......................................................

9 5 .0 3 1

9 5 .5 0 8

..............................................................

4 1 .5 9 1

4 2 .8 6 2

N o n d u r a b le s le s s f o o d ................................................................

1 8 .3 2 7

1 9 .3 8 5

1 3 .1 8 9

P e rs o n a l c a re

.....................................................................
.......................................................................

T o ile t g o o d s ' a n d p e r s o n a l c a r e a p p lia n c e s
P r o d u c t s f o r t h e h a ir , h a ir p ie c e s , a n d . w ig s

I

S p e c ia l g r o u p s ;
A ll ite m s le s s f o o d

..........................................................................

8 2 .2 8 1

8 0 .7 0 2

C o s m e t ic s , b a t h a n d n a il p r e p a r a tio n s ,
m a n ic u r e a n d e y e m a k e u p im p le m e n t s

...

.2 4 8

.271

O t h e r t o i l e t g o o d s a n d s m a ll p e r s o n a l
c a r e a p p lia n c e s

.........................................................

P e r s o n a l c a r e s e r v i c e s .........................................y .
B e a u ty p a r lo r s e r v ic e s f o r f e m a l e s ............ . / . .
.......................................................................

U n p r ic e d it e m s 1 ..............................................................
P e r s o n a l a n d e d u c a t io n a l e x p e n s e s
S c h o o l b o o k s a n d s u p p lie s

......................

.1 9 5

N o n d u r a b le s le s s f o o d a n d a p p a r e l

.961

.9 4 2

N o n d u r a b le s .....................................................................................

3 6 .0 4 5

3 8 .6 8 3

S e r v ic e s le s s r e n t

3 5 .0 6 6

3 2 .5 1 8

3 6 .5 8 0

3 4 .1 2 8

9 .9 0 5

1 0 .8 9 6

.6 5 2

.611

.3 0 8
.001

(2 )

1 441

P e rs o n a l e x p e n s e s

1.4 2 1

1 .4 7 8

2 .3 3 1

2 .5 9 8

In s u r a n c e a n d f in a n c e

.1 4 7

...

4 .2 9 6

4 .8 9 3

1 2 .7 0 2

1 2 .4 1 8

6 .7 1 0

6 .4 8 6

4 .8 5 3

.2 4 0

3 .8 8 2

------

.2 1 0
..................................................................................................

A ll ite m s le s s e n e r g y

...................................................................

A ll ite m s le s s f o o d a n d e n e r g y
1 0 0 .0 0 0
5 9 .3 1 0

...........................................

C o m m o d it ie s le s s f o o d a n d e n e r g y

1 0 0 .0 0 0
6 2 .1 6 0

...............................

8 .5 7 9

9 .1 5 5

9 1 .4 2 1

9 0 .8 4 5

7 3 .7 0 2

7 1 .5 4 7

3 6 .3 9 7

3 7 .0 7 7

..............................................................

5 .1 9 4

5 .7 8 5

S e r v ic e s le s s e n e r g y ................................................................

3 7 .3 0 5

3 4 .4 7 0

E n e rg y c o m m o d itie s

1 N o t a c t u a lly p r ic e d ; im p u t e d f r o m p r ic e d ite m s .
2 L e s s t h a n 0 .0 0 1 p e r c e n t .




......................................

H o u s e k e e p in g a n d h o m e m a in t e n a n c e s e r v ic e s

E n e rg y

C o m m o d it ie s 1
2

..................................

................................................................

U t ili t i e s a n d p u b lic t r a n s p o r t a t io n
.171

C o m m o d it y a n d s e r v ic e g r o u p
A ll i t e m s ..........

......................................

G a s o lin e , m o t o r o il, c o o la n t, a n d o t h e r p r o d u c t s

.................

.......................................................

D o m e s t ic a lly p r o d u c e d f a r m f o o d
I m p o r te d f o o d a n d f is h e r y p r o d u c t s

...............................

