View original document

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

U. S. DEPARTMENT OF LABOR

BUREAU OF LABOR STATISTICS
ROYAL MEEKER, Commissioner

BULLETIN OF THE UNITED STATES\
BUREAU OF LABOR S T A T IS T IC S /
W H O L E S A L E

P R I C E S

(WHOLE 1 7 ^
‘ * * /NU M BER 1 / J

S E R I E S :

No .

INDEX NUMBERS OF WHOLESALE
PRICES IN THE UNITED STATES
AND




FOREIGN

COUNTRIES

JULY, 1915

WASHINGTON
GOVERNMENT PRINTING OFFICE
1915

3




CONTENTS.

Part I.—The making and using of index numbers:
Page.
The aim and scope of the present bulletin................................................................
5
I. The history of index numbers............................................................................ 5-8
II. The difficulties of measuring changes in the level of prices............................. 8-10
III. The characteristics of price fluctuations........................................................... 10-24
IY. Varieties of methods used in making index numbers.................................... 25-93
1. The relations between methods and uses............................................... 25-27
2. Collecting and publishing the original quotations............................... 27-30
3. Market prices, contract prices, and import-export values..................30-33
4. Relative versus actual prices................................................................. 34,35
5. Base periods............................................................................................... 36-44
6. The numbers and kinds of commodities included...............................44-71
7. Problems of weighting.............................................................................. 71-80
8. Averages and aggregates........................................................................... 80-93
V. A comparison of the leading American index numbers for the years 1890 to
1913.................................................................................................................. 93-112
1. Analysis of the similarities and differences.......................................94-109
2. Critical evaluation................................................................................ 109-112
VI. Conclusions........................................................................................................ 112-114
LIST OF CHARTS.
Chart 1.—Conspectus of yearly changes in prices, 1891-1893. (Based on
Table 2 .)..................................................................................................... (Facing) 15
Chart 2.—Distribution of 5,578 price variations (percentages of rise or fall from
prices of preceding year).......................................................................................... 20
Chart 3.—Distribution of the price variations of 241 commodities in 1913 (per­
centages of rise or fall in price).............................................................................. 22
Chart 4.—Index numbers made from the market prices and from the import
or export values of identical lists of commodities. England, 1871-1902.
(Based on Table 5.).................................................................................................. 33
Chart 5.—General-purpose index numbers including 25, 50, and 242 com­
modities. (Based on Table 8.).............................................................................. 50
Chart 6.—Index numbers of the prices of 20 raw materials and of 20 products
manufactured from them. (Based on Table 9.)................................................ 56
Chart 7.—Index numbers of the prices of wool, cotton, hides, wheat, and pig
iron in their raw, partially manufactured, and finished forms. (Based on
Table 9.)..................................................................................................................... 57
Chart 8.—Index numbers of the prices of 19 mineral products and of 18 farm
crops. (Based on Table 10.).................................................................................. 59
Chart 9.—Index numbers of the prices of manufactured goods used for family
consumption and for industrial purposes. (Based on Table 11.)................... 62
Chart 10.—Index numbers of the prices of 25 food products and of 25 miscel­
laneous commodities. (Based on Table 15.)....................................................... 69
Chart 11.—A comparison of medians and arithmetic means of 145 commod­
ities. (Based on Table 17)...................................................................................... 89
Chart 12.—Dun’s, Bradstreet’s, and the Bureau of Labor Statistics’ index
numbers reduced to a common basis. (Based on Table 18.)........................... 99



3

4

CONTENTS.

Part II.—Index numbers of wholesale prices in the United States and foreign
countries:
United States:
t
Page.
Index numbers of the United States Bureau of Labor Statistics............. 115-127
Index numbers of the United States Senate Committee on Finance....... 128-138
Index numbers of the Annalist............................................................... 138-140
Index numbers of Bradstreet’s................................................................. 141-148
Index numbers of Dun............................................................................. 148-152
Index numbers of Gibson........................................................................ 153-156
Australia:
Index numbers of the Commonwealth Bureau of Census and Statistics.. 157-166
Austria-Hungary:
Index numbers of Dr. Bela von Jankovich.............................................. 166-168
Index numbers of Mario Alberti............................................................. 168-172
Belgium:
Index numbers of Hector Denis............................................................. 172-175
Canada:
Index numbers of the Department of Labor........................................... 176-186
Denmark:
Index numbers of the State Statistical Bureau....................................... 186-188
France:
Index numbers of the Annuaire Statistlque de la France...................... 188-192
Index numbers of the Statistique Generale de la France........................ 192-195
Index numbers of La Reforme Economique........................................... 195-203
Index numbers of Emile Levasseur........................................................ 204-207
Germany:
Index numbers of the Imperial Statistical Office.................................... 208-218
Index numbers of the Jahrbucher fur Nationalokonomie und Statistik. 219-239
Index numbers of Otto Schmitz.............................................................. 240-250
Index numbers of Adolf Soetbeer............................................................ 250-255
Great Britain:
Index numbers of the Board of Trade..................................................... 255-261
Index numbers of the Economist............................................................ 261-269
Index numbers of Augustus Sauerbeck................................................... 269-276
India:
Index numbers (rupee prices) of Fred. J. Atkinson................................. 276-282
Italy:
Index numbers of the Annuairio Statistico Italiano............................... 282-284
Index numbers of Achille Necco............................................................. 285-288
Japan:
Index numbers of the Department of Agriculture and Commerce........... 288-292
Netherlands:
Index numbers of the Netherlands Statistical Office............................. 293,294
New Zealand:
Index numbers of James W. Mcllraith................................................... 295-300
Norway:
Index numbers of Einar Ruud............................................................... 300-304
Russia:
Index numbers of Ministry of Commerce and Industry, Petrograd....... 305-309
Spain:
Index numbers of Francisco Bernis......................................................... 309-313
Conference on index numbers of the International Institute of Statistics, Sep­
tember, 1911 ................................................................................................ 313-319
Select bibliography of additional index numbers......................................... 319-324




BULLETIN OF THE

U.S. BUREAU OF LABOR STATISTICS.
WHOLE NO. 173.

W A S H IN G T O N .

JULY, 1915.

INDEX NUMBERS OF WHOLESALE PRICES IN THE UNITED STATES
AND FOREIGN COUNTRIES.
PART I.—THE MAKING AND USING OF INDEX
NUMBERS.
B Y W ESLEY C. M ITCHELL

.1

THE AIM AND SCOPE OF THE PRESENT BULLETIN.

The aim of this bulletin is to make the index numbers of wholesale
prices currently published in the United States and foreign countries
more accessible, more intelligible, and therefore more useful.
To this end the leading series of index numbers, compiled by
official bureaus, financial journals, and private investigators both at
home and abroad are described in detail. The history of each series,
the source from which quotations are taken, the number and descrip­
tion of the commodities included, the methods of averaging, the
statistical results obtained, etc., are set forth as far as the facts
could be learned, so that anyone wishing to use the figures in question
may know how they are derived and what they mean. This syste­
matic description of the series now being published is preceded by a
critical analysis of the various methods employed to measure changes
in the level of prices—an analysis which shows the advantages and
the defects of these methods, the purposes for which the different
index numbers may properly be employed, the reasons for the dis­
crepancies which usually appear when two or more series for the same
time and country are put side by side, the safeguards which are neces­
sary in making comparisons between different series, and the confi­
dence which index numbers merit as measures of price fluctuations.
I.—THE HISTORY OF INDEX NUMBERS.

The honor of inventing the device now commonly used to measure
changes in the level of prices probably belongs to an'Italian, G. R.
Carli. In an investigation into the effect of the discovery of America
i The writer has received generous help from Prof. Irving Fisher, Prof. Allyn A. Young, Dr. Royal
Meeker, and Mr. C. H. Verrill, all of whom read the first draft of this paper and made many effective criti­
cisms. To his wife he is indebted not only for a critical reading of the manuscript, but also for the drawing
of the charts.




6

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

upon the purchasing power of money, he reduced the prices paid for
grain, wine, and oil in 1750 to percentages of change from their prices
in 1500, added the percentages together, and divided the sum by
three, thus making an exceedingly simple index number. Since his
book was first published in 1764, index numbers are now 150 years
old.1
It was in England, however, where practically the same device had
been hit upon by Sir George Schuckburg-Evelyn in 1798,2 that the
theory and practice of index numbers were chiefly developed. The
generation that created the classical political economy was deeply
interested in the violent price fluctuations that accompanied the
Napoleonic wars and the use of an irredeemable paper currency from.
1797 to 1821. Several attempts were made to measure these fluc­
tuations, and in 1833 G. Poulett Scrope suggested the establishment
of a “ tabular standard of value.” 3
Interest in the study of price fluctuations lagged somewhat in the
forties; but the great rise of prices after the Californian and Aus­
tralian gold discoveries started fresh investigations. W. S. Jevons
in England and Adolf Soetbeer in Germany gave a powerful impetus
to the theoretical discussion and the practical computation of index
numbers. The problem changed somewhat in form but received even
more attention after 1873, when a prolonged fall of prices began. In the
sixties the chief aim of investigation had been to discover the relations
between the rise of prices and the increased production of gold; in the
seventies and eighties the chief aim was to find the relations between
the fall of prices and the restrictions placed upon the free coinage of
silver. The weightiest theoretical contributions of this period were
made by Prof. F. Y. Edgeworth, who served as secretary of a com­
mittee appointed by the British Association for the Advancement of
Science “ for the purpose of investigating the best methods of ascer­
taining and measuring variations in the value of the monetary stand­
ard.” 4
The problem of price fluctuations did not enter upon its present
phase until the world-wide rise of prices which began in 1896-97 had
been under way for several years. After 1900, and more insistently
after 1910, complaints about the rising cost of living became common
in practically all civilized countries. Efforts to measure this increase
as well as efforts to explain it multiplied. Index numbers are both
1 Del Valore e della Proporzione d©’ Metalli Monetati eon i generi in Italia prima deile Scoperte dell’ Indie
col confronto del Valore e della Proporzione de’ Tempi nostri. Republished by Custodi in his Scrittori
Italiani de Economia Politiea. Parte Modema, Vol. X III, pp. 297-366, especially pp. 335-354.
2 “ An account of some endeavors to ascertain a standard of weight and measure,” Philosophical Trans­
actions of the Royal Society of London, 1798, Part I, Art. VIII, pp. 133-182, especially pp. 175 and 176.
s Principles of Political Economy, London, 1833, pp. 405-408.
* For the reports of this committee, see the Reports of the British Association, 1887, pp. 247-254; 1888,
pp. 181-188; 1889, p. 133; 1890, pp. 485-488. See particularly the memoranda by Prof. Edgeworth subjoined
to these reports.




THE MAKING AND USING OF INDEX NUMBERS.

7

troublesome and expensive to compile, yet now in the United States
not less than five series are currently maintained, four of them by
financial papers. In England there are three important series; in
France two; in Germany three; while the Governments of Italy,
Denmark, Netherlands, Russia, Canada, Australia, and Japan publish
official index numbers, and private investigators have made series for
Belgium, Norway, Austria, Spain, New Zealand, and India. This list
may well be incomplete even at present, and is almost certain to
require additions within a short time.
Most of the series just mentioned have been established but recently.
The oldest—that of the London Economist—was begun in 1869.1
Sauerbeck’s English series dates from 1886, Conrad’s German series
from 1887 (though in a sense it continues investigations made by
Laspeyres in 1864), and Bradstreet’s American series from 1897.
Of the remaining index numbers regularly published at present, all
date from years since 1899, and the majority from years since 1909.2
With this increase in numbers there has come an improvement in
quality. The early index numbers were made by private investi­
gators, at irregular intervals, from such price quotations as chance
had preserved. As public appreciation of the importance of meas­
uring changes in price levels has developed, the work has more and
more been assumed by financial journals and Government bureaus.
This shift has produced a greater measure of continuity in the series,
as well as greater frequency, regularity, and promptness in the pub­
lication of the results. Even more important is the improvement
in the character and the scope of the price quotations from which
the index numbers are made. Whereas the individual investigator
had to take what he could get in the way of data, financial journals
and Government bureaus can collect those current prices that are
best adapted for statistical treatment, and can give better assurance
of the representative value of their quotations and the uniform
quality of the commodities included in successive years.
This improvement in the quantity and quality of index numbers
is as marked in the United States as elsewhere. Price quotations
had been published with more or less care and system by various
newspapers and periodicals for many years before the first effort to
compile an average of price variations was made. In 1881, Mr. H. C.
Burchard, Director of the Mint, made an index number covering the
years 1825 to 1880 from quotations that had been printed in certain
reports of the Secretary of the Treasury, supplemented by quotations
from a New York newspaper. But his data were of uncertain quality
1 From 1864 to 1869 the Economist published the relative prices of commodities, but such separate figures
without a sum or an average do not constitute an index number proper.
2 The years mentioned are the dates of first publication, not the earliest dates for which relative prices
are shown. In most cases the computers carried their investigations back into the past, frequently for a
decade or more.




8

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

and his series was allowed to lapse after 1884.1 After an interval
of eight years, the Senate Committee on Finance authorized a more
ambitious effort. Under the direction of Dr. Roland P. Falkner,
the statistician of this committee, the (then) Department of Labor
made a huge collection of price quotations, running back as far as
1840, and compiled an index number including more than 200 com­
modities for the years 1860 to 1891, and 85 commodities for 1840 to
1891.2 But this also was a single investigation, and the United States
did not have an index number regularly maintained year after year
until the establishment of Bradstreet’s series in 1897. A quasi
continuation of the Senate Finance Committee’s work, covering
the years 1890-1899, was prepared by Dr. R. P. Falkner, and pub­
lished by the Department of Labor in March, 1900.3 Another short­
lived series was begun by Prof. John R. Commons and Dr. N. I. Stone
in the Quarterly Bulletin of the Bureau of Economic Research later
in the same year.4 In January, 1901, the second continuous Ameri­
can series was started by Dun’s Review and gradually carried back
to 1860; the third, covering the years 1890 to date, was added by
the Federal Department of Labor in March, 1902. Other series of
this type were begun by Thomas Gibson’s weekly market letters in
1910, and by the New York Times Annalist in 1913.
This recent activity in the making of index numbers has been
accompanied by a rapid growth of the literature of the subject.
Among the later contributions dealing with theoretical issues, the
first place belongs to the work of an American scholar, Mr. C. M.
Walsh. His great treatise upon The Measurement of General
Exchange-Value, published in 1901, is still the most comprehensive
book upon the subject. But the bibliographies that aim to cover
the field now include hundreds of items, and to them must go the
student who wishes a guide to further reading.5
H.—THE DIFFICULTIES OF MEASURING CHANGES IN THE
LEVEL OF PRICES.
It is a curious fact that men did not attempt to measure changes
in the level of prices until after they had learned to measure such
subtle magnitudes as the weight of the atmosphere, the velocity of
sound, fluctuations of temperature, and the precession of the equi­
1 See Finance Reports, 1881, pp. 312-321; 1882, pp. 252-254; 18.83, pp. 316-318; Report of the Director of
the Mint on the Production of the Precious Metals, 1884, pp. 497-502. Compare the criticism of this series
by Prof. J. Laurence Laughlin, Quarterly Journal of Economics, April, 1887, pp. 397 and 398.
2 See the description given on pp. 128-139.
3 See Bulletin of the Department of Labor, No. 27, March, 1900.
4 See the issues for July and October, 1900.
5 For such bibliographies see Walsh, The Measurement of General Exchange-Value, pp. 553-574, and
J. L. Laughlin, Principles of Money, pp. 221-224. The most important contribution of later date than
Laughlin’s entries is Prof. Irving Fisher’s Purchasing Power of Money, pp. 385-429.




THE MAKING AND USING OF INDEX NUMBERS.

9

noxes. Their tardiness in attacking that problem is the more strange
because price changes had frequently been a subject of acrimonious
debate among publicists and a cause of popular agitation. Long
before the high development of the credit system and the wage-eaming class practical issues of grave importance were raised by the insta­
bility of prices, as the disturbances created in sixteenth-century
Europe by the inflow of American silver and gold abundantly show.
Perhaps disinclination on the part of “ natural philosophers” to soil
their hands with such vulgar subjects as the prices of provisions was
partly responsible for the delay;1 but after all a number of eminently
“ respectable” men wrote upon economic topics in every generation
after the days of Columbus—to go no further back. Nor can the
technical difficulties of the problem explain this tardiness; for the
mathematical intricacy of index numbers, and even the necessity
of allowing for changes in the pure silver content of coins, are obstacles
far less formidable than those surmounted long before in other fields
of research.
Probably the chief cause of delay was that averages of price fluctu­
ations did not promise to command much confidence after they had
been made. The quotations available for use by the early investi­
gators were few in number and often of doubtful accuracy. Carli,
for example, dealt with only 3 commodities; Shuckburg-Evelyn with
only 12. About the vastly greater number of unrecorded price
fluctuations the one firmly established fact was that they exhib­
ited bewildering diversity. Under these circumstances, could an
average made from a few samples be accepted as a reliable measure
of changes in the general level of prices ? And if averages could not
be trusted, why trouble to devise a plan of making them ? So writers
upon prices long contented themselves with statements about the fluc­
tuations of particular commodities, and with indefinite assertions that
the purchasing power of money had changed little or changed much.
So, also, when certain bold investigators did finally venture to make
index numbers, no one was particularly impressed by the significance
of their achievement.
This lack of faith in the validity of averages of price variations was
overcome rather slowly, partly in consequence of improvements in
business organization. The multiplication of commercial news­
i One of the early, British writers on prices, Bishop Fleetwood, remarked: “ * * * as the World now ,
goes, the greatest (tho’ I will not think the best) Part of Readers will be rather apt to despise than to com­
mend the Pains that are taken in making Collections of so mean Things as the price of Wheat & Oats, of
Poultry, and such like Provisions * *
Chronicon Preciosum, 1707, 2d ed., 1745, p. 6. Sir G. Shuckburg-Evelyn, in the paper referred to above, also felt himself on the defensive in presenting the first English
index number: “ * * * However, I may appear to descend below the dignity of philosophy, in such
economical researches, I trust I shall find favour with the historian, at least, and the antiquary.” Shuckburgs-Evelyn's discussion of index numbers, indeed, was merely a minor appendix to his discussion of
standards of weights and measures. But it has become his chief claim to remembrance.




10

BULLETIK OF THE BTJEEAU OE LABOR STATISTICS.

papers and the more systematic keeping of private and public records
provided a larger and more accurate body of quotations. Improved
means of transportation made wholesale prices in the larger cities
basic for many local markets. The grading and standardizing of
commodities increased the number of articles which could be safely
accepted as substantially uniform in quality from one year to the
next. More important still was the discovery by statisticians that
social phenomena of most kinds, though seeming to result from the
uncontrolled choice of individuals, yet reveal a striking regularity
when studied in large numbers. The demonstration that a formerly
unsuspected regularity lay hidden in one set of numerical data after
another encouraged economists to believe that the known price varia­
tions might after all be fair samples of the more numerous unknown
variations. The general similarity of the results reached by different
investigators using dissimilar data confirmed this faith. Thus em­
boldened, economic statisticians devoted much time to extending the
scope and improving the technique of index numbers. And their
growing confidence in the trustworthiness of their series was gradually
imparted to the public.
To-day few, if any, competent judges doubt the validity of index
numbers or the substantial accuracy of the results they show when
properly constructed from carefully collected data. Indeed the
danger at present is rather that the figures as published will be taken
too absolutely as a complete representation of the facts about price
fluctuations. It is therefore well to begin a study of index numbers,
not by analyzing the finished series, but by inspecting the actual
changes in prices from which they are made, and which they purport
to summarize. In no other way, indeed, can the value and the
limitations of index numbers be learned.
HI.— THE CHARACTERISTICS OF PRICE FLUCTUATIONS.

An excellent collection of materials for the study of changes in
wholesale prices is found in the Bulletin of the Bureau of Labor
Statistics, No. 149. Here are given the average annual prices at
wholesale of more than 230 commodities for a period of almost a
quarter of a century. Comparison of the changes in these actual
prices is facilitated by the publication of two series of relative prices
for each commodity. One series reduces the quotations in dollars
and cents to percentages of the average actual prices in the decade
1890-1899. The second series, which may be called ‘ i chain relatives/ ’




TH E MAKlHGr AND CJSING OF INDEX LUM BERS.

11

shows the percentage by which each article rose or fell in price each
year as compared with the year before.1
A survey of these relative figures for the 230 commodities throws
the diversity of price fluctuations into high relief. (1) During the
24 years 1890-1913 no two of the commodities quoted have under­
gone the same changes in price. Some articles have risen rather stead­
ily in price and fluctuate on a much higher level in 1913 than in 1890;
for example, rosin and crude petroleum. Other articles have fallen
much more than they have risen and fluctuate on a much lower level
at the end than at the beginning; for example, soda and wood alcohol.
Some articles are steady in price, seldom changing from one year to
the next; for example, bread and certain kinds of tools. Other
articles change in price every year, for example, cotton and pig
iron. (2) In every year a considerable proportion of the com­
modities rise in price, a considerable proportion fall, and a somewhat
smaller proportion remain unchanged. (3) The range covered even
by the fluctuations from one year to the next is very wide. For exam­
ple, in 1896 potatoes fell 54.6 per cent, while coke rose 41.5 per cent;
i The reader may follow the discussion more easily if he runs over the following sample of the figures
referred to:
Cotton, upland, middling.

Year.

Average, 1890-1899........................
1890...............................................

Average
price per
pound.

$0.07762

Relative
price.

Per cent of
increased+ )
or de­
crease ( —)
compared
with pre­
ceding year.

100.0

.11089
.08606
.07686
.08319
.07002

142.9
110.8
99.0
107.2
90.2

-2 2.4
-1 0.7
+ 8.2
-1 5.8

1896...............................................
1897...............................................
1898...............................................
1899...............................................

.07298
.07918
.07153
.05972
.06578

94.0
102.0
92.2
76.9
84.7

+ 4.2
+ 8.5
- 9.7
-1 6.5
+10.1

1900...............................................
1901...............................................
1902...............................................
1903...............................................
1904...............................................

.09609
.08627
.08932
.11235
.12100

123.8
111.1
115.1
144.7
155.9

+46.1
-1 0.2
+ 3.5
+25.8
+ 7.7

1905...............................................
1906...............................................
1907...............................................
1908...............................................
1909...............................................

.09553
.11025
.11879
.10463
.12107

123.1
142.0
153.0
134.8
156.0

-2 1 .0
+15.4
+ 7.7
-1 1 .9
+15.7

1910...............................................
1911...............................................
1912...............................................
1913...............................................

.15118
.13037
.11503
.12792

194.8
168.0
148.2
164.8

+24.9
-1 3.8
-1 1 .8
+11.2




12

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

in 1899 wheat flour fell 20.2 per cent, while steel billets rose 103.3 per
cent; in 1913 onions fell 38.5 per cent, while cabbage rose 58.5 per
cent.1
Such extreme diversities as have been cited, however, give a mis­
leading impression of chaos among the fluctuations. A just impres­
sion can be had only from some scheme of presentation which takes
account of all the commodities at once. Table 1 is a first rough
approximation toward this end. It shows for each year how many
of the commodities quoted rose, remained unchanged, or fell in price,
and divides those which rose and those which fell into six groups,
according to the magnitude of their fluctuations.
i All of these figuresshow percentages of rise or fall from the average prices of the commodities in question
in the preceding year.




T

able

1.—CONSPECTUS OF THE CHANGES IN W HOLESALE PRICES IN THE UNITED STATES, B Y Y EA R S, 1891 TO 1913.

[Based upon the percentages of increase or decrease in price from one year to the next, given in Table II of Bulletin of the United States Bureau of Labor Statistics, No. 149, May, 1914.]

Year.

Number
of com­
Less
modities
2.0
to
5.0
to
10.0
to
20.0
to
50.0
per
than 4.9 per 9.9 per 19.9 per 49.9 per
thatrose
cent or
2.0 per cent.
cent.
cent. more. in price.
cent.
cent.

tel

OF
INDEX
NUM BERS,

82
55
78
19
77
73
92
135
169
184
89
143
128
113
131
167
162
55
124
146
75
137
133

8

USING

106
140
114
192
138
133
118
73
46
38
128
61
92
106
89
47
48
155
98
81
147
80

Number of commodities that rose in price by—

AND




232
232
234
236
237
240
241
242
242
242
242
242
242
242
242
242
242
242
253
253
253
253
252

Number
of com­
modities
that
did n o t
ch an ge
in price.

MAKING

1891.
1892.
1893.
1894.
1895.
1896.
1897.
1898.
1899.
1900.
1901.
1902.
1903.
1904
1905.
1906.
1907.
1908.
1909.
1910.
1911.
1912.
1913.

Number of commodities that fell in price b y Total
number Number
of com­
of com­
Less
50.0 per 20.0 to 10.0 to 5.0 to 2.0 to
modities modities
that
fell cent or 49.9
than
quoted
per 19.9 per 9.9 per 4.9 per 2.0
in
price.
per
more.
cent.
cent.
each year.
cent.
cent.
cent.

00

14

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

A more significant presentation of the same set of price fluctua­
tions is given by Table 2. To make this table a tally sheet was
drawn up for each year from 1891 to 1913, on which the changes
from prices in the preceding year were entered in the order of their
magnitude, beginning with the greatest percentage of fall and run­
ning up through “ no change’ ’ to the greatest percentage of rise.
Then the whole number of recorded fluctuations for each year was
divided into 10 numerically equal groups, again beginning with the
case of greatest fall and counting upward. Finally the nine dividing
points between these 10 equal groups were marked off in the percent­
age scale of fall, “ no change, ” or rise. For example, the tally sheet
for 1913 showed how the average prices of 252 commodities in that
year differed from their average prices in 1912. One-tenth of these
252 commodities showed a fall of prices ranging between 38.5 per
cent and 10.4 per cent, the second tenth ranged between a fall of
10.4 per cent and one of 3.7 per cent; the third tenth ranged between
a fall of 3.7 per cent and one of 1 per cent; the fourth tenth between
a fall of 1 per cent and “ no change” ; the fifth tenth between “ no
change” and a rise of 0.5 per cent, and so on. These dividing points
( —10.4 per cent, —3.7 per cent, —1 per cent, ± 0 per cent, +0.5
per cent, etc.) between the successive tenths into which the data
were split are called “ decils.” The midmost decil, which of course
divides the whole number of observations into two equal groups, is
called the “ median.” Table 2 presents the results drawn from the
tally sheets—that is, the nine decils for each year, together with the
percentages of greatest fall and of greatest rise from prices in the
year before.




S"§

S

*

£

•-

4

tr p -

2?
J
O
v

2

2?
2*
- S ■; s
« ® S
j i - 50 «5
|
o

I
o

£

0
«>
&

*

*
■*

-J-1 I I » i I I I I i i I I I I I I 1.1 I I I I I i i 11 I i i I i i i 1 I I I i I I I i i i I I I I I i I i I I I I i i I I i I I I I I I I I I I I l l I l l I I I I I I II I i I i i i i I I I I I l l I I I I I I I ! I I I I I I I Ll.l I I I.LI 1 I I I I I I I I I I I I I I I I I I I I I I I 11.1 I I I I I U U J -J. J-l
94261°—15. (To iace page 15.)




15

THE M AKING AND USING OF INDEX NUMBERS.
T able

2.—CHAIN IN D E X NUMBERS OF PRICES AT W HOLESALE IN THE UNITED
STATES, BY YEARS, 1891 TO 1913.

[The decils are those points in the percentage scale of rise or fall in price which divide the whole numbe1*
of price changes recorded each year into 10 equal groups. Based upon the percentages of increase or
decrease in price from one year to the next, given in Table II of Bulletin of the United States Bureau of
Labor Statistics, No. 149, May, 1914.J
(— Indicates a fall; + Indicates a rise; ± 0 indicates “ no change.**)

Year.

Great­
est
fall.

1st
decil.

2d
decil.

3d
decil.

4th
decil.

Me­
dian.

6th
decil.

7th
decil.

8tb
decil.

9th
decil.

1891. .*.........
1892............
1893............
1894............
1895............
1896............
1897............
1898............
1899............
1900............
1901............
1902............
1903............
1904............
1905............
1906............
1907............
1908............
1909............
1910............
1911............
1912............
1913............

Per ct.
-30.5
-4 1 .2
-2 7.5
-4 4.3
-3 8 .0
-54. 6
-5 0.9
-2 1 .9
-2 0.2
-2 9.2
-4 2 .6
-4 0 .6
-3 3 .7
-4 3.8
-4 4 .9
-39.1
-4 3 .0
-3 9.5
-2 9 .8
-3 7 .7
-4 7 .4
-36.1
-3 8.5

Per ct.
-1 3.2
-1 6 .0
-1 1 .9
-2 1 .4
-1 4 .0
-1 7 .8
-1 1 .5
- 7.0
- 3.8
- 3.6
-1 5 .0
- 7.4
-1 2 .6
-1 5 .0
- 7.6
- 4.8
- 3.2
-2 1.3
- 7.7
- 6.1
-15.1
- 6.8
-1 0.4

Per ct.
- 8.0
-1 1 .2
- 8.0
-1 5.8
- 9.6
-1 1.3
- 7.2
- 3.3
± o
+ o
-1 0.2
- 1.6
- 5.3
- 7.6
- 3.9
± o
± o
-1 6.0
- 3.7
- 2.4
- 9.8
- 2.9
- 3.7

Perct.
- 4.8
- 8.5
- 5.5
-1 3 .4
- 6.5
- 7.5
- 4.4
- .4
± 0
+ 3.2
- 6.1
± 0
- 2.1
- 3.5
- 1.0
± o
± o
-10.-8
- 1.1
- .4
- 7.0
- .5
- 1.0

Per ct.
- 1.4
- 5.4
- 2.4
-1 0 .8
- 4.1
- 3.0
- 1.7
± o
+ 2.6
+ 5.1
- 3.7
± 0
± o
- .6
± o
+ 2.8
+ 1.2
- 5.8
± o
± o
- 4.2
± o
± 0

Perct.
±0
-3 .1
±0
-7 .1
-2 .4
-1 .2
±0
+1.8
+5.5
+7.5
-1 .5
+2.2
+1.3
±0
+ .7
+5.1
+3.9
-3 .8
±0
+1.5
- .9
+1.0
+ .5

Perct.
±0
-0 .5
±0
-5 .0
±0
±0
±0
+5.0
+7.6
+9.6
±0
+4.7
+3.7
+1.3
+3.2
+6.4
+6.6
- .9
+1.7
+3.6
±0
+3.6
+2.4

Perct.
+ 1.5
± o
+ 1.1
- 3.3
+ .7
+ .3
+ 2.9
+ 8.3
+10.6
+12.7
+ 1.3
+ 7.1
+ 5.3
+ 3.0
+ 5.9
+ 9.7
+ 8.9
± 0
+ 5.0
+ 6.3
± 0
+ 6.7
+ 4.5

Perct.
+ 5.0
+ 1.1
+ 4.8
- 1.3
+ 4.2
+ 4.3
+ 6.2
+13.3
+16.4
+17.4
+ 4.9
+12.1
+ 8.3
+ 5.9
+ 9.6
+14.5
+12.3
+ .8
+ 8.1
+ 9.2
+ 2.9
+11.0
+ 7.5

Perct. Per ct.
+15.3 + 53.0
+ 5.5 + 28.0
+11.0 + 59.1
+ 31.1
± o
+12.1 + 61.9
+10.2 + 41.5
+12.7 +101.6
+19.8 + 60.4
+30.8 +103.3
+25.6 + 72.8
+13.2 + 53.0
+20.4 + 58.9
+14.1 + 37.4
+11.7 + 39.9
+15.9 + 46.0
+18.9 . + 40.7
+17.6 + 67.8
+ 6.2 + 44.9
+16.0 + 70.1
+18.6 + 49.5
+11.0 + 86.1
+17.7 + 46.2
+12.7 + 58.5

Averages... -3 8.0

-11.0

- 6.2

- 3.5

- 1.4

+ .5

+2.3

+ 4.3

+ 7.8

+14.6

Great­
est
rise.