E le m e n t a r y a n d h ig h s c h o o l t u i t i o n

.......................................................

S e le c te d b e e f c u t s ..........................................................................

..............................................................

U n p r ic e d it e m s 1 ............................................................

..........................................................................

S e r v ic e s le s s m e d ic a l c a r e

P e r s o n a l a n d e d u c a t io n a l s e r v i c e s ........................
C o lle g e t u i t i o n

....................................

.331

......................................

T u it io n a n d o t h e r s c h o o l fe e s

1 4 .1 8 4

.1 7 7

H a ir c u t s a n d o t h e r b a r b e r s h o p s e r v ic e s ^
f o r m a le s

C o m m o d it ie s le s s f o o d

30

A pp en d ix ©= N o n -P O P S IM@
ini=[n]©y©®[hi©l(ol
S am ple Ddiigm s

4. Specific nursing home service.

For each Non-pops entry level item (electricity, for
example), the following information is given below:

5. Outlets, 34; quotes, 34.

1. S ource o f the universe d ata,
2. S am pling units for outlets,

Lodging away from home

3. M easure o f size,
4. D esired final pricing unit,

1.

5.

2. For a given sample area, facilities in each of the
remaining 84 sample areas are eligible for selec­
tion.

N um ber o f designated outlets an d designated quotes.

Electricity

bls

Unemployment Insurance file.

3. Number of employees multiplied by number of
months per year for which the facility is open, by
the inverse of the probability of selecting the
sample area.

1. a. National Electric Rate Book.
b. Statistics o f Privately Owned Electric Utilities in
the United States 1972, 1973.
c. Statistics o f Publicly Owned Electric Utilities in
the United States, 1972,1973.

4. Specific accommodation.
5. Outlets, 298; quotes, 298.

All three of the above were provided by the Federal
Power Commission.

Sntracity transit—
-bus, subway, streetcar, ©ommyttgr
railroads

2. Electric utility companies serving the 85 cpi sample
areas.

1.

3. Annual revenue from sales of electricity to residents
of the respective sample areas.

Unemployment Insurance file cross-referenced
with data from the American Transit Association
and the American Association of Railroads.

bls

4. Specific type of service for a specific number of kilo­
watt hours.

2. Establishments within 1 of the 85 sample areas.

5. Outlets, 415; quotes, 1,245.

3. Total employment.
4. Specific trip.

MaturaS ga§ (piptd)

5. Outlets, 238; quotes, 238.

1. Brown’ Directory o f North American Gas Com­
s
panies.

insurant©—auto, GioygelhoSd, flr©5and personal
property

2. Gas (or electricity and gas) companies serving the 85
cpi

sample areas.

3. Annual revenue from sales of natural gas to residents
of the respective sample areas.

1. 1978 Best Executive Data Service file of insurance
companies by State.

4. Specific type of service and specific number of cubic
feet or therms of gas.

2. Insurance companies writing insurance in the States
in which the respective sample areas are located.

5. Outlets, 238; quotes, 600.

3. Total revenue for noncommercial policies by type
of insurance.

Convalescent H d myrgimg hom© ©
in i
sir©

4. Specific policy within the sample area.
5. Auto: Outlets, 298; quotes, 596.
Homeowners: Outlets, 200; quotes, 400.
Fire and extended coverage: Outlets, 38; quotes, 76.
Personal property: Outlets, 238; quotes, 238.

1. 1977 Health, Education, and Welfare Census o f
Nursing and Related Care Facilities.
2. Individual facilities in 85

cpi

sample areas.

3. Number of beds.




31

0®ll®g© tuition, tsoBfeg© books, to y sin g wlhol© sit
sohooi

c. 1973 Highway statistics, by
interviews.

fha

and telephone

1. U.S. Department of Education tape file of all uni­
versities, colleges, junior colleges, and profes­
sional and technical schools.