+ 57.0

Chart 1, based upon Table 2, gives a more vivid idea of these price
fluctuations. It shows for each year the whole range covered by
the recorded changes from prices in the preceding year by vertical
lines, which connect the points of greatest rise with the points of
greatest fall. These lines differ considerably in length, which indi­
cates that price changes cover a wider range in some years than in
others. The heavy dots upon the vertical lines show the positions
of the decils. One-tenth of the commodities quoted in any given
year rose above their prices of the year before by percentages scat­
tered between the top of the line for that year and the highest of
the dots. Another tenth fell in price by percentages scattered
between the bottom of the line and the lowest of the dots. The
fluctuations of the remaining eight-tenths of the commodities were
concentrated within the much narrower range between the lowest
and the highest dots. The dots grow closer together toward the
central dot, which is the median. This concentration indicates, of
course, that the number of commodities showing fluctuations of
relatively slight extent was much larger than the number showing
the wide fluctuations falling outside the highest and lowest decils,
or even between these decils and the decils next inside them.




16

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

The middle dots or medians in successive years are connected by a
heavy black line, which represents the general upward or downward
drift of the whole set of fluctuations. To make this drift clear the
median of each year is taken as the starting point from which the
upward or downward movements in the following year are meas­
ured. Hence the chart has no fixed base line. But in this respect
it represents faithfully the figures from which it is made; since these
figures are percentages of prices in the preceding year, a price fluc­
tuation in any year establishes a new base for computing the percent­
age of change in the year following. The fact that prices in the
preceding year are the units from which all the changes proceed is
further emphasized by connecting the nine decils, as well as the
points of greatest rise and fall with the median of the year before by
light diagonal lines. The chart suggests, and not inaptly, a series of
bursting bomb shells, the bombs being represented by the median
dots of the years before and the scattering of their fragments by the
lines which radiate to the decils and the points of greatest rise and
fall.1*
Time is well spent in studying this chart, because it is capable of
giving the mind a more just impression of the characteristics of price
changes than any other device.2 The marked diversity of the fluctu­
ations of different commodities in the same year—some rising, some
falling, some remaining unchanged—the wide range covered by these
fluctuations, and the erratic occurrence of extremely large changes
are strikingly shown; but so also are the much greater frequency of
rather small variations, the dense concentration near the center of
the field, the existence of a general drift in the whole complex of
changes, and the frequent alterations in the direction and the de­
gree of this drift. But if the chart is effective in giving these im­
pressions, it leaves them rather vague. To render certain of them
more definite, recourse must be had to the figures from which the
chart was drawn.
These figures, already given in Table 2, enable us to measure the
concentration of the mass of fluctuations about the center of the field.
One way to measure this concentration is to compute the differences
between the successive decils; that is, to find the range within which
successive tenths of the fluctuations fall. This “ range” is, of course,
1 Owing to the constant shifting of the base line, no fixed scale of relative prices can be shown on the
margin of the chart. But the offsets on the margin indicate the vertical distances allotted to a rise or fall
of 1 per cent from the prices of the preceding year.
The scale used here, that is, the ratio between the horizontal distances which represent one year, and the
vertical distances which represent 1 per cent of rise or fall in prices, is the same as that used in the other
charts in this bulletin, and in the earlier bulletins in this series. Because of its intricacy, the chart had to
be reproduced on a larger scale than in the other cases, but of course that fact does not alter the slant of
the lines, and this slant is the matter of importance.
2 Except, perhaps, a similar chart
logarithmic scale.




drawn to a-

17

THE M AKING AND USING OF INDEX NUMBERS.

a certain number of points in the percentage scale of change from
prices in the year before. When this computation is made for the
whole period covered by the table, we get the results presented in
Table 3. As heretofore, the successive tenths of the fluctuations
represented are reckoned by starting with cases of greatest fall in price
and counting upwards to cases of greatest rise. The central division
of the table shows that the average range covered by the fluctuations
diminishes rapidly as we pass from the cases of greatest fall toward
the cases of little change, and then increases still more rapidly as we
go onward to the cases of greatest rise. The right-hand group of
columns shows how the range increases if we start with the two
middle tenths, take in the two tenths just outside them, then the two
tenths outside the latter, and so on until we have included the whole
body of fluctuations. The left-hand group of columns, on the other
hand, combines in succession the two tenths on the outer boundaries,
then the two tenths immediately inside them, and so on until we get
back again to the two central tenths. Perhaps the most striking single
result brought out by this table is that eight tenths of all the fluctua­
tions are concentrated within a range (25.7 per cent) slightly nar­
rower than that covered by the single tenth that represents the
heaviest declines (27 per cent), and much narrower than that covered
by the single tenth that represents the greatest advances (42.4 per
cent).
T able 3.— AVERAGE CONCENTRATION OF PRICE FLUCTUATIONS AROUND THE ME­

DIAN, 1891 TO 1913.
[Based upon Table 2. The fluctuations represent percentage changes from average prices in the preceding
year.]
Average range covered by the—
1st and
10th
tenths
of the
price
fluctu­
ations.

69.4

2d and
9th
tenths
of the
price
fluctu­
ations.

3d and
8th
tenths
of the
price
fluctu­
ations.

11. 8.

6.1

4th and
7th
tenths
of the
price
fluctu­
ations.

5th and
Central Central Central Central
6th
four
two
six
eight
tenths Successive tenths tenths tenths tenths tenths
of the price fluc­ of the of the of the of the
of the
tuations.
price
price
price
price
price
fluctu­ fluctu­ fluctu­ fluctu­
fluctu­
ations. ations. ations. ations.
ations.
1st
2d
3d
4th
3.6j 5th
6th
7th
8th
9th
10th

94261°—Bull. 173—15------2




tenth, 27.0
tenth, 4.9
tenth,
tenth,
tenth,
tenth,
tenth,
tenth,
tenth, 6.9
tenth, 42.4

3.6

13.9

25.:

Whole
number
of the
price
fluctu­
ations.

95.1

18

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

Such results as these gain greatly in significance by being put
beside corresponding results for other groups of statistical data. The
best comparison to make, however, is one between the actual distri­
bution of our price fluctuations about their average and a “ normal”
distribution of the same data—that is, a distribution according per­
fectly with the so-called “ normal law of error.” This law shows how
phenomena are distributed about their average when the number of
phenomena observed is very large, and when each phenomenon is the
resultant of numerous independent factors, none of which is of pre­
ponderating importance. It has been found that many kinds of phe­
nomena tend to conform rather closely to this normal distribution; for
example, human heights, errors of observation, shots at a target, wage
rates in different occupations, etc.1 When it can be shown that phe­
nomena are distributed approximately in this fashion, their average
can safely be accepted as a significant measure of the whole set of varia­
tions, since even the deviations from the average are then grouped
in a tolerably definite and symmetrical fashion about the average.
With such a comparison in view we may treat each recorded per­
centage of rise or fall in price as an observation of the degree and
direction in which prices vary from one year to the next. Taking
all the commodities and all the years covered by the bureau’s chain
relatives, we have 5,578 observations for analysis. Table 4 shows
how these cases are distributed along a percentage scale of rise or
fall in prices which jumps two points at a time. The columns headed
“ number of cases” show how many price variations of the given mag­
nitudes and directions occur, and the columns headed “ proportion
of cases” show the same numbers in the rather clearer form of per­
centages of their sum (5,578).
i See, for example, Prof. F. Y . Edgeworth’s article “ Probability/’ Part II, Encyclopaedia Britajonica.
11th ed., and the references there given.




19

THE M AKING AND USING OF INDEX NUMBERS.

T able 4.— DISTRIBUTION OF 5,578 CASES OF CHANGE IN THE W HOLESALE PRICES

OF COMMODITIES FROM ONE Y E A R TO THE N E XT , ACCORDING TO THE MAGNI­
TUDE AND DIRECTION OF THE CHANGES.
[Based upon the chain relatives in Table II of Bulletin of the Bureau of Labor Statistics, No. 149.]
Rising prices.
Per cent of
change from
the average
price of the
preceding
year.
102-103.9
100-101.9
98- 99.9
96- 97.9
94- 95.9
92- 93.9
90- 91.9
88- 89.9
86- 87.9
84- 85.9
82- 83.9
80- 81.9
78- 79.9
76- 77.9
74- 75.9
72- 73.9
70- 71.9
68- 69.9
66- 67.9
64- 65.9
62- 63.9
60- 61.9
58- 59.9
56- 57.9
54r~ 55.9
52- 53.9
50- 51.9
48- 49.9

Num­
ber of
cases.

Propor­
tion of
cases.

1
1

0.018
.018

1
1
1
1

.018
.018
.018
.018

1
4
1
3
4

.018
.072
.018
.054
.072

4
6
1
3
4
1
5

.072
.108
.018
.054
.072
.018
.090

Per cent of
change from
the average
price of the
preceding
year.

Falling prices.

Num­
ber of
cases.

Proportion
of cases.

46-47.9
44-45.9
42-43.9
40-41.9
38-39.9
36-37.9
34-35.9
32-33.9
30-31.9
28-29.9
26-27.9
24-25.9
22-23.9
20-21.9
18-19.9
16-17.9
14-15.9
12-13.9
10-11.9
8- 9.9
6- 7.9
4- 5.9
2- 3.9
Under 2.

11
10
6
14
17
11
18
17
22
30
29
47
45
65
73
1102
106
115
167
i 237
261
1356
355
1410

0.197
.179
.108
.251
.305
.197
.323
.305
.394
.538
.520
.843
.807
1.165
1.308
1.828
1.900
2.062
2.994
4.249
4.679
6.382
6.364
7.350

No change.

1697

12.494

Per cent of
change from
the average
price of the
preceding
year.
Under 2.
2- 3.9
4 -5 .9
6- 7.9
8- 9.9
10-11.9
12-13.9
14-15.9
16-17.9
18-19.9
20-21.9
22-23.9
24-25.9
26-27.9
28-29.9
30-31.9
32-33.9
34-35.9
36-37.9
38-39.9
40-41.9
42-43.9
44-45.9
46-47.9
48-49.9
50-51.9
52-53.9
54-55.9

Propor­
tion of
cases.

Number
of cases.

1405
1375
329
1238
200
173
1120
107
76
71
45
39
32
17
27
16
7
10
7
5
5
4
2
1
1
1

7.261
6.723
5.898
4.267
3.585
3.101
2.151
1.918
1.362
1.273
.807
.699
.574
.305
.484
.287
.125
.179
.125
.090
.090
.072
.036
.018
.018
.018

1

.018

Summary.

Number
of cases.

Proportion
of cases.

Kfefrig prices...............................................................................................................
No change...................................................................................................................
Falling prices..............................................................................................................

2,567
697
2,314

46.021
12.494
41.485

Total..................................................................................................................

5,578

100.000

1Location of the decils.

Such is the actual distribution of the phenomena under analysis.
To compare it with the “ normal” distribution, we put these figures
on a chart, which presents the facts clearly to the eye. Here the
horizontal scale represents percentages of rise or fall in price, and
the vertical scale represents the number of times each percentage of
change is observed. The dotted line shows how our 5,578 cases
would have been distributed had they followed strictly the normal
law of error. The areas included by the unbroken line and the dotted
line are equal.
There are three significant points to notice here: (1) The two
forms of distribution, the actual and the "normal,” are of the same



20

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

type. (2) The concentration about the central tendency is greater
in the actual than in the “ normal” distribution; but on the other
hand, the extreme variations diverge further from this central tend­
ency in the actual distribution than in the other.1 (3) Unlike the
“ normal” distribution, the actual distribution is not perfectly sym­
metrical. Two closely related aspects of this difference may be
pointed out: First, the outlying cases of the “ normal” distribution
Ch art

2 —DISTRIBUTION OF 5,578 PRICE VARIATIONS (PERCENTAGES OF RISE OR
FALL FROM PRICES OF PRECEDING Y E A R ).

extend precisely the same distance from the central tendency in both
directions, whereas in the actual distribution the outlying cases run
much farther to the right (in the direction of a rise in prices) than to
the left (in the direction of a fall).2 Second, the central tendency
1 The greater concentration in the actual than in the “ normal” distribution is perhaps best shown by the
fact that the probable deviation (half of the range between the quartiles) is only 5.6 points in the first case
as against 8.9 points in the second. The quartiles correspond in nature to the decils and medians. They
are the points which divide the whole number of observations into four equal part s. In the case of the
actual distribution the quartiles are +6.6 and —4.6.
2 The extremes are +103.3 per cent and —54.6 per cent. From another viewpoint, however, these ex­
tremes may be regarded as nearly symmetrical. The greatest rise represents a price of the year before mul­
tiplied by 2.033; the greatest fall a price of the year before divided by 2.203. The significance of these two
ways of comparing the magnitude of a rise and a fall in prices is discussed in Section IV, subsection 8.




THE MAKING AND USING OF INDEX NUMBERS.

21

itself is free from ambiguity in one case but not in the other. In the
u normalM distribution this tendency may be expressed indifferently
by the median, the arithmetic mean, or the mode (the point of great­
est density); for these three averages coincide. In the actual dis­
tribution, on the contrary, these averages differ slightly; the median
and mode stand at ±0, while the arithmetic mean is +1.36 per cent.1
These departures of the actual distribution from perfect symmetry
possess a certain significance; but, after all, they are minor qualifica­
tions of the important proposition; namely, year-to-year price fluc­
tuations are grouped about their central tendency in a strikingly
regular fashion.
This study of the actual distribution of price fluctuations from one
year to the next will be found to throw light upon several problems
presently to be faced in discussing the methods of making index
numbers. For the moment we have use primarily for the demonstra­
tion that these fluctuations are highly concentrated about a central
tendency. This conclusion strengthens the hope that we may make
measurements of price fluctuations that fairly represent the net
resultant of all the changes, miscellaneous as they seem to be. For
properly constructed averages have clearly a better chance of being
representative and significant when the phenomena for which they
stand have a strongly marked central tendency about which devia­
tions are symmetrically grouped than when the phenomena are irregu­
larly scattered over their range.
But it must be remembered, and with the reminder doubt reenters,
that the variations just analyzed are percentages of increase or de­
crease from the prices of the year before. Most index numbers,
however, attempt to measure price fluctuations, not with reference to
the preceding year, but with reference to a period considerably more
remote. For example, the Bureau of Labor Statistics measures prices
in 1913 in terms of average prices in the decade 1890-1899. Are price
variations computed in this manner highly concentrated around their
central tendency like the price variations with which we have been
dealing ?
Chart 3 answers this question emphatically in the negative. It
represents the distribution of the price variations of 241 commodities
quoted by the Bureau of Labor Statistics for the year 1913.2 These
variations are computed in two ways: (1) as percentages of rise or
fall from the prices of 1912; (2) as percentages of rise or fall from
1 That the arithmetic mean is slightly above zero arises partly from the fact that there are 33 percentages
of rise greater than any percentage of fall. But it also arises partly from the fact that our data come from
a period (1890-1913) when the trend of year-to-year fluctuations was more often upward than downward;
there were 2,567 cases of advance in price against 2,314 cases of fall. The median is kept from rising above
zero because the cases of “ no change/’ 697 in number, more than offset the difference between the numbers
of advances and of declines in price.
2The Bureau quotes 252 commodities in 1913; but 11 can not be included in the present comparison
because no quotations are given for them in 1890-1899.




to
to

3.—DISTRIBUTION OF THE PRICE VARIATIONS OF 241 COMMODITIES IN 1913 (PERCENTAGES OF RISE OR FALL IN PRICES).

OF
THE
BUREAU
OF * LABOR
STATISTICS.




BULLETIN

Ch ar t

THE M AKING AND USING OF INDEX NUMBERS.