2. For a given p s u , all facilities in the Nation have a
chance of selection.

2. For a given sample area, all schools in the Nation
are eligible for selection.

4. Specific toll fee for specified use of the facility.

3. Total toll revenue.
5. Outlets, 60; quotes, 60.

3. Estimated total full-time-equivalent enrollment by
residents of each sample area at each institution.

P©stagi©

4. Specific fee for the college.

1. The distribution of household mail by type of postal
service, by postal zones as determined by the
Household Mailstream Study, Final Report,
prepared by the Survey Research Center, Institute
for Social Research, University of Michigan,
under contract to the U.S. Postal Service.

5. Tuition: Outlets, 238; quotes, 476.
Books: Outlets, 238; quotes, 952
Housing: Outlets, 238; quotes, 476.
Elem entary and s®o©ndary §©lh®©i tuito©n

2. U.S. Postal Service.

1. Curriculum Information Center, Inc., data file of all
private elementary and secondary schools in the
United States.

3. Postal revenue.
4. Specific postal service and postal zones traveled.

2. Schools in 85 sample areas.

5. Outlets, 1; quotes, 238.

3. Enrollment of each school.
Magazines

4. Specific tuition fee.
5. Outlets, 181; quotes, 181.

1. 1975 Ayer Directory o f Publications.
2. Selection of magazine names.

Airlin© fares

3. Monthly circulation.

1. Civil Aeronautics Board data file consisting of a 10percent sample of all passenger itineraries origi­
nating in the United States.

4. Single copy or subscription plan.
5. Outlets, 125; quotes, 144.

2. All airlines providing service from the 85 cpi sample
areas.

lnt©reity by®
1. Russell’ Official National Motor Coach Guide.
s

3. Number of nonbusiness passengers per airline, per
trip itinerary, per fare class.

2.

4. A specific trip itinerary and fare class for the selected
airline.

3. Number of trips per week.
4. Specific trip (origin and destination) and class of
service.

5. Outlets, 476; quotes, 476.

5. Outlets, 355; quotes, 355.

Tsini far©§
1.

b ls

Intercity train

Unemployment Insurance file.

2. Taxi companies in the sample

Bus companies serving a given sample area.

psu

1. Federal Railroad Administration (
Matrix System City Table Report.

.

3. Total employment.

f r a

)

Am trak

Amtrak.

4. Specific taxi ride.

2.

5. Outlets, 238; quotes, 238.

3. Number of tickets sold.
4. Specific trip and class.

T®Sls

5. Outlets, 1; quotes, 476.

1. a. Toll Facilities in the JJ.S., the Federal Highway
Administration.
b. Toll Rates—U.S. Toll Roads and Toll Rates—
Bridges and Tunnels, the International Bridge,
Tunnel, and Turnpike Association.



Telephone main station s©rwic@s
1. Telephony’ Directory o f Telephone Industry, b l s
s
Unemployment Insurance file, and Bell com­
panies.

32

State vehicle registration, driver’s license, and State
vehicle inspection

2. Rate group and exchanges serving the 85 psu ’s within
a company.
3. Total revenue for each rate group and exchange or
number of residential customers.

1. Survey identifying the various State departments of
motor vehicles.

4. Specific service such as main station costs, addi­
tional message units, extension costs, etc.

2. State motor vehicle departments.
3. Revenue generated by the fee.

5. Outlets, 126; quotes, 985.

4. Specific class/vehicle registration, type of license,
or inspection service.

Intrastate l©ng°dostane@ telephone
1. Special

at&
t

5. Vehicle inspection: Outlets, 41; quotes, 117.
Driver’s license: Outlets, 41;-quotes, 117.
Vehicle registration: Outlets, 41; quotes, 117.

tabulations.

2. Type of call and rate period originating from

ps u .

3. Revenue.

Books noli from book clubs

4. Specification of a particular telephone call.

1. New York Times list of paperback bestsellers and a
universe of 22 publishers.

5. Outlets, 60; quotes 484.

2. F our samples of three publishers plus the New York
Times list each. Since the sampling frame is arbi­
trarily limited to 22 publishers, this component
does not have a strict probability sample.