23

the average prices of 1890-1899. Of course the first set of varia­
tions corresponds in character to the variations represented above in
Chart 2. The distribution of these variations, shown by the area in­
closed by the unbroken line, is similar in type to the actual distri­
bution in Chart 2; although it is less regular—a difference to be
expected, since the number of observations is only 241 here as against
5,578 there. But the distribution of the second set of variations
(percentages of change from the average prices of 1890-1899) as repre­
sented by the area inclosed within the dotted line belongs to a
different type. It has no pronounced central tendency; it shows no
high degree of concentration around the arithmetic mean ( + 30.4
per cent) or median ( + 26 per cent). It is more like an oblong than
like the bell-shaped normal curve; it has a range between the greatest
fall (52.2 per cent) and greatest rise (234.5 per cent) so extreme that
two of the cases could not be represented on the chart; and its prob­
able deviation is five times as great as that of the corresponding
variations from 1912 prices— 18.5 points as against 3.6.1
Price variations, then, become dispersed over a wider range and less
concentrated about their mean as the time covered by the variations
increases. The cause is simple: With some commodities the trend
of successive price changes continues distinctly upward for years at a
time; with other commodities there is a consistent downward trend;
with still others no definite long-period trend appears. In any large
collection of price quotations covering many years each of these
types, in moderate and extreme form, and all sorts of crossings among
them, are likely to occur. As the years pass by the commodities that
have a consistent trend gradually climb far above or subside far below
their earlier levels, while the other commodities are scattered between
these extremes. Thus the percentages of variation for any given
year gradually get strung out in a long, thin, and irregular line, with­
out a marked degree of concentration about any single point.
The consequence is that the measurement of price fluctuations
becomes difficult in proportion to the length of time during which the
variations to be measured have continued. In other words, the
farther apart are the dates for which prices are compared, the wider
is the margin of error to which index numbers are subject, the greater
the discrepancies likely to appear between index numbers made by
different investigators, the wider the divergencies between the aver­
ages and the individual variations from which they are computed, and
the larger the body of data required to give confidence in the repre­
sentative value of the results.
From this preliminary survey of the characteristics of price fluc­
tuations it appears (1) that year-to-year changes in the price level
* The probable deviation, to repeat, is half the inter-quartile range.




24

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

can be measured with, good prospects of success, because such varia­
tions show a symmetrical distribution and a marked degree of con­
centration about their central tendency; but (2) that measurements
of variations between years far apart have a more problematical
value. The practical question whether the index numbers in current
use can be trusted, then, may have two answers. Perhaps they give
results that are reliable as between successive years, and at the same
time doubtful for dates between which 10, 20, or 50 years have
intervened.
The best way to test the reassuring conclusion about index num­
bers for successive years and to resolve the disturbing doubt about
index numbers covering long periods is to compare different series
of index numbers that purport to measure price changes in the same
country during the same time. If the results turn out to be con­
sistent with one another our faith will be confirmed. If the results
are not consistent we must find a valid reason for the discrepancies,
or become skeptical about the present methods of measuring changes
in the price level.
When this test is applied, the first impression is unfavorable.
For example, the five currently published American index numbers
show the following results for 1912 and 1913:

Year.

Bureau of
Labor Sta­ Bradstreet’s
index
tistics’ index number.
number.

Annalist
index
number.

Gibson’s
index
number.

Dun’s index
number.

1912.......................................................
.......................................................
1913

133.6
135.2

$9.1867
9.2076

143.25
139.98

62.6
58.1

$124.44
120.89

Changes................................................

+1.6

+.0209

-3.27

-4 .5

-3.55

Here no two of the series are as closely consistent with each other
as one could wish. On the contrary, the five series disagree not only
as to the degree but also as to the direction of the change in prices.
And this is a comparison between successive years, where measure­
ments should be especially accurate.
Such offhand comparisons as the above, however, are not fair,
and the conclusion they suggest as to the unreliability of index num­
bers can not be accepted without further study, for these various
index numbers mean different things. They do not all undertake
to measure the same quantity, hence they do not all employ the same
methods, and hence the discrepancies among their results may reveal
no real inconsistency. No valid comparison of index numbers can
be made, indeed, without a careful examination of what is measured
and how the measurement is made. Such an examination accord­
ingly we must make before we can satisfy our minds upon the question
whether index numbers yield trustworthy results.




THE MAKING AND USING OF INDEX NUMBERS.

25

IV.— VARIETIES OF METHODS USED IN MAKING INDEX
NUMBERS.

Making an index number involves several distinct operations: (1)
Defining the purpose for which the final results are to be used; (2) de­
ciding the numbers and kinds of commodities to be included; (3) de­
termining whether these commodities shall all be treated alike or
whether they shall be weighted according to their relative importance;
(4) collecting the actual prices of the commodities chosen, and, in
case a weighted series is to be made, collecting also data regarding
their relative importance; (5) deciding whether to measure the aver­
age variations of prices or the variations of a sum of actual prices; (6)
in case average variations are to be measured, choosing the base upon
which relative prices shall be computed; and (7) settling upon the
form of average to be struck.
At each one of these successive steps choice must be made among
alternatives that range in number from two to thousands. The pos­
sible combinations among the alternatives chosen are indefinitely
numerous. Hence there is no assignable limit to the possible varie­
ties of index numbers, and in practice no two of the known series are
exactly alike in construction. To canvass even the important
variations of method actually in use is not a simple task.
1. THE RELATIONS BETWEEN METHODS AND USES.

The first step, framing a clear idea of the ultimate use of the results,
is most .important, since it affords the clue to guide the compiler
through the labyrinth of subsequent choices. It is, however, the
step most frequently omitted.
When the end in view is specific and capable of precise statement
the problem of choosing methods is comparatively simple. Straight­
forward logic then determines what commodities should be included,
what sources of quotations should be drawn upon, and how the origi­
nal data should be worked up to give the most significant results.
Puzzles a-plenty are left, but most of them are limited to finding the
best compromise between what logic marks out as desirable and
what is feasible in view of the time and money at the investigator’s
disposal.
Few of the widely-used index numbers, however, are made to serve
one special purpose. On the contrary, most of them are “ generalpurpose” series, designed with no aim more definite than that of
measuring changes in the price level. Once published they are used
for many ends—to show the depreciation of gold, the rise in the cost
of living, the alternations of business prosperity and depression, and
the allowance to be made for changed prices in comparing estimates
of national wealth or private income at different times. They are




26

BULLETIN OF . THE BUREAU OF LABOR STATISTICS,

cited to prove that wages ought to be advanced or kept stable; that
railway rates ought to be raised or lowered; that utrusts’7 have
manipulated the prices of their products to the benefit or the injury
of the public; that tariff changes have helped or harmed producers
or consumers; that immigration ought to be encouraged or restricted;
that the monetary system ought to be reformed; that natural re­
sources are being depleted or that the national dividend is growing.
They are called in to explain why bonds have fallen in price and why
interest rates have risen, why public expenditures have increased,
why social unrest prevails in certain years, why farmers are prosper­
ous or the reverse, why unemployment fluctuates, why gold is being
imported or exported, and why political “ landslides” come when
they do.
The compiler of a general-purpose index number, then, can not
foresee to what uses and misuses his figures will be put. For each of
the legitimate uses he might conceivably devise an appropriate series.
But he can not conceivably devise a single series that will serve all
uses equally well. For the very qualities that make an index number
good, say, for the man of affairs concerned with the business outlook,
may make it bad for other men interested in the fortunes of farmers,
in the effects of the tariff, in the relation between gold output and
prices, in comparing changes in price levels in different countries, etc.
The day has not yet come when the uses of index numbers are suffi­
ciently differentiated and standardized to secure the regular publi­
cation of numerous special-purpose series. Until that day does come
the making of general-purpose series will continue and the makers will
go on choosing their methods perforce on rather vague and general
grounds. So long also must most of the users of index numbers put
up with figures imperfectly adapted to their ends.
The critical student of contemporary index numbers is in the same
uncomfortable position as the compiler. He has no single rule of
right and wrong to apply in judging the different general-purpose
series, for methods that are legitimate for certain uses are question­
able for others. Nevertheless, it is futile (though not uncommon)
for him to discuss methods without reference to uses, since a statis­
tical method has neither merits nor defects except as a means to
certain ends. The one course that is open to him is to invert the
problem. Instead of studying methods in the light of uses, he must
study uses in the light of methods. That is, he must analyze the
effect of the different methods followed in practice and so determine
what the resulting figures mean and the uses to which they may
properly be put.
The following discussion proceeds upon this plan. It deals primarily
with the popular general-purpose series and endeavors to show how




THE M AKING AND USING OF INDEX NUMBERS.

27

the various methods used in constructing these index numbers deter­
mine the uses to which they are severally adapted.
2. COLLECTING AND PUBLISHING THE ORIGINAL QUOTATIONS.

The reliability of an index number obviously depends upon the
judgment and the accuracy with which the original price quotations
were collected. This field work is not only fundamental, it is also
laborious, expensive, and perplexing beyond any other part of the
whole investigation. Only those who have tried to gather from the
original sources quotations for many commodities over a long series
of years appreciate the difficulties besetting the task. The men who
deal with data already published are prone to regard all this prelimi­
nary work as a clerical compilation requiring much industry but
little skill. To judge from the literature about index numbers, one
would think that the difficult and important problems concern meth­
ods of weighting and averaging. But those who are practically
concerned with the whole process of making an index number from
start to finish rate this office work lightly in comparison with the
field work of getting the original data.
We commonly speak of the wholesale price of articles like pig iron,
cotton, or beef as if there were only one unambiguous price for
any one thing on a given day, however this price may vary from
one day to another. In fact there are many different prices for
every great staple on every day it is dealt in, and most of these differ­
ences are of the sort that tend to maintain themselves even when
markets are highly organized and competition is keen. Of course
varying grades command varying prices, and so as a rule do large
lots and small lots; for the same grade in the same quantities, differ­
ent prices are paid by the manufacturer, jobber, and local buyer; in
different localities the prices paid by these various dealers are not the
same; even in the same locality different dealers of the same class
do not all pay the same price to everyone from whom they buy the
same grade in the same quantity on the same day. To find what
really was the price of cotton, for example, on February 1, 1915,
would require an elaborate investigation, and would result in show­
ing a multitude of different prices covering a considerable range.
Now the field worker collecting data for an index number must
select from among all these different prices for each of his commodi­
ties the one or the few series of quotations that make the most repre­
sentative sample of the whole. He must find the most reliable
source of information, the most representative market, the most
typical brands or grades, and the class of dealers who stand in the
most influential position. He must have sufficient technical knowl­
edge to be sure that his quotations are for uniform qualities, or to




28

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

make the necessary adjustments if changes in quality have occurred
in the markets and require recognition in the statistical office. He
must be able to recognize anything suspicious in the data offered
him and to get at the facts. He must know how commodities are
made and must seek comparable information concerning the prices of
raw materials and their manufactured products, concerning articles
that are substituted for one another, used in connection with one
another, or turned out as joint products of the same process. He
must guard against the pitfalls of cash discounts, premiums, rebates,
deferred payments, and allowances of all sorts. And he must know
whether his quotations for different articles are all on the same basis,
or whether concealed factors must be allowed for in comparing the
prices of different articles on a given date.
Difficult as it is to secure satisfactory price quotations, it is still
more difficult to secure satisfactory statistics concerning the relative
importance of the various commodities quoted. What is wanted
is an accurate census of the quantities of the important staples,
at least, that are annually produced, exchanged, or consumed. To
take such a census is altogether beyond the power of the private
investigators or even of the Government bureaus now engaged in
making index numbers. Hence the compilers are forced to confine
themselves for the most part to extracting such information as they
can from statistics already gathered by other hands and for other
purposes than theirs. In the United States, for example, estimates
of production, consumption, or exchange come from most miscella­
neous sources: From the Department of Agriculture, the Census
Office, the Treasury Department, the Bureau of Mines, the Geo­
logical Survey, the Internal Revenue Office, the Mint, associations of
manufacturers or dealers, trade papers, produce exchanges, traffic
records of canals and railways, etc. The man who assembles and
compares estimates made by these various organizations finds among
them many glaring discrepancies for which it is difficult to account.
Such conflict of evidence when two or more independent estimates
of the same quantity are available throws doubt also upon the seem­
ingly plausible figures coming from a single source for other articles.
To extract acceptable results from this mass of heterogeneous data
requires intimate familiarity with the statistical methods by which
they were made, endless patience, and critical judgment of a high
order, not to speak of tactful diplomacy in dealing with the authori­
ties whose figures are questioned. The keenest investigator, after
long labor, can seldom attain more than a rough approximation to
the facts. Yet it is only by critical use of the data now available
that current index numbers can be weighted, and the best hope of




THE MAKING AND USING OF INDEX NUMBEBS.

29

improving weights in the future lies in demonstrating not only the
imperfections of our present statistics of production, consumption,
and exchange, but also the importance of making them better.
When all this preliminary work has been done, the original quota­
tions and the weights should be published at length. Unfortunately,
many compilers of index numbers publish only the final results of
their computations, upon the ground of expense or lack of interest
in the detailed information. But much is sacrificed by taking this
easy course. First, the reputation of the index number itself is
compromised, and deservedly. No one can really test whether a
series is accurately compiled from representative quotations unless
the data and their sources are given in full. Second, and more
important, the publication of actual quotations greatly extends the
usefulness of an investigation into prices. Men with quite other
ends in view than those of the original compilers can make index
numbers of their own adapted to their peculiar purposes if provided
with the original data.
Nor is the importance of such unplanned uses to be rated lightly.
If we are ever to make the money economy under which we live
highly efficient in promoting social welfare we must learn how to
control its workings. What wares our business enterprises produce
and what goods our families consume are largely determined by
existing prices, and the production and consumption of goods are
altered by every price fluctuation. What we waste and what we
save, how we divide the burden of labor and how we distribute its
rewards, whether business enjoys prosperity or suffers depression,
whether debts of long standing become easier or harder to pay—all
these and many other issues turn in no small measure upon what
things are cheap and what are dear, upon the maintenance of a due
balance within the system of prices, upon the upward or downward
trend of the price changes that are always taking place. But if the
prices of yesterday are powerful factors in determining what we
shall do and how we shall fare to-day, what we do and how we fare
to-day are powerful factors in determining what prices shall be
to-morrow. If prices control us, we also control them. To control
them so that they shall react favorably upon our economic fortunes
we need more insight than we have at present. It is, then, one of
the great tasks of the future to master the complicated system of
prices which we have gradually developed—to find how prices are
interconnected, how and why they change, and what consequences
each change entails. For when men have learned these things they
will be vastly more skillful in mending what they find amiss in
economic life, and in reenforcing what they find good. As yet our




30

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

knowledge is fragmentary and uncertain. But of all the efforts
being made to extend it none is so certain to prove fruitful as the
effort to record the actual prices at which large numbers of com­
modities are bought and sold. For such data are the materials with
which all investigators must deal, and without which no bits of
insight can be tested. Indeed, it is probable that long after the best
index numbers we can make to-day have been superseded, the data
from which they were compiled will be among the sources from
which men will be extracting knowledge which we do not know
enough to find.
3. MARKET PRICES, CONTRACT PRICES, AND IMPORT-EXPORT VALUES.

All the American index numbers are made from “ market prices.”
These prices are usually obtained directly from manufacturers,
selling agents, or wholesale merchants; from the records of produce
exchanges and the like; or from trade journals and newspapers
which make a specialty of market reporting in their respective
fields.
Several of the important foreign index numbers are made wholly
or partly from “ import and export values” ; that is, from the average
prices of important articles of merchandise as officially declared by
the importing or exporting firms, or as determined by governmental
commissions. For example, Soetbeer’s celebrated German series,
and the British Board of Trade’s official series are made mainly from
such material, and the official French series was made wholly from
import values until 1911.
A third source of quotations often drawn upon in Europe is the
“ contract prices” paid for supplies by such institutions as hospitals,
normal schools, poorhouses, army posts, and the like. The official
Italian series, Alberti’s series for Trieste, and Levasseur’s French
series are examples.
These three classes of quotations—market prices, import and export
values, and contract prices—usually differ somewhat, not only with
respect to the prices prevailing on a given date, but also with respect
to the degree of change from time to time. Accordingly it is desir­
able to inquire into the several advantages possessed by each source
of quotations.
Contract prices may be set aside promptly, because index numbers
made from them have a limited range of usefulness. Though the
institutions whose records are drawn upon often make purchases on
a considerable scale, yet the common description of their contract
rates as “ semiwholesale” prices points to the peculiar and there­
fore unrepresentative character of such data. Moreover, there is




THE M AKING AND USING OF INDEX NUMBERS.