Interstate long-dlstane© telephone
1.

a t &t

Long Lines special

bls

summary data file.

2. Type of call, rate period, and mileage band from 85
PSU’s.

3. Each publisher’s market share of books.

3. Total revenue for specific call cells.

4. Mean price derived from the mean prices of the four
subsamples.

4. Special phone call defined by type of call, time of
placement, length of haul and length of conversa­
tion, and specific location of exchange.

5. Outlets, 9; quotes, 9.

5. Outlets, 60; quotes, 484.




33

Appendix D. Pricing cycles for sample areas
Percent
of index
population1

Sample area

PSU

Pricing cycle
1
(odd
months)

2
(even
months)

Northeast Region:

9.6670

X

X

A 102
A 103
A104
A 105
A 106

New York City (Bronx, Kings, New York, Queens, Richmond)
\
New York suburban (Nassau, Putnam, Rockland, Suffolk, Westchester) 1 ..................
New Jersey suburban (Bergen, Essex, Hudson, Middlesex, Morris,
Passaic, Somerset, Union)
Philadelphia, Pennsylvania—New Jersey, S M S A .............................................................
Boston, Massachusetts, SMSA ...........................................................................................
Pittsburgh, Pennsylvania, S M S A ........................................................................................
Buffalo, New York, SMSA .................................................................................................
Northeast Pennsylvania SMSA ...........................................................................................

2.8556
1.7258
1.4231
.7966
.3675

X
X

X

B102
B104
B106
B108

Providence-Warwick-Pawtucket, Rhode Island—Massachusetts, SMSA ....................
Rochester, New York, SMSA .............................................................................................
Allentown-Bethlehem-Easton, Pennsylvania—New Jersey, S M S A ................................
Springfield-Chicopee-Holyoke, Massachusetts—Connecticut, SMSA .........................

1.0495
.9868
.9565
1.0239

C101
Cl 04
Cl 06
C108

Norwalk, Connecticut ..........................................................................................................
Binghamton, New York—Pennsylvania, S M S A ...............................................................
Portland, Maine ...................................................................................................................
Johnstown, Pennsylvania, SMSA ......................................................................................

.9362
.9325
.9561
.9517

D101

Centerville, East Falmouth, Falmouth Center, Harwich Center, Hyannis, Otis,
Provincetown Center, South Yarmouth, West Yarmouth, Massachusetts ................
Ansonia city, Seymour town, Connecticut .........................................................................
Canton village, Gouverneur village, Massena village, Ogdensburg city,
Potsdam village, New York .............................................................................................
Ellport borough, Ellwood City borough, New Castle city, New Wilmington
borough, Oakwood, other urbanized areas, Pennsylvania ...........................................

A101
A110
A 111

D103
D106
D107

X
X
X
X
X
X
X
X
X
X
X

.4436
.4433

X

.4436

X

X

X

.4435

North Central Region:
X
X
X

A207
A208
A209
A210
A211
A212
A213
A214

Chicago, Illinois, S C A ..........................................................................................................
Detroit, Michigan, SMSA ....................................................................................................
St. Louis, Missouri—Illinois, SMSA .................................................................................
Cleveland, Ohio, SMSA ......................................................................................................
Minneapolis-St. Paul, Minnesota—Wisconsin, S M S A ....................................................
Milwaukee, Wisconsin, S M S A .............................................................................................
Cincinnati, Ohio—Kentucky—Indiana, SMSA ...............................................................
Kansas City, Missouri—Kansas, S M S A .............................................................................

.5146
.6264
.4148
.2246
.1418
.8280
.8120
.7447

B210
B211
B213
B215

Columbus, Ohio, SMSA ......................................................................................................
Grand Rapids, Michigan, S M S A ........................................................................................
Indianapolis, Indiana, SMSA .............................................................................................
Toledo, Ohio—Michigan, S M S A ........................................................................................

1.1263
1.1221
.6412
1.1533

C210
C211
C214
C215
C218
C220

Racine, Wisconsin, S M S A ....................................................................................................
Saginaw, Michigan, SMSA .................................................................................................
Davenport Rock Island-Moline, Iowa—Illinois, SMSA ................................................
Canton, Ohio, SMSA ..........................................................................................................
Decatur, Illinois, SMSA ......................................................................................................
Terre Haute, Indiana, S M S A ...............................................................................................

.9011
.9126
.9089
.9050
.9110
.9043

D210

Mexico city, Vandalia city, Troy city, Bowling Green city, Louisiana city,
Hannibal city, Missouri ....................................................................................................
Grand Island city, Aurora city, Central City city, Nebraska ...........................................
Detroit Lakes city, Park Rapids village, Wadena village, M in n e so ta.............................
Webster City city, Fort Dodge city, I o w a ...........................................................................

.0700
.0698
1.0698
1.0700

X
X

X

X

D211
D214
D216

X
X
X
X
X
X
X
X
X
X
X
X
X
X
X

X
X
X

South Region:
A315
A316
A317
A318
A319
A320

Washington, D.C.—Maryland—Virginia, S M S A .............................................................
Dallas-Fort Worth, Texas, S M S A ......................................................................................
Baltimore, Maryland, S M S A ...............................................................................................
Houston, Texas, S M S A ........................................................................................................
Atlanta, Georgia, SMSA ......................................................................................................
Miami, Florida, SMSA ........................................................................................................

1.7253
1.4005
1.2226
1.1831
.9440
.7546

B318
B319
B321
B324
B325
B328
B329
B332

Tampa St. Petersburg, Florida, SMSA .............................................................................
Raleigh-Durham, North Carolina, SMSA ........................................................................
New Orleans, Louisiana, SMSA ........................................................................................
Richmond, Virginia, S M S A .................................................................................................
San Antonio, Texas, SMSA ...............................................................................................
Nashville-Davidson, Tennessee, SMSA .............................................................................
Louisville, Kentucky—Indiana, SMSA .............................................................................
Memphis, Tennessee—Arkansas—Mississippi, SMSA ....................................................

1.0122
.9759
.9906
.9966
.9336
1.0069
.9625
1.0035




34

X
X
X
X
X
X
X
X
X
X
X
X
X

Appendix D. Pricing cycles for sample areas - Continued
Percent
of index
population1

Sample area

PSU

C321
C323
C326
C327
C330
C332
C333
C336

Huntsville, Alabama, SMSA ...............................................................................................
West Palm Beach-Boca Raton, Florida, SMSA ...............................................................
Albany, Georgia, SMSA ......................................................................................................
Baton Rouge, Louisiana, SMSA ............. ..........................................................................
Pine Bluff, Arkansas, SMSA ...............................................................................................
Corpus Christi, Texas, SMSA ............................................................................................
Huntington-Ashland, West Virginia—Kentucky—Ohio, S M S A ....................................
Brownsville-Harlingen-San Benito, Texas, S M S A ...........................................................

D317

D’Iberville, Moss Point city, Ocean Springs city, Pascagoula city, other
urbanized areas, Mississippi .............................................................................................
Beaufort city, Capehart, Parris Island, Port Royal town, South Carolina ....................
Smithfield town (part), Suffolk city, Virginia ....................................................................
Wadesboro town, East Rockingham city, Hamlet town, Rockingham city,
North Carolina .................................................................................................................

D319
D321
D323

0.9881
.9868
.9878
.9863
.9891
.9857
.9883
.9872

1.2665
1.2669
1.2669

Pricing cycle
1
(odd
months)

2
(even
months)

X
X
X
X
X
X
X
X

X
X
X
X

1.2669

West Region:

A422
A423
A424
A425
A426
A427

Los Angeles, California, SMSA; Anaheim-Santa Ana-Garden Grove,
California, SMSA ............................................................................................................
San Francisco-Oakland, California, SMSA ......................................................................
Seattle-Everett, Washington, SMSA .................................................................................
San Diego, California, SMSA .............................................................................................
Portland, Oregon—Washington, SMSA ..........................................................................
Honolulu, Hawaii, SMSA ...................................................................................................
Anchorage, Alaska, SMSA .................................................................................................