31

often more doubt about the strictly uniform character of the supplies
furnished to these institutions than about the uniformity of the
standardized goods which are usually quoted in the market reports.
If the aim of the investigation is to find the average variations in the
cost of supplies to public institutions, doubtless contract prices are
the best data to use. But if the aim is to measure the average
variations in the wholesale prices paid by the business world at
large, then market prices are distinctly the better source. Indeed,
contract prices are seldom used for the latter purpose except when
well-authenticated market quotations can not be had.
The theory on which import and export values are sometimes pre­
ferred to market prices is that the former figures show more nearly
the variations in the prices actually paid or received by a country for
the great staples which it buys and sells than do market quotations
for particular brands or grades of these commodities. For example,
England buys several different kinds of cotton in proportions that
vary from year to year. A price obtained by dividing the total de­
clared values of all the cotton consignments imported by their total
weight will show the average cost per pound actually paid by Eng­
lishmen for cotton with more certainty than will Liverpool market
quotations for a single grade of cotton like “ Middling American” —
provided always that the “ declared values” are trustworthy. Now,
if the aim of the investigation is to find out the variations in the
average prices paid or received for staples—irrespective of minor
changes in their qualities—then the preference for import and export
values is clearly justified, again granted the trustworthiness of the
returns. But if the aim is to measure just one thing—the average
variation in prices—market prices for uniform grades are clearly bet­
ter data. For index numbers made from import and export values
measure the net resultant of two sets of changes, and one can not tell
from the published figures what part of the fluctuations is due to
changes in prices and what part is due to changes in the qualities of
the goods bought and sold.
As might be expected, import and export series generally pursue
a more even course than market-price series. But this difference
may be due less to the sources from which the quotations are obtained
than to differences in the lists of commodities used. Fortunately, we
can arrange a more certain test than any of the common series pro­
vide. In 1903 the British Board of Trade published the average
import or export prices of 25 commodities for which Mr. Sauerbeck




32

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

has published market prices.1 Index numbers made from these two
sets of data for the same commodities for the years 1871 to 1902 are
given in Table 5. The results confirm the expectation: As compared
with the import and export index number, the market-price index
number starts on a higher level in 1871, falls to a lower point dur­
ing the middle nineties, rises to a higher level in 1900, and again
drops to as low a level in 1902.
T

5.—COMPARISON OF IN D E X NUMBERS MADE FROM IMPORT AND E XP O R T
VALUES W ITH IN DEX NUMBERS MADE FROM THE M ARKET PRICES OF THE SAME
COMMODITIES, B Y YEARS, 1871 TO 1902.

able

[Data from the British Board of Trade and from Sauerbeck.]
(Arithmetic means of relative prices.

Year.

Average prices in 1890-1899=100.

Import
ana ex­
port
values.

Market
prices.

158
169
170
162
152
149
150
139
128
136
133
129
125
118

170
185
182
168
155
152
152
138
131
137
130
125
123
116

105

107

1871.
1872.
1873.
1874.
1875.
1876.
1877.
1878.
1879.
1889.
1881.
1882.
1883.
1884.
1885.

110

112

Year.

25 commodities.)

Import
and ex­
port
values.

1887.
1888.
1889.
1890.
1891.
1892.
1893
1894.
1895.
1896.
1897.
1898.
1899.
1900.
1901.
1902.

Market
prices.

104
108
108
109
111
105
103
95
93
94
93
95
101
114
107
104

107
110
110
111
111
103
104
94
94
93
91
95
105
117
106
104

i Wholesale and Retail Prices. Return to an Order of the . . . House of Commons . . . for ‘ ‘ Report
on Wholesale and Retail Prices in the United Kingdom in 1902, with Comparative Statistical Tables for a
Series of Years.” For Sauerbeck’s figures see his annual articles in the Journal of the Royal Statistical
Society. The list of commodities in question is as follows:

Commodity.

Bacon.........................
Barley........................
Coal............................
Coffee.........................
Copper.......................
Cotton........................
Flax...........................
Hides.........................
Iron, pig....................
Ju te...l......................
Lead...........................
Linseed......................
Maize.........................
Oats...........................
Oil, olive....................
Oil, palm...................
Petroleum.................
Rice...........................
Silk............................
Sugar, refined............
Tea.............................
T in.............................
Wheat........................
Wool..........................
Do...........................




Quotations given by
Board of Trade.
Average import values
....... d o ..........................
Average export values.
Average import values
.......d o..........................
.......d o..........................
.......d o..........................
.......d o..........................
Average export values.
Average import values
.......d o..........................
.......d o..........................
.......d o..........................
.......d o..........................
.......d o..........................
.......d o..........................
.......do..........................
.......d o..........................
.......d o..........................
.......d o..........................
.......d o..........................
.......d o..........................
.......do..........................
d o..........................
Average export values.

Brands quoted by Sauerbeck.

Waterford.
English Gazette.
Wallsend, Hetton, in London.
Rio, good channel.
Chile bars.
Middling American.
St. Petersburg.
RiverPlata, dry.
Good medium.
English pig.
Linseed.
American mixed.
English Gazette.
Olive oil.
Palm oil.
Petroleum, refined.
Rangoon, cargoes to arrive.
Tsatlee.
Java, floating cargoes.
Congou, common.
Straits.
English Gazette.
Merino, Adelaide, average grease,
English, Lincoln, half hogs.

j

THE MAKING AND USING OF INDEX NUMBERS.

33

4.—IN D E X NUMBERS MADE FROM THE M ARKET PRICES AND FROM THE IMPORT
AND E X P O R T VALUES OF IDENTICAL LISTS OF COMMODITIES. ENGLAND, 1871-1902.
(BASED ON TABLE 5.)

Ch art




34

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

4. RELATIVE VERSUS ACTUAL PRICES.

In February, 1864, Hunt’s Merchants’ Magazine published the fol­
lowing statement to show how rapidly prices rose after the suspension
of specie payments in December, 1861, and the issue of the irredeem­
able United States notes.1 These figures are the total prices of 55
articles quoted by their customary commercial units.

Value of 55 leading articles of New York commerce.
January, 1862................................................................................. $804
April, 1862.....................................................................................
844
January, 1863................................................................................. 1,312
March, 1863.................................................... ............................... 1,524
July, 1863......................................................................... ! ........... 1,324
October, 1863................................................................................. 1,455
January, 1864................................................................................. 1,693

For example, in January, 1862, coal oil is entered as 30 cents per
gallon and pig iron as $24 per ton; molasses is entered as 42£ cents
per gallon and whalebone as $69 per ton; oats is entered as 38 cents
per bushel and corn as $59.25 per hundred bushels, etc.2
Clearly, this simple method of measuring changes in the price level
by casting sums of actual prices is not trustworthy. For a relatively
slight fall in the quotation for whalebone would affect the total, as
Hunt’s Merchants’ Magazine computes it, much more than a rela­
tively enormous increase in the price of molasses. The fact that corn
happens to be quoted by the hundred bushels makes a 1 per cent
change from its price in January, 1862, equal to a 43 per cent change in
the price of wheat and to a 156 per cent change in the price of oats,
both of which are quoted by the bushel.
It was to avoid such patent absurdities that Carli threw his actual
prices of grain, wine, and olives in 1750 into the form of percentages
of rise or fall from their prices in 1500, and then struck the average
of the three percentages. When this operation is performed it makes
no difference whether the commodities are quoted by large or by
small units. The obvious common sense of this precedent has caused
it to be followed or reinvented by most makers of index numbers to
this day—with one slight modification. To avoid the awkwardness
of the plus and minus signs necessary to indicate whether prices have
advanced or receded, it is usual to substitute for percentages of rise
or fall relative prices on the scale of 100. For example, a rise of 10
per cent and a fall of 10 per cent are expressed by relatives of 110
and 90, respectively. Occasionally, however, percentages of rise or
fall are still used as by Carli; as, for instance, in the chain relatives
published by the Bureau of Labor Statistics in Bulletin No. 149 and




i Vol. 50, p. 132.

* See vol. 48, p. 129.

THE M AKING AND USING OF INDEX NUMBERS.

85

averaged in the first four tables of this bulletin. A second unim­
portant variant, long practiced by the Economist, but now seldom
used, is to publish as the final result the sums of relative prices,
instead of their averages.1
In recent years a few statisticians have gone back from the use of
relative to the use of actual prices, adopting various devices to avoid
such crude errors as those perpetrated in the figures cited from Hunt’s
Merchants’ Magazine. In 1897 Bradstreet’s began reducing all its origi­
nal quotations by the gallon, ton, dozen, square yard, etc., to prices
by the pound, and presenting as its index number the aggregate
prices per pound of 98 articles.2 Four years later, Dun’s Review
followed this lead, with an important difference. Instead of reducing
actual quotations to quotations by the pound, it multiplied the actual
quotation for each article included by the quantity of that article sup­
posed to be consumed in the course of a year by the average indi­
vidual. These products were then cast up, and the sums, in dollars
and cents, were presented as an index number purporting to show
the changes in the per capita cost of a year’s supplies.3
Still later (1912), the method practiced by Dun was adopted by
the Commonwealth statistician of Australia as the basis of his official
series. However, after he had calculated the aggregate expenditure
of Australians upon his bill of goods in terms of pounds sterling, he
threw these pecuniary sums back into the form of relative numbers
on the scale of 1,000.
Accordingly, three types of index numbers are now in general use:
(1) Averages of relative prices or average percentages of change in
prices; (2) sums in dollars and cents showing changes in the aggre­
gate cost of certain definite quantities of certain commodities; (3)
relative figures made from series of the second sort. The first type
shows average variations, the second type shows the variations of an
aggregate, the third type turns these variations of an aggregate into
percentages of the aggregate itself as it stood at some selected
time. Certain of the advantages and shortcomings of the several
types can be brought out in connection with the next topic, base
periods; but the relation between average variations of relative
prices and the variations of aggregate actual prices can not be ade­
quately treated until we reach the section devoted to forms of
averages.
1 Gibson’s index number is such a sum. See p. 94. The difference between sums of relative prices and
these sums divided by the number of articles included is, of course, purely formal. Averages have
displaced sums in current use mainly because it is easier to make comparisons on the scale of 100 than on
the scale of 2,200, or whatever number is given by the addition of relative prices.
2 For a criticism of this method, see p. 101.
8 The confidence merited by this index number is discussed in Section V.




36

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

5. BASE PERIODS.

When relative prices are used it is necessary to select the quota­
tions of some given period as a base. The actual prices in this base
period are called 100; all antecedent and subsequent prices are
divided by the base prices, and the quotients, multiplied by 100,
make the relatives which are usually summed and divided by the
number of commodities to get the final index number.
In some cases the prices of a single day have been used as the base,
but as a rule average prices for a year, five years, a decade, or an even
longer period have been preferred. For this preference there is a
simple justification when arithmetic means are used as averages of
the relative prices.1 If the price of any commodity happens to be
unusually high or unusually low in the base period, its relative prices
at other periods will be correspondingly high or low, and very high
relative prices, especially, may exercise an undue influence upon arith­
metic means. If an appreciable proportion of the commodities in the
list be very high or very low, the final index number may be distorted.
Though numerically correct, the results have less significance than if
they showed changes in terms of prices that men consider “ normal.” 2
Of course exceptionally high or exceptionally low quotations are less
likely to last for a year than for a day, and less likely to last for a
decade than for a year.
The period chosen as base should be that period with which accu­
rate comparisons are most significant for the purpose in hand. Prob­
ably most users of general-purpose index numbers prefer to make
their comparisons with recent dates. Hence the case for “ chain”
indexes is very strong—that is, for indexes like the medians of Table 2,
which show the average rise or fall of prices on the basis of prices in
the preceding year.3 Hence, also, any index number with a fixed
base becomes in one respect less significant the longer it is main­
tained. For example, when the Bureau of Labor Statistics’ series
was established in 1902, the public was interested to know how much
prices in that year had changed in terms of average prices in the
decade 1890-99. In 1915, however, we care less about a measure­
ment of change in terms of what prices were 16 to 25 years ago than we
care about how much prices have changed with reference to 1914.
Similarly, Sauerbeck’s index number suffers in significance now be­
cause it forces one to make all comparisons in terms of prices in a
i If geometric means are used the ratios between the index numbers for different dates are not influenced
at all by the selection of the base, and if medians are used they are likely to be affected but slightly, provided
the number of commodities included be large. See the discussion of forms of averages, subsec. 8.
* The selection of a proper base period, however, does not guarantee immunity from the exercise of undue
influence by certain articles. More important than the base is the choice of proper weights. See subsec. 7.
* This form of index number was invented by Prof. Alfred Marshall. See Contemporary Review, March,
1887




THE MAKING AND USING OF INDEX NUMBERS.

37

period that ended before most of the people now living were old
enough to know the meaning of prices.
A further advantage of chain index numbers is that they make the
dropping of obsolescent and the adding of new commodities espe­
cially easy. It is difficult to keep the list of commodities included in a
fixed-base system really representative of the markets over a long
period of time. Barring perhaps thirty or so staple raw materials that
hold their importance for centuries at a time, most commodities have
their day of favor and then yield to new products. Consequently
the compilers can hardly let two decades pass without revising
their lists, in certain details, or seeing them lose in significance.
But since a chain index does not profess to give accurate compari­
sons except between successive years the compiler feels himself free
to improve his list whenever he can. It is very much easier to in­
clude many commodities on this plan. And if the index number be
weighted, the chain index has a similar advantage in facilitating the
frequent revision of the weights./
Once more, year-to-year variations of prices can be measured with
a closer approach to accuracy than variations covering a longer
period of time. For, as was shown in Section III of this bulletin, the
former variations are highly concentrated about their central tend­
ency while the variations from what prices used to be years ago are
widely dispersed. The longer a fixed-base system is maintained,
indeed, the more scattered become the relative prices as a rule.
Hence the variations are less and less aptly represented by any
average that can be devised, and the margin of error to which the
results are subject grows wider. In other words, with a given body
of quotations to build upon, chain relatives are more trustworthy
than their rivals; and, as has just been said, it is feasible to provide
a larger body of quotations for chain relatives than for a fixed-base
series.
Finally, another aspect of the wide dispersion that becomes char­
acteristic of fixed-base relatives with the lapse of years merits sepa­
rate mention. The commodities that have a consistent long-period
trend gradually climb far above or fall far below the average relative
prices.2 Then the high relative prices of the first group come to
exercise much more influence upon the position of the average itself
than do the low relative prices of the second group.3 A 10 per cent
change in the price of an article whose price has already doubled will
count four times as much as a 10 per cent change in the price of an
article whose price has dropped by half. For most purposes, this
1 Compare Irving Fisher, The Purchasing Power of Money, revised edition, p. 204. On weighting, see
subsec. 7 of this bulletin.
2 Compare p. 23.
s Medians and geometric means are not distorted by such cases, as are the much commoner arithmetic
means. See subsec. 8.




38

BULLETIN OF THE BUKEAU OF LAB Oil STATISTICS.

development is to be regarded as a defect of the fixed-base series.
For commodities seldom gain in importance because of a great rise in'
price; on the contrary, the commodities that become cheaper are
likely to be consumed and produced on an increasing scale.1 Against
this danger of magnifying the influence of articles that are becoming
costly and minimizing the influence of articles that are becoming
cheap, no care in the selection of a base avails for long if the base be
fixed.2
Chain relatives have their drawbacks also. Makers of index num­
bers find them more laborious to compute than fixed-base series,
since most of the actual prices used as divisors change every year.
And users of index numbers find a chain series difficult to interpret
when they seek to know how much prices have risen or fallen over
considerable periods of time. Of course, chain relatives for succes­
sive years can be multiplied together to form a continuous series, but
it is not easy to give the later members of the series a concrete mean­
ing. To know, for example, that in 1891 prices fell, on the average,
0.2 per cent below their level in 1890; that in 1892 they fell 4.4 per
cent below their new level in 1891, and so on through ups and downs
on an ever-changing base for every year to 1915, enables one to make
a series beginning, say, with 100 in 1890 and running on with 99.8 in
1891, 95.4 in 1892, etc., to some result for 1915.3 But such a series
does not enable one to say in terms of what a comparison is made
between prices in 1915 and in 1890. Any fixed-base series covering
these years, on the contrary, would show the level of prices both in
1890 and in 1915 in terms of a common denominator—namely, the
level at which prices stood in the base period, whatever that was.
Hence it is an excellent plan to make from the original quotations
two series of index numbers—one a chain index and the other a
fixed-base series.
Even this combination, however, is far from meeting all the needs
of users of index numbers. For certain users may require for special
purposes accurate measurements of price fluctuations in terms of the
price level in any given month or year, or any given stretch of time
in the whole period covered by the investigation. If such users are
few as compared with all the people who note or quote the popular
index numbers, they are precisely the few most interested in price
fluctuations and most likely to increase knowledge by their use of the
figures. But of course compilers can not foresee what base periods
would serve best all these special purposes, and they can not be
1 Compare, A. W . Flux, “ Modes of constructing Index numbers,” Quarterly Journal of Economics,
August, 1907. Vol. 21, p. 615. See pp. 88 and 89 of this bulletin.
2 In other words, the objection stated on p. 36 to accepting as the base a period when prices were not
“ normal” extends in a measure to accepting any long-past period as a base for measuring current fluctua­
tions of prices.
* For an example of this method see pp. 42 to 44.