5.0340
1.8679
.8416
.8058
.5861
.3740
.0741

B433
B434
B436
B440

Denver-Boulder, Colorado, SMSA ....................................................................................
San Jose, California, SMSA ...............................................................................................
Riverside-San Bernardino-Ontario, California, SMSA ..................................................
Fresno, California, S M S A ...................................................................................................

.7276
1.2055
1.1468
.8906

C438
C440
C441
C443

Colorado Springs, Colorado, SMSA .................................................................................
Tuscon, Arizona, SMSA ......................................................................................................
Salinas-Seaside-Monterey, California, S M S A .................................................................
Bakersfield, California, S M S A .............................................................................................

.7543
.7067
.7620
.7193

D425
D427
D429
D431

Corvallis city, Lincoln City city, Newport city, Toledo city, Oregon .............................
Alamogordo city, Holloman, Tularosa village, New M e x ic o ...........................................
Logan city, Smithfield city, Utah ........................................................................................
Dillon city, Anaconda city, Butte city, Floral Park, Silver Bow Park, Montana .........

.4707
.4932
.4950
.4913

A421

1The percent of population for the all urban consumer index based on the 1970
census. For the A PSU’s, B213, and B433, the percentage reflects only the popula-




X

X
X

X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X

tion of the sample areas. The remaining areas were noncertainty selections, and
the percentage reflects the population of the strata represented by the sample area.

35

Appoinidlix E„ S@as®inal Adjystmsinit
IMIetlh®dl®D®gy at B IS

An economic time series may be affected by regular
intrayearly (seasonal) movements which result from
climatic conditions, model changeovers, vacation prac­
tices, holidays, and similar factors. Often such effects
are large enough to mask the short-term, underlying
movements of the series. If the effect of such intrayearly
repetitive movements can be isolated and removed, the
evaluation of a series may be made more perceptive.
Seasonal movements are found in almost all
economic time series. They may be regular, yet they do
show variation from year to year and are subject to
changes in pattern over time. Because these intrayearly
patterns are combined with the underlying growth or
decline and cyclical movements of the series (trendcycle) and also random irregularities, it is difficult to
estimate the pattern with exactness.
More than a half-century ago, attempts were made to
isolate seasonal factors from time series. Some early
methods depended upon smoothing curves by using per­
sonal judgment. Other formal approaches were
periodogram analysis, regression analysis, and correla­
tion analysis. Because these methods involved a large
amount of work, relatively little application of seasonal
factor adjustment procedures was carried out.
In the mid-1950’s, new electronic equipment made
more elaborate approaches feasible in seasonal factor
methods as well as in other areas. Using a computer, the
Bureau of the Census developed seasonal factors based
on a ratio-to-moving-average approach. This was a ma­
jor forward step, as it made possible the uniform ap­
plication of a method to a large number of series at a
relatively low cost.1 Subsequent improvements in
methods and in computer technology have led to more
refined procedures which are both faster and cheaper
than the original technique.
The Bureau of Labor Statistics began work on
seasonal factor methods in 1959. Prior to that time,
when additional data became available and seasonal
factors were generated from the lengthened series, the
new factors sometimes differed markedly from the cor­
responding factors based on the shorter series. This dif­