THE MAKING AND USING OF INDEX NUMBEKS.

expected to work out index numbers on all the bases made possible
by their original data. It is therefore highly desirable to have index
numbers that can be shifted from one base to another both readily
and accurately.
It is this desideratum, in large part, that has led to the recent
reaction against index numbers made by striking arithmetic means
of relative prices and in favor of index numbers made by adding
actual prices. For the latter form of index, being a sum of dollars
and cents, can be thrown into the form of a series of relative prices
upon any base that is desired, with slight labor and without inac­
curacy; whereas arithmetic means of relative prices can not con­
sistently be shifted to a new base without recomputing the relative
prices, commodity by commodity, and striking new averages from
these new relatives.1 Such recomputations are so laborious that a
short method of shifting the base of this kind of index numbers is
often practised even by persons quite aware of the ambiguity of the
new results. This method consists in dividing the figures for other
dates by the figures for the date desired as base and multiplying the
quotients by 100. Of course this process results in a relative price of
100 for the new base period, and the other figures look as if they showed
average relative prices as percentages of prices at this period. But
there is no mathematical justification for assuming that results reached
in this way must agree with results reached by recomputing relative
prices for each commodity on the new base. For such recomputation
usually alters considerably the relative influence exercised upon the
arithmetic means by the price fluctuations of certain commodities.
Those articles which are cheaper in the new than in the old base period
get higher relative prices and therefore increased influence. Vice
versa, articles that are dearer in the new base period get lower rela­
tive prices and therefore diminished influence. Of course the short
method of shifting the base, which retains the old relative prices,
does not permit any such alteration in the influence exercised by the
fluctuations of different commodities. Hence the two methods of
shifting the base seldom yield precisely the same results. To present
a series of arithmetic means shifted by the short method as showing
what the index numbers would have been if they had been computed
upon the new base is therefore misleading.
i On the forms of index numbers that can be shifted from one base to another without altering the ratios
between the successive members of the series, see Irving Fisher, Purchasing Power of Money, table opposite
p. 418, test 7.




40

BULLETIN

OF THE BUREAU OF LABOR STATISTICS.

It is easy to arrange examples in which wide discrepancies appear
between the results of the two methods of shifting the base.1 But
the difficult and the important thing is to find out how serious the
discrepancies are in actual practice. For to use index numbers
effectively, it is often necessary to shift the base, and sometimes the
short method must be followed, either because recomputation in full
requires a prohibitive amount of labor, or because the original data
necessary for recomputation have not been published. The next
table gives three pertinent examples. In the first case when Sauer­
beck’s index is shifted from 1867-1877 = 100 to 1890-1899 = 100 the
discrepancies are fairly regular and rather small both absolutely and
relatively. In the last case, when the same series is shifted to 1860 =
100, the discrepancies are highly irregular from year to year, and are
rather large both absolutely and relatively—several times exceeding
5 per cent of the recomputed figures. In the remaining case the
discrepancies are small absolutely, though often large relatively to
the recomputed figures, and also highly variable from year to year.2
The conclusion which these experiments suggest is that the two
i For example, suppose that an index number includes only wheat and corn, and that their prices are as
follows:
__________________________________________
1913
Wheat, per bushel. . .
Com, per bushel........

$1.00
.40

1914

|

$0.50
.40

If 1913 be made the base, the relative prices and index numbers will be:
1913

1914

Wheat, relative prices
Com, relative prices..

100
100

50
100

Sums...................
Index numbers..........

200
100

150
75

If now the base be shifted from 1913 to 1914 by the short method, the index number for 1913 willbe (100-*-75)
100= 1 3 3 But if the figures be recomputed on the basis of prices in 1914, the result is an index number of
150 in 1913:
________________________________________
1913

1914

Wheat, relative prices
Com, relative prices..

200
100

100
100

Sums...................
Index numbers..........

300
150

200
100

2 The discrepancies shown in the table do not result wholly from the mathematical inconsistency of the
short method; but partly from the fact that when an index number is shifted to a new base by recomputa­
tion in full it is commonly impossible or undesirable to utilize all the original data. Some commodity,for
example, may not be quoted for the dates used as the new base, and therefore has either to be dropped or
Introduced at a later date b y means of some doubtful assumption as to what its price would have been had
it been quoted for the full period. Of course this observation makes the objection to using the short method
stronger rather than weaker. It means that this method often leads the statistician into uses of the original
data which he would have avoided had he undertaken the recomputation of the index number.




41

THE MAKING AND USING OF INDEX NUMBERS.

methods almost always give different results; that the discrepancies
are by no means constant from year to year in a given case, and that
their magnitude both absolutely and relatively differs much from one
case to another. Hence it is well to avoid the short method of
shifting bases whenever possible; and when that method must be
used, its results should not be treated as showing what the index
number would have been had it been made originally on the new base.
T able

6.-—EXAM PLES OF DISCREPANCIES BETW EEN THE RESULTS OF TWO METHODS
OF SHIFTING THE BASES ON WHICH IN DEX NUMBERS ARE COMPUTED.
(Arithmetic means.)

Sauerbeck’s index number,
1890-1913.

Bureau of Labor Statistics’
index number.

Sauerbeck’s index number,
1860-1891.

Recom­
Bu­
Shifted
Shifted puted
Chain Chain
Years. Origi­
Years. Origi­
reau’s index
to
to
index
on
basis Dis- series
nal
nal
Dis­
on
18901860
form, =100,
form, 1899= 1890- crep- basis made made crepan­
by
186718671899= an- 1890by
cies.
by 100,
by
long
by cies. 1899-= short
1877= 100,
1877= short
short
method.
method.
100. method.
long
100. method.
100.
method.
1890..
1891..
1892..
1893..
1894..
1895..
1896..
1897..
1898..
1899..
1900..
1901..
1902..
1903..
1904..
1905..
1906..
1907..
1908..
1909..
1910..
1911..
1912..
1913..

72
72
68
68
63
62
61
62
64
68
75
70
69
69
70
72
77
80
73
74
78
80
85
85

109
109
103
103
95
94
92
94
97
103
114
106
105
105
106
109
117
121
111
112
118
121
129
129

109
109
103
103
95
94
9i‘
93
97
104
115
107
106
106
108
111
119
123
112
114
120
123
130
130

1
1
1
1
1
1
2
2
2
2
1
2
2
2
1
1

112.9
111.7
106.1
105.6
96.1
93.6
90.4
89.7
93.4
101.7
110.5
108.5
112.9
113.6
113.0
115.9
122.5
129.5
122.8
126.5
131. G
129.2
133.6
135.2

-1 .1
- 5 .0
- .5
- 9 .0
- 2 .6
-3 .4
- .8
+4.1
+8.9
+8.7
- 1 .8
+4.1
+ .6
- .5
+2.6
+5.7
+5.7
- 5 .2
+3.0
+4.0
-1 .8
+3.4
+1.2

- 0.2
- 4.4
- .2
- 8.7
- 1.5
- 2.8
+ •2
+ 4.8
+10.4
+ 9.4- 1.1
+ 4.6
+ 1.2
- .1
+ 2.9
+ 5.8
+ 6.0
- 5.6
+ 3.2
+ 4.1
- 1.9
+ 3.4
+ 1.2

0.9
.6
.3
.3
1.1
.6
.4
.7
1.5
.7
.7
.5
.6
.4
.3
.1
.3
.4
.2
.1
.1

1860.
1861.
1862.
1863.
1864.
1865.
1866.
1867.
1868.
1869.
1870.
1871.
1872.
1873.
1874.
1875.
1876.
1877.
1878.
1879.
1880.
1881.
1882.
1883.
1884.
1885.
1886.
1887.
1888.
1889.
1890.
1891.

99
98
101
103
105
101
102
100
99
98
96
100
109
111
102
96
95
94
87
83
88
85
84
82
76
72
69
68
70
72
72
72

Re­
com­
puted
on
basis
1860=
100.

Discrepancies.

100.0 100.0
99.0
99.6 *0. 6
102.0 105.5 3.5
104.0 109.3
5.3
106.1 112.3
6.2
102.0 105.8
3.8
103.0 106.5
3.5
101.0 103.9
2.9
100.0 103.1
3.1
99.0 101.9
2.9
97.0 100.3
3.3
101.0 102.6
1.6
110.1 112.5
2.4
112.1 116.6
4.5
103.0 107.0
4.0
97.0 100.3
3.3
96.0
97.5
1.5
95.0
97.4
2.4
91.2
87.9
3.3
83.8
86.7
2.9
2.9
88.9
91.8
85.9
88.5
2.6
88.0
3.1
84.9
86.0
3.2
82.8
2.5
76.8
79.3
75.4
72.7
2. 7
72.4
69.7
2.7
2.0
68.7
70.7
70.7
73.9
3.2
72.7
76.7
4.0
76.0 3.3
72.7
72.7
75.4
2.7

The second of the preceding examples of discrepancies arising from
the two ways of shifting bases merits especial attention because it
refers to the new and important chain index number published by the
Bureau of Labor Statistics in Bulletin No. 149. All of the “ per­
centages of increase or decrease compared with each preceding year or
month ” on pages 9 to 16 of Bulletin No. 149 were made by dividing
the 1890-1899 index number for each date by the corresponding
index number for the preceding date. Consequently these results
are not precisely what the captions, under which they appear, suggest.



42

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

The fact that the discrepancies between the two sets of results are
small, never exceeding 1.5 points in the scale of percentage changes,
affords striking confirmation of a conclusion drawn in Section III
from the distribution of price variations. Because variations from
prices in the preceding year are highly concentrated about a central
tendency, while variations from the prices of a remoter period are
widely scattered, it was argued that the measurement of price changes
is easy in proportion as the time during which these changes have
been accumulating is short. So, now, we find that dissimilar methods
of manipulating the same data yield nearly the same results when they
are applied to the easy problem of making a chain index number.
The use of the short method in making the new chain indexes was
the natural result of a practice begun by the bureau in 1908— a
practice that illustrates, from another angle, the problem of shifting
bases. In that year 11 new commodities were introduced into the
bureau’s index number.1 Since quotations were not secured for
these commodities prior to 1907, relative prices could not be com­
puted for them on the 1890-1899 base. How, then, could these
new articles be included in the index numbers of the groups affected ?
The bureau solved this problem by (1) computing relative prices
for both the new and the old commodities in 1908 on the basis, prices
in 1907=100, (2) averaging these relatives, and (3) shifting these
new index numbers for 1908 from the 1907 base to the 1890-1899
base. This shift was effected by multiplying the group index num­
bers for 1908, on the 1907 base, by the group index numbers for 1907,
on the 1890-1899 base.
The process may be illustrated from the group of farm products.
The index number for this group in 1907, on the base, prices in 18901899 = 100, was 137.1, while in 1908, on the base, prices in 1907 = 100,
the index number was 97.1. The bureau assumed that since prices of
farm products in 1908 were, on the average, 97.1 per cent of their
prices in 1907, and since their prices in 1907 had been 137.1 per cent
of their prices in 1890-1899, therefore prices in 1908 were 97.1 per
cent of 137.1 per cent of prices in 1890-1899; that is, 133.1 per cent
of prices in 1890-1899. By repeating this process in later years, the
bureau forged its successive chain indexes from 1908 to 1913 into a
continuous series.
The merits and defects of series made in this fashion have already
been canvassed.2 The one fact important for present purposes is
that the results of this method, however excellent in other ways, do
not agree with results worked out on a fixed-base system. Hence the
i.Among farm products, horses, mules, poultry, and leaf tobacco; among food products, canned com,
canned peas, canned tomatoes, fresh beef in Chicago, dressed poultry, and cabbage; among lumber and
building materials, yellow-pine flooring.
a See pp. 36-38.




THE MAKING AND USING OF INDEX N UMBERS.

48

bureau’s present index numbers of farm products, foods, and lum­
ber and building materials—the three groups into which new com­
modities were introduced—are not accurately comparable in 19081913 either with its figures for these same groups in 1890 to 1907, or
with its 1908-1913 figures for the six remaining groups. Even the
general index number of all commodities is affected appreciably by
the admission of the inconsistent elements into the grand totals from
which averages were struck.
The question remains: How much difference did the change in
method make ? To answer one must find a better way of introducing
the 11 new commodities into the old list. To effect this introduction,
some assumption is necessary concerning the relation of their prices
in 1907, the first year for which quotations were obtained, to their
unknown prices in the base period. Perhaps the bureau’s implicit
assumption that the 1907 index number of the old commodities on
the 1890-1899 base should be taken as the starting point for comput­
ing relative prices for the new commodities is as defensible as any
other guess that can be made. If this guess be accepted, the relative
prices for the four new farm products should be computed on the base,
actual prices in 1907 = 137.1; for the six new foods on the base, actual
prices in 1907 = 117.8, and for the one new kind of lumber on the base,
actual prices in 1907 = 146.9. Then the relatives for the new com­
modities can be added to and averaged with the relatives for the old
without more ado. If this method be applied to farm products the
result is an index number for 1908 of 133.4/ whereas the bureau’s
method gives 133.1. In the case of foods we get 121.5 instead of the
bureau’s 120.6, and in the case of lumber and building materials 131.8
instead of 133.1.
Now the discrepancies between these two sets of results for 1908
seem small. But the bureau soon made them large, by building its
index for 1909 on the discrepancy for 1908, again building in 1910 on
the discrepancy accumulated in 1908 and 1909, and so on. By 1913
1 The details of this new computation are as follows:
Relative
prices in
1908 on
base, actual
prices in
1907=137.1.

New farm products.

1907

1908

Horses.....................................................................................................
Mules......................................................................................................
Poultry...................................................................................................
Leaf tobacco...........................................................................................

$211.17
203.80
.1409
12.4381

$196.18
189.13
.1327
15.0625

127.4
127.2
129.1
166.0

Sum nf rAlativAS for t.h« 4 new commodities______________________
811m of relatives for the 16 old commodities.. .......... .................................. ................................

549.7
2,117.4

A ??re?ate relative nricea of 20 farm nrodncts_____________________
Average relative prices of farm products.......................................................................................

2,667.1
133.4




44

BULLETIN OF THE BUKEAU OF LABOK STATISTICS.

the results of the two methods for farm products are far apart; the
two figures for food products and for lumber and building materials
are seriously at variance, and even the general index numbers for all
commodities show a difference of 4.2 points. Table 7 presents the
two sets of results side by side.1
T able

7.—EFFECT OF SHIFTING THE BASE OF IN D E X NUMBERS B Y SIMPLE M ULTI­
PLICATION.

[Bureau of Labor Statistics’ index numbers for farm products, foods, lumber and building materials, and
all commodities after the inclusion of 11 new commodities in 1908.]
(Arithmetic means.)

Farm products.

Years.

Lumber and building
materials.

Food products.

All commodities.

Index
numbers
on base,
prices in
preceding
year= 100,
multi­
plied by
index
numbers
based on
1890-1899
= 100.

Base for
old com­
modities,
1890-1899
= 100.
Base for
new com­
modities,
prices in
1907=
index
number of
old com­
modities
in that
year.

Index
numbers
on base,
prices in
preceding
year= 100,
multi­
plied by
index
numbers
based on
1890-1899
= 100.