ference could affect any portion of the series. It was dif­
ficult to accept a process by which the addition of recent
information could affect significantly the seasonal fac­
tors for periods as much as 15 years earlier, especially
since this meant that factors could never become final.
The first b l s method, introduced in 1960, had two
goals: First, to stabilize the seasonal factors for the
earlier part of the series; second, to minimize the revi­
sions in the factors for the recent period.
Since 1960, the Bureau has made numerous changes
and improvements in its techniques and in methods of
applying them. Thus far, all the changes have been
made within the scope of the ratio-to-moving-average or
difference-from-moving-average types of approaches.
The b l s 1960 method, entitled “ The b l s Seasonal Fac­
tor M ethod,” was further refined, with the final version
being produced in 1966. It was in continuous use for
many Bureau series (especially employment series based
on the establishment data) until 1980. In 1967, the
Bureau of the Census introduced “ The X -ll Variant of
the Census Method II Seasonal Adjustment Program ,”
better known as simply X -ll. The X -ll provided some
useful analytical measures along with many more op­
tions than the b l s method. Taking advantage of the
X - l l ’s additional flexibility, b l s began making increas­
ing use of the X-l 1 method in the early 1970’s, especial­
ly for seasonal adjustment of the labor force data based
on the household data. Later in the 1970’s, Statistics
Canada, the Canadian national statistical agency,
developed an extension of the X -ll called “ The X -ll
a r i m a seasonal adjustment method.” The X -ll a r i m a
provided the option of using a r i m a (Autoregressive In­
tegrated Moving Average) modeling and forecasting
techniques to extrapolate some extra data at the end of a
time series to be seasonally adjusted. The extrapolated
data help to alleviate the effects of the inherent limita­
tions of the moving average techniques at the ends of
series. After extensive testing and research showed that
use of X-l 1 a r i m a would help to further minimize revi­
sions in factors for recent periods, BLS began using the
X -ll a r i m a procedure in 1980 for most of its official
seasonal adjustment.
The standard practice at b l s for current seasonal ad­
justment of data as it is initially released is to use pro­
jected seasonal factors which are published ahead of

1 Julius Shiskin, Electronic Computers and Business Indicators,
Occasional Paper No. 57 (New York, National Bureau of Economic
Research, 1957).




36

time. The time series are generally run through the
seasonal adjustment program once a year to provide the
projected factors for the ensuing months and the revised
seasonally adjusted data for the recent history of the
series, usually the last 5 or 6 years. It has generally been
unnecessary to revise any further back in time because
the programs which have been used have all accomplish­
ed the objective of stabilizing the factors for the earlier
part of the series, and any further revisions would pro­
duce only trivial changes. For the projected factors, the
factors for the last complete year of actual data were
selected when the X -ll or the b l s method programs
were used. With the X -ll a r i m a procedure, the pro­
jected year-ahead factors produced by the program are
normally used. For the labor force data since 1980, only
the first 6 months of factors projected from the annual
run are used—a special midyear run of the program is
done, with up-to-date data included, to project the fac­
tors for the remaining 6 months of the year.
The alternative to the use of projected factors is con­
current adjustment where all data are run through the
seasonal adjustment program each month, and the cur­
rent observation participates in the calculation of the
current factor. Of course, the concurrent approach
precludes the prior publication of factors and requires
substantially more staff and computer resources to run,
monitor, and evaluate the seasonal adjustment process.
However, recent research has shown potentially signifi­
cant technical advantages in the area of minimization of
factor revisions that are possible with concurrent adjust­
ment. If future findings suggest the desirability of a
change to a concurrent procedure or to some other type
of methodology, such a change will be seriously con­
sidered in consultation with the Government’s working
group on statistics.