Base for
old com­
modities,
1890-1899
= 100.
Base for
new com­
modities,
prices in
1907=
index
number of
old com­
modities
in that
year.

Index
numbers
on base,
prices in
preceding
year=100,
multi­
plied by
index
numbers
based on
1890-1899
= 100.

Base for
old com­
modities,
1890-1899
= 100.
Base for
new com­
modities,
prices in
1907=
index
number of
old com­
modities
in that
year.

Index
numbers
on base,
prices in
preceding
year= 100,
multi­
plied by
index
numbers
based on
1890-1899
= 100.

Base for
old com­
modities,
1890-1899
= 100.
Base for
new com­
modities,
prices in
1907=
index
numbers of
their
groups in
that year.

133.1
153.1
164.6
162.0
171.3
165.8

133.4
150.2
159.8
156.3
162.6
153.8

120.6
124.7
128.7
131.3
139.5
137.1

121.5
126.5
129.0
126.4
134.2
131.9

133.1
138.4
153.2
151.4
148.2
151.8

131.8
136.4
151.2
152.1
148.2
147.1

122.8
126.5
131.6
129.2
133.6
135.2

122.0
125.5
130.5
126.8
130.5
131.0

1908...............
1909...............
1910...............
1911...............
1912...............
1913...............

6. THE NUMBERS AND KINDS OP COMMODITIES INCLUDED.

Since the earlier makers of index numbers had to use such price
quotations as they could find, the problems how many and what
kinds of commodities to include were practically solved for them.
As Prof. Edgeworth remarks, “ Beggars can not be choosers.”
i The revised figures should be substituted for the bureau’s figures for 1908-1913, or the 11 new commodi­
ties should be dropped altogether. The latter course has been followed in other tables in this paper, but it
makes little difference which set of results is used, as the following figures show:
Recomputation of the Bureau of Labor Statistics’ index number for all commodities in the years 1908 to 1918.

After revising the method of treating the 11 new com­
modities........................................................................
After dropping the 11 new commodities altogether.........




1908

1909

1910

1911

1912

122.0
121.7

125.5
124.5

130.5
130.0

126.8
126.4

130.5
130.3

1913

131.0
130.4

THE MAKING AND USING OF INDEX NUMBERS.

45

Paucity of data still hampers contemporary efforts to measure
variations of prices in the past; but the compilers of index numbers
for current years have a wider range of choice. The scope of their
data is limited not by the impossibility but by the expense of col­
lecting quotations. And in the case of governmental bureaus or
financial journals the limits set by expense are neither narrow nor
rigid. Such organizations can choose many commodities if they will
or content themselves with few.
One principle of choice is generally recognized. Those commodities
are preferable that are substantially uniform from market to market
and from year to year. Often the form of quotation makes all the
difference between a substantially uniform and a highly variable com­
modity. For example, prices of cattle and hogs are more significant
than prices of horses and mules, because the prices of cattle and hogs
are quoted per pound, while the prices of horses and mules are quoted
per head.
It is often argued that the application of this common-sense prin­
ciple rules out almost all manufactured goods, because such articles
are continually being altered in quality to suit the technical exigencies
of new industrial processes or the varying tastes of consumers. But
minor changes in quality, provided their occurrence is known, do not
necessarily unfit a commodity for inclusion. When the brand for­
merly sold is replaced by a variant it is usually possible to get over­
lapping quotations for the old and new qualities during the time of
transition. Then the new series may be spliced upon the old by
means of the ratio borne by the price of the new grade to the price of
the old grade in the years when the substitution is made. Statis­
ticians willing to take the extra precautions and trouble involved by
such operations can legitimately include not only a large number of
staple raw materials and their simplest products, but also an even
larger number of manufactured goods.
Some of the modem index numbers, accordingly, have long lists of
commodities. Dun’s index number seems to be built up from 310
series of quotations, the official Canadian index number includes 272
articles, the Bureau of Labor Statistics’ index number for 1913 had
25?, and the new weighted index number for 1914 contains 297 quota­
tions of 201 distinct articles. On the other hand, many of the bestknown index numbers use less than 50 series of quotations. Forty-five
is a favorite number, largely because of the high reputation early
established by Sauerbeck’s English series. The British Board of
Trade’s series, the new official French series, the New Zealand series,
Von Jankovich’s Austrian series, and Atkinson’s series for British India




46

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

all have just 45 commodities, while the new series of the London
Economist and the relative prices published by the Imperial Statis­
tical Office of Germany include 44 articles. Even shorter lists are
often used. For example, Schmitz’s German series has only 29 com­
modities, the New York Annalist series 25, and Gibson’s series 22.
Private investigators working with limited resources sometimes con­
fine themselves to a bare dozen commodities, or even less.1
These differences of practice raise important questions of theory.
Does it make any substantial difference in the results whether 25
or 50 or 250 commodities be included—provided always that the
lists be well chosen in the three cases ? If differences do appear in
the results, are they merely haphazard, or are they significant differ­
ences? If there are significant differences, which set of results is
more valuable, that made from the long or from the short lists?
And what does the proviso that the lists be well-chosen mean ? In
short, do the index numbers including hundreds of commodities pos­
sess any real advantages over those including 50 or 25 to compensate
for the greater trouble and expense of compiling them ?
The best way to answer these questions is to experiment with large
and small index numbers, made on a strictly uniform plan for the
same country and the same years. Table 8 presents six such index
numbers which differ only in respect to the number and kind of
commodities included. The first column includes all the commod­
ities quoted by the Bureau of Labor Statistics, except the 11 whose
prices do not run back of 1908.2 Many of the commodities in this
list are merely different varieties of the same article; for example,
there are two kinds of corn meal, four kinds of leather, six kinds of
women’s dress goods, eleven kinds of steel tools, etc. The second
column gives an index number in which all such groups are repre­
sented by single averages, so that the number of series which enter
1 These statements refer to the number of series of relative prices averaged to get the final results as now
presented. Often two or more different varieties of an important article are counted as separate com­
modities, and, on the other hand, the relative prices of slightly different articles are sometimes averaged
to make one of the series which enters into the final averages. In view of the diversity of practice in this
respect, a perfectly consistent counting of the number of distinct “ commodities” included in the general
series is impossible. Moreover, the figures are often published with such imperfect explanations as to
make the counting of the commodities included doubtful or impossible on any interpretation of that term.
2 To facilitate comparison, decimals have been dropped and the index for each year rounded off to the
nearest whole number. Further, the results for 1908-1913 are changed for the reasons explained on pp.
42-44. Regarding the changes in the number of commodities included, see Bulletin No. 149, p. 11.




THE M AKING AND USING OF INDEX NUMBERS.

47

directly into the final results is cut down to 145.1 The third column,
which includes 50 commodities, is made up from the list adopted for
the Gibson index number in its original form.2 The fourth series is
i This experimental list of 145 commodities is given below. When the relative prices of closely related
articles are averaged to make a single series, the number of these articles quoted by the bureau and in­
cluded in the group is indicated. Most of the bureau’s series which do not cover the whole period, 18901913, are dropped altogether. As the basis of a general-purpose index number, this revised list is worse
than the bureau’s list in certain respects and better in others. See Section V.
FARM PRODUCTS.

1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.

Barley.
Cattle, 2.
Com.
Cotton.
Flaxseed.
Hay.
Hides.
Hogs, 2.
Hops.
Oats.
Rye.
Sheep, 2.
Wheat.
FUEL AND LIGHTING.

1.
2.
3.
4.
5.
6.
7.

Candles.
Coal, anthracite, 4.
Coal, bituminous, 3.
Coke.
Matches.
Petroleum, crude.
Petroleum, refined, 2.
FOOD, ETC.

1. Apples, evaporated.
2. Beans.
3. Bread, crackers, 2.
4. Bread, loaf, 3.
5. Butter, 3.
6. Cheese.
7. Coffee.
8. Currants.
9. Eggs.
10. Fish, 4.
11. Flour, buckwheat.
12. Flour, rye.
13. Flour, wheat.
14. Lard.
15. Meal, com, 2.
16. Meat, beef, 3.
17. Meat, pork, 4.
18. Meat, mutton.
19. Milk.
20. Molasses.
21. Onions.
22. Potatoes.
23. Prunes.
24. Raisins.
25. Rice.
26. Salt.
27. Soda.
28. Spice, pepper.
29. Starch, com.
30. Sugar, 3.
31. Tallow.
32. Tea.
33. Vinegar.

CLOTHS AND CLOTHING.

1. Bags.
2. Blankets, 3.
3. Boots ana shoes, 3.
4. Broadcloths.
5. Calico.
6. Carpets, 3.
7. Cotton flannels, 2.
8. Cotton thread.
9. Cotton yarns, 2.
10. Denims.
11. Drillings, 2.
12. Flannels.
13. Ginghams, 2.
14. Horse blankets.
15. Hose.
16. Leather, 4.
17. Linen tnread.
18. Overcoatings, 2.
19. Print cloths,
20. Sheetings, 7.
21. Shirtings, 5.
22. Silk, 2.
23. Suitings.
24. Tickings.
25. Underwear, 2.
26. Women’s dress goods, 6.
27. Wool, 2.
28. Worsted yams, 2.
METALS AND IMPLEMENTS.

1. Bar iron, 2.
2. Barb wire.
3. Builders’ hardware, 3.
4. Copper, ingot.
5. Copper, wire.
6. Lead, i>ig.
7. Lead pipe.
8. Nails, 2.
9. Pig iron, 4.
10. Quicksilver.
11. Silver.
12. Spelter.
13. Steel billets.
14. Steel rails.
15. Tin. pig.
16. Tools, 11.
17. Wood screws.
18. Zinc.

LUMBER AND BUILDING MATE­
RIALS.

1. Brick.
2. Carbonate of lead.
3. Cement.
4. Doors.
5. Hemlock.
6. Lime.
7. Linseed oil.
8. Maple.
9. Oak, 2.
10. Oxide of zinc.
11. Pine, white, 2.
12. Pine, yellow.
13. Plate glass, 2.
14. Poplar.
15. Putty.
16. Rosin.
17. Shingles, 2.
18. Spruce.
19. Tar.
20. Turpentine.
21. Window glass, 2.
HOUSE-FURNISHING GOODS.

1.
2.
3.
4.
5.

Earthenware, 3.
Furniture, 4.
Glassware, 3.
Table cutlery, 2.
Woodenware, 2.

1.
2.
3.
4.
5.
6.

Cottonseed meal.
Cottonseed oil.
Jute.
Malt.
Paper, 2.
Proof spirit.
Rope.
Rubber.
Soap.
Starch, laundry.
Tobacco, 2.

MISCELLANEOUS.

7.

8.
9.
10.
11.

DRUGS AND CHEMICALS.

1.
2.
3.
4.
5.
6.
7.
8.
9.

Alcohol, grain.
Alcohol, wood.
Alum.
Brimstone.
Glycerine.
Muriatic acid.
Opium.
Quinine.
Sulphuric acid.

2
The list is as follows: Wheat, wheat flour (two kinds), barley, oats, com, corn meal, potatoes, rye,
sugar 89°, sugar 96°, coffee, tea, stoers, fresh beef, salt beef, sheep, mutton, hogs, bacon, hams, butter,
cotton, cotton yarns (two kinds), jute, wool (two kinds), worsted yams, raw silk (two kinds), pig iron,
bar iron, cement, copper ingots, copper sheets, lead, anthracito coal, bituminous coal (two kinds), hides,
leather, cottonseed oil, linseed oil, petroleum (crude and refined), rubber, spruce lumber, yellow-pine
lumber, and paper. See J. P. Norton, “ A revised index number for measuring the rise in prices,” Quar­
terly Journal of Economics, August, 1910, vol. 24, pp. 750-758.




48

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

made from the prices of 20 pairs, each commodity being given in two
forms, raw and manufactured, e. g., barley and malt, cattle and beef,
copper ingots and copper wire, etc.1 The last two columns contain
index numbers each made from the prices of 25 important articles
selected at random, the two lists having no items in common.2
T

able

SIX IN D E X NUMBERS FOR THE UNITED STATES MADE FROM QUOTATIONS
FOR DIFFE R E N T NUMBERS OF COMMODITIES, B Y YEARS, 1890 TO 1913.

8 .—

[Data from the Bulletin of the Bureau of Labor Statistics, No. 149.]
(Arithmetic means.

Year.

1903...............................................................

Average prices in 1890-1899=100.)

242 to 261
25 com­ 25 com­
commod­ 145 com­ 50 com­ 40 com­ modities, modities,
modities.
modities.
modities.
ities.
first list. second
list.
113
112
106
106
96
94
90
90
93
102
111
109
113
114
113
116
123
130
122
125
130
126
130
130

114
113
106
105
96
93
89
89
93
103
111
no
114
114
114
116
122
130
121
124
131
130
134
131

114
114
105
105
94
94
87
89
95
103
112
109
116
115
116
118
123
132
125
132
135
129
138
138

•

113
114
105
101
93
95
88
89
95
108
115
116
122
118
118
122
128
138
129
135
141
135
142
139

115
112
103
103
. 92
95
88
90
96
107
113
111
116
118
122
123
130
132
124
133
133
129
140
142

113
118
112
107
96
93
85
84
90
103
109
107
117
117
110
115
122
132
122
128
134
131
138
133

Averages 1890-1899.......................................
1900-1909.......................................
1910-1913......................................
Number of points by which prices rose ( + )
or fell (—) in—
1890-1896................................................
1896-1907................................................
1907-1908................................................
1908-1912................................................

100
118
129

100
118
132

100
120
135

100
124
139

100
122
136

100
118
134

-2 3
+40
- 8
+ 8

-2 5
+41
- 9
+13

-2 7
+45
- 7
+13

-2 5
+50
- 9
+13

-2 7
+44
- 8
+16

-2 8
+47
-1 0
+16

Difference between highest and lowest rel­
ative prices...............................................

40

45

51

54

54

54

Average change from year to year..............

I
4.0

4.1

4.9 |

5.5

5.0

6.2

1 The remaining 17 pairs are corn and corn meal, cotton and cotton textiles, flaxseed and linseed oil,
window glass and glassware, hides and leather, hogs and pork, lead (pig) and lead pipe, milk and cheese,
petroleum (crude and refined), pig iron and nails, pine boards and pine doors, rye and rye flour, sheep and
mutton, spelter and zinc, steel billets and steel tools, wheat and wheat flour, wool and woolen textiles.
2 The first list includes cotton, com, wheat, hides, cattle, hogs, coffee, wheat flour, salt, sugar, tea, pota­
toes, wool, silk, anthracite coal, bituminous coal, crude petroleum, pig iron, steel billets, copper ingots,
lead (pig), brick, average of nine kinds of lumber, jute, and rubber.
The second list includes hay, oats, rye, eggs, sheep, lard, beans, corn meal, butter, rice, milk, prunes,
cotton yarns, worsted yarns, coke, cement (Rosendale 1890-1899, Portland domestic 1900-1913), tallow,
spelter, bar iron, tin (pig), quicksilver, lime, tar, paper, proof spirit.




49

THE M AKING AND USING OF INDEX NUMBERS.

8.—SIX IN D E X NUMBERS FOR THE UNITED STATES MADE FROM QUOTATIONS
F OR D IF FE R E N T NUMBERS OF COMMODITIES, B Y YEARS, 1890 TO 1913—Concluded.

table

Number o f points by which the selected index numbers were greater (+ ) or less (— ) than the
Bureau of Labor Statistics* series•

25 com­
145 com­ 50 com­ 40 com­ 25 com­ modities,
modities. modities. modities. modities,
first list. second
list.

Year.

+ 1
+ 1
± 0
1
± 0
1
_ 1
_ 1
± 0
+ 1
± 0
+ 1
+ 1
± 0
+ 1
± 0
1
± 0
1
_ 1
+ 1
+ 4
+ 4
+ 1

+ 1
+ 2
- 1
- 1
—2
± 0
- 3
- 1
+ 2
+ 1
+ 1
± 0
+ 3
+ 1
+ 3
+ 2
± 0
+ 2
+ 3
+ 7
+ 5
+ 3
+ 8
+ 8

Arithmetic sums...........................................................
Algebraic sums..............................................................
Average differences computed from the—
Arithmetic sums....................................................
Algebraic sums.......................................................