In applying any method of seasonal adjustment, the
user should be aware that the result of combining series
which have been adjusted separately will usually be a lit­
tle different from the direct adjustment of the combined
series. For example, the quotient of seasonally adjusted
unemployment divided by seasonally adjusted labor
force will not be quite the same as when the unemploy­
ment rate is adjusted directly. Similarly, the sum of
seasonally adjusted unemployment and seasonally ad­
justed employment will not quite match the directly ad­
justed labor force. Separate adjustment of components
and summing of them to the total usually provides series
that are easier to analyze; it is also generally preferable
in cases where the relative weights among components
with greatly different seasonal factors may shift radical­
ly. For other series, however, it may be better to adjust
the total directly if high irregularity among some of the
components makes a good adjustment of all com­
ponents difficult.
Finally, it is worth noting that the availability of a
fast, efficient procedure for making seasonal adjust­
ment computations can easily lead to the processing of
large numbers of series without allotting enough time to
review the results. No standard procedure can take the
place of careful review and evaluation by a skilled
analyst. A review of all results is strongly recommend­
ed. And it should also be remembered that, whenever
one applies seasonal factors and analyzes seasonally ad­
justed data, seasonal adjustment is a process which
estimates a set of not directly observable components
(seasonal, trend-cycle, irregular) from the observed
series and is, therefore, subject to error. Because of the
complex nature of methods such as X -ll a r i m a , the
precise statistical properties of these errors are not yet
known.

Teetoieai References
Organization for Economic Co-operation and Development.
Seasonal Adjustment on Electronic Computers. Paris,
1961.
The report and proceedings of an international con­
ference held in November 1960. Describes experience in
the United States, Canada, and several European coun­
ties. Includes theoretical sections relating to calendar
(trading day) variation and general properties of moving
averages.

Barton, H.C., Jr. “Adjustment for Seasonal Variation,’’
Federal Reserve Bulletin, June 1941.
The classic account of the frb ratio-to-moving aver­
age method, in which the analyst uses skilled judgment
to draw freehand curves at key stages of the pro­
cedure.
Dagum, Estela Bee. The X -ll a r i m a Seasonal Adjustment
Method. Ottawa, Statistics Canada, February 1980
(Statistics Canada Catalogue No. 12-564E).

Shiskin, Julius. Electronic Computers and Business In­
dicators, Occasional Paper No. 57. New York, National
Bureau of Economic Research, 1957. Also published in
Journal of Business, Vol. 30, October 1957.
Describes applications of the first widely used com­
puter program for making seasonal adjustments.

Macaulay, Frederick R. The Smoothing of Time Series, nber
No. 19. New York, National Bureau of Economic
Research, 1931.
An early discussion of moving averages and of the
criteria for choosing one average rather than another.




37

T e c h n ic a l References—Continued
Proceedings of a 1976 conference jointly sponsored
by the National Bureau of Economic Research and
the Bureau of the Census.

U.S. Department of Commerce, Bureau of the Census. The
X -ll Variant of the Census Method II Seasonal Adjust­
ment Program. Technical Paper No. 15, (1967 revi­
sion).

U.S. Department of Labor, Bureau of Labor Statistics. The
b l s Seasonal Factor Method, 1966.
U.S. Department of Commerce, Bureau of the Census. Sea­
sonal Analysis of Economic Time Series, Economic
Research Report, ER-1, issued December 1978.




U.S. Department of Labor, Bureau of Labor Statistics. Em­
ployment and Earnings, January 1980.

38

3 2356 00251305 7

Bureau © L©®@[r stau su cs
f?
^©glQouiil O ffices

R©gi©m 1
Suite 1603
John F. Kennedy Federal Building
Government Center
Boston, Mass. 02203
Phone: (617) 223-6761

Region 1
1
Suite 3400
1515 Broadway
New York, N.Y. 10036
Phone: (212) 944-3121

Region U
S
3535 Market Street
P.O. Box 13309
Philadelphia, Pa. 19101
Phone: (215) 596-1154



Region SV
1371 Peachtree Street, N.E.
Atlanta, Ga. 30367
Phone: (404) 881-4418

Rogoons ¥11 snd ¥018
911 Walnut Street
Kansas City, Mo. 64106
Phone: (816) 374-2481

Region ¥

R@gi®eia IX and X

9th Floor
Federal Office Building
230 S. Dearborn Street
Chicago, III. 60604
Phone: (312) 353-1880

Region W
S
Second Floor
Griffin Square Building
Dallas, Tex. 75202
Phone: (214) 767-6971

450 Golden Gate Avenue
Box 36017
San Francisco, Calif. 94102
Phone: (415) 556-4678