23
4- 9

Maximum differences...................................................

+ 4
4- 0

1890.................................................................................
1891................................................................................
1892................... t ...........................................................
1893......................................................... .......................
1894................................................................................
1895................................................................................
1893................................................................................
1897................................................................................
1898 ................................................................. ............
1899................................................................................
1900 ...............................................................................
1901................................................................................
1902................................................................................
1903................................................................................
1904................................................................................
1905................................................................................
1907................................................................................
1908................................................................................
ig09................................................................................
1910................................................................................
1911................................................................... ............
1912................................................................................
1913................................................................................

Minimum rjiffp.rpmnfts.................................... .........................

+

1.0
.4

±
+

0
2
1
5
3
1
2
1
2
6
4
7
9
4
5
6
5
8
7
10
11
9
12
9

+ 2
db 0
- 3
- 3
—4
+ 1
—2
± 0
+ 3
+ 5
+ 2
+ 2
+ 3
+ 4
+ 9
+ 7
+ 7
+ 2
+ 2
+ 8
+ 3
+ 3
+10
+12

0
6
6
1
0
— 1
—5
—6
—3
+ 1
—2
—2
+ 4
+ 3
- 3
- 1
— 1
+ 2
± 0
+ 3
4
+ 5
+ 8
+ 3

60
+44

129
4-105

97
+73

70
+22

2.5
+ 1.8-

+

4.0
+ 3.0

+

+ 8
± o

4- 12
± 0

+12
± 0

+ 8
± 0

_

+
_
+
+
+
+
+
+
+
+
+
+

4+
4+

+
+

5.4
4.4

±
+
+
+
±

2.9
.9

Number o f points by which each index number rose (+ ) or fell (—) in each successive year.

Year.

1891...............................................................
1892...............................................................
1893...............................................................
1894...............................................................
1895...............................................................
1896...............................................................
1897...............................................................
1898...............................................................
1899...............................................................
1900...............................................................
1901...............................................................
1902...............................................................
1903...............................................................
1904...............................................................
1905...............................................................
1906...............................................................
1907.................- ............................................
1908...............................................................
1909...............................................................
1910...............................................................
1911...............................................................
1912.....................................................................
1913.....................................................................

94261°—Bull. 173—15— 4




25 com­
242 to 261 145 com­ 50 com­ 40 com­ 25 com­ modities,
commod­ modities.
modities. modities. modities,
first list. second
ities.
list.
—l
- 6
± 0
-1 0
- 2
- 4
db 0
+ 3
+ 9
+ 9
- 2
+ 4
+ 1
- 1
+ 3
+ 7
+ 7
- 8
+ 3
+ 5
—4
+ 4
± 0

—i
- 7
- 1
—9
- 3
—4
± 0
+ 4
+10
+ 8
- 1
+ 4
± 0
± 0
+ 2
+ 6
+ 8
- 9
+ 3
+ 7
- 1
+ 4
- 3

± 0
- 9
± 0
-1 1
db 0
- 7
+ 2
+ 6
+ 8
+ 9
- 3
+ 7
- 1
+ 1
+ 2
+ 5
+ 9
- 7
+ 7
+ 3
- 6
+ 9
± 0

+ 1
- 9
- 4
- 8
+ 2
- 7
+ 1
+ 6
+13
+ 7
+ 1
+ 6
- 4
db 0
+ 4
+ 6
+10
- 9
+ 6
+ 6
- 6
+ 7
- 3

- 3
- 9
db 0
-1 1
+ 3
- 7
+ 2
+ 6
+11
+ 6
- 2
+ 5
+ 2
+ 4
+ 1
+ 7
+ 2
- 8
+ 9
± o
- 4
+11
+ 2

+ 5
- 6
- 5
-1 1
- 3
- 8
- 1
+ 6
+13
+ 6
- 2
+10
± 0
- 7
+ 5
+ 7
+10
-1 0
+ 6
+ 6
- 3
+ 7
- 5

50

BULLETIN

OF THE BUREAU OF LABOR STATISTICS.

Now, these six index numbers, large and small, certainly have a
strong family likeness. The great movements of American prices
since 1890 stand out boldly in them all—the heavy fall of prices in
1890-1896, the distinctly greater rise in 1896-1907, the sharp decline
in 1908, the recovery in 1909, and the wavering course in 1910-1913.
Chart

5:—GENERAL-PURPOSE IN D E X NUMBERS, INCLUDING 25, 50, AND 242 COMMODI­
TIES. (BASED ON TABLE 8.)

If index numbers could pretend to nothing more than to show roughly
the trend of price fluctuations, then it would indeed matter little
which of these series were used. Either of the sets including only
25 commodities would serve that limited purpose as well as the set
containing nearly ten times as many commodities, though doubtless
the longer lists would command more confidence.




THE MAKING AND USING OF INDEX NUMBERS.

51

But the very success with which index numbers, even when made
from scanty and dissimilar data, bring out the broader features of
price movements encourages one to hope, from this device, for more
than an indication of the direction and a rough approximation to the
degree of change. Instead of concluding that an easy compilation,
based on a few series of quotations “ will do,” we may hope that
careful work covering a wide field will enable us to improve upon our
first results and attain measurements that have a narrow margin of
error.
When we make these more exacting demands upon our six index
numbers we attach importance to the fact that their general similarity
does not preclude numerous differences of detail. For example, two
series indicate that prices rose in 1891, one indicates that prices did
not change, and three indicate a fall; three put the lowest point in
1896, one in 1897, and two make the price level the same in these
years; one series shows a rise in 1901, five show a fall; in 1913 again
one series indicates a rise of prices, three indicate a fall, and two indi­
cate no change; the general level of prices in the final year is made
to vary between an average rise of 30 per cent and one of 42 per cent
above the level of 1890-1899; there is also a difference in steadiness,
the small series fluctuating through a wider range than the large ones,
etc.
To what are these discrepancies due ? Are they discreditable to
the large series, or to the small ones, or to neither set ? Can they be
accounted for except as the results of random differences in sampling ?
If an index number made from the wholesale prices of 25, or 50,
or 250 commodities can measure approximately the changes in all
wholesale prices, it must be because the known fluctuations in the
prices of these selected commodities are fair samples of the unknown
fluctuations in the prices of the vastly larger number of other com­
modities for which quotations are not collected. Now if (1) the
price fluctuation^ of each commodity that is bought and sold were
strictly independent of the price fluctuations of every other com­
modity, and if (2) each commodity had just the same importance as
an element in the general system of prices as every other commodity,
then any series of price quotations collected at random would be a
fair sample for determining th6 average changes in the wholesale
prices of commodities in general. Of course, the larger the number
of commodities included, the more trustworthy would be the index
number. In Table 8, for example, the first index number would be
adjudged the best, and the divergencies between it and its fellows
would be held to result from the scantier material from which the lat­
ter are made.
In fact, however, the situation is by no means so simple, because
neither of the above-mentioned conditions holds true. Commodities




52

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

are far from being all of the same importance as elements in the whole
system of prices. With the complications arising from this fact the
section on the problems of weighting will deal. Neither are the price
fluctuations of different commodities independent of each other. On
the contrary, the price changes of practically every commodity in the
markets of the whole country are causally related to the changes in
the prices of a few or of many, perhaps in the last resort of all, other
commodities that are bought and sold. Most of these relations are
so slight that they can not be traced by statistical methods. But
certain bonds are so close and so strong that they establish definite
groups of related prices which fluctuate in harmony with one another
and which differ in definable ways from the fluctuations of o.ther such
groups. The present task is to show the existence of these groups
and the effects which they exercise upon index numbers.
First, the price fluctuations of a raw material are usually reflected
in the prices of its manufactured forms. Hence to quote in some
cases both the raw material and several of its finished products, and
to quote in other cases the raw material alone, assigns certain groups
of related prices a larger influence upon the results than is assigned
the other groups. When the aim is to secure a set of samples which
fairly represent price fluctuations as a whole, the existence of these
groups must be taken into account. Neglect on this score may give
a misleading twist to the final index numbers. A celebrated case in
point is that of the Economist index number in 1863-1865. Out of
the 22 commodities included in the Economist’s list as then consti­
tuted 4 consisted of cotton and its products. Hence when the
blockade of Southern ports during the Civil War raised the price of
cotton, the Economist index numbers grossly exaggerated the aver­
age rise in the price level, as appears from the following comparison
between the Economist’s results for 1860-1865 and the corresponding
English figures compiled by Sauerbeck:1

Year.

1860.......................................................................................................................
1861................................................................................. ' ...................................
1862.......................................................................................................................
1863.......................................................................................................................
1864.......................................................................................................................
1865.......................................................................................................................

Economist Sauerbeck’s
index number index number
(prices in
(prices in
1860=100).
1860=100).
100
102
109
136
145
136

100
100
106
109
112
106

1 To make the comparison as fair as possible, both series are here given, not in their original form, but
recomputed on a common basis. See Wholesale Prices, Wages, and Transportation, report by Mr.
Aldrich from the Committee on Finance, March 3,1893,52d Cong., 2d sess., Senate Report No. 1394, Part I,
pp. 226 and 255,




THE MAIvINtt AND USINCJ OF INDEX NUMBERS.

53

Directly opposing the relations which unite the prices of finished
goods with the prices of their raw materials is a second set of
influences which make the price fluctuations of manufactured goods
considered as a group characteristically different from the price
fluctuations of their raw materials considered as a separate group.
Table 9 presents several sets of index numbers designed to throw
these characteristic differences into high relief. The first two columns
compare the relative prices of the 49 raw materials quoted by the
Bureau of Labor Statistics and of the 183 to 193 more or less manu­
factured commodities in its list.1 The second pair of columns con­
tains index numbers made from the prices of 20 raw materials and
of 20 products manufactured from these same materials.2 Then
come three columns giving index numbers made from the prices of
five great staples at three successive stages of manufacture: Wheat,
flour, and bread; cotton, cotton yarns, and cotton textiles; wool,
worsted yams, and woolen textiles; pig iron, steel billets, and steel
tools; hides, leather, and shoes.3 The later sections of the table give
the data for each of these last-mentioned groups separately. These
several comparisons establish the conclusion that manufactured
goods are steadier in price than raw materials. The manufactured
goods fell less in 1890-1896, rose less in 1896-1907, again fell less in
1907-1908, and rose less in 1908-1913. Further, the manufactured
goods had the narrower extreme range of fluctuations, the smaller
average change from year to year, and the slighter advance in price
from one decade to the next.4 It follows that index numbers made
from the prices of raw materials, or of raw materials and slightly
manufactured products, must be expected to show wider oscillations
than index numbers including a liberal representation of finished
commodities.
Third, there are characteristic differences among the price fluctua­
tions of the groups consisting of mineral products, forest products,
1 See Bulletin No. 149, pp. 13 and 14. The differences between tbe original figures and those given here
are due (1) to the dropping of decimals, (2) to the exclusion of 11 commodities which the Bureau of Labor
Statistics quotes in the years 1908-1913 only, (3) to the computation of the arithmetic means in these years
by the method applied in 1890-1907 in place of the bureau’s roundabout method. See Bulletin No. 149,
p. 32 and pp. 42 to 44 of this bulletin.
2 The articles included here are those from which the index number of 40 commodities in Table 8 was
made. For the list, see p. 48 and note.
3 For the lists of textiles and of tools, see Bulletin of the Bureau of Labor, No. 99, March, 1912, pp.
554-556 and 682-683.
4 Like most generalizations about price changes, these statements are strictly valid only in the case of
averages covering several commodities, but the exceptions are not numerous, even in the case of single
commodities, as detailed study of the wheat, cotton, wool, iron, and leather groups will show.




54

BULLETIN OF TH E BUREAU OF LABOR STATISTICS.
TABLE

9.—IN DEX NUMBERS MADE FROM THE PRICES OF R A W M ATERIALS
[Data from the Bulletin of the Bureau
(Arithmetic means. Average
Twenty
pairs.

Five triplets.

"Wheat group.

49
raw
ma­
teri­
als.

183 to
193
man­
ufac­
tured
prod­
ucts.

1

2

2

115
116
108
104
93
92
84
88
94
106
112
111
122
123
120
121
127
133
124
131
135
135
145
139

112
111
106
106
97
94
92
90
93
101
110
108
111
112
111
115
122
129
121
123
129
124
127
128

113
114
104
99
91
94
85
88
98
114
118
120
127
122
123
127
135
146
135
143
149
144
151
149

112
114
105
103
94
96
92
89
92
103
111
113
118
114
113
117
120
131
124
127
132
127
132
128

125
117
103
95
79
89
87
94
101
111
120
no
123
125
128
132
136
145
130
149
149
135
141
143

119
116
109
100
86
89
88
90
95
107
no
102
110
114
115
115
119
126
117
126
125
115
119
122

108
107
106
105
98
95
95
94
95
98
105
102
103
106
110
114
121
125
120
121
124
120
124
127

119
128
105
90
74
80
85
106
118
95
94
96
99
105
138
135
106
121
132
160
146
131
140
127

121
126
104
89
78
84
91
110
109
88
88
87
90
97
125
122
97
109
119
139
126
112
122
109

101
101
101
101
101
98
97
101
101
101
101
101
101
101
106
110
no
no
113
116
118
118
122
123

Averages, 1890-1899..........................
1900-1909 ..........................
1910-1913 ..........................
Number of points by which prices
rose (+ ) or fell (—) in—

100
122
139

100
116
127

100
130
148

100
119
130

100
130
142

100
115
120

100
113
124

100
119
136

100
107
117

100
107
120

1890-1896 .....................................
1896-1907 .....................................
1907-1908................................... .
1908-1913 .....................................
Difference between highest and
lowest relative prices.

-3 1
+49
- 9
+15
61

-2 0
+37
- 8
+ 7
39

—28
+61
-1 1
+14
66

-2 0
+39
- 7
+ 4
43

-3 8
+58
-1 5
+13
70

-3 1
+38
- 9
+ 5
40

-1 3
+30
- 5
+ 7
33

-3 4
+36
+11
- 5
86

-3 0
+18
+10
-1 0
61

- 4
+13
+ 3
+10
26

Average change from year to year...

5.5

4.0

6.4

4.9

8.4

5.5

3.1

13.6

11.6

1.3

Year.

Raw
ma­
teri­
als.

Maiv
ufaetured
goods.

Raw
ma­
teri­
als.

Inter­
medi­ Fin­
ate ished Wheat. Wheat Bread.
flour.
prod­ goods.
ucts.

Number of commodities included. .
1890....................................................
1891....................................................
1892....................................................
1894....................................................
1896....................................................
1897....................................................
1898......................... ..........................
1899....................................................
1900....................................................
1901....................................................
1902....................................................
1903.................... ...............................
1904....................................................
1905......................: ............................
1906....................................................
1907....................................................
1908....................................................
1909....................................................
1910....................................................
1911....................................................
1912....................................................
1913....................................................




55

THE MAKING AND USING OF INDEX N UMBERS.
AND OF MANUFACTURED . GOODS, B Y YEARS, 1890 TO 1913.
of Labor Statistics, No. 149.]
prices in 1890-1899-100.)

Cotton group.

Leather group.

Iron group.

Wool group.

Year.

Wool­
Leath­
Steel
Raw Cot­ Cot­
en
ton Raw Worst­
Pig bil­ ‘Steel Hides.
er.
Shoes.
ed
cot­ ton tex­
tex­
wool.
ton. yams. tiles.
yams. tiles. iron. lets. tools.

16

4

1

11

1

122
123
117
110
91
74
73
83
101
107
118
102
112
118
117
125
129
128
118
130
124
116
119
113

111
112
112
109
96
88
87
90
98
100
111
105
106
111
112
119
125
124
121
122
124
120
123
123

131
116
106
96
83
91
88
78
77
134
140
112
155
141
104
124
145
175
125
127
124
112
118
122

142
118
110
95
77
86
88
70
71
145
116
112
142
130
103
112
128
136
122
114
118
100
104
120

107
106
105
103
99
95
96
95
94
101
112
110
115
118
118
128
134
138
134
129
131
123
124
126

100
102
93
80
68
110
87
106
123
132
127
132
143
125
124
153
165
155
143
176
165
158
188
196

101
101
97
97
92
108
95
96
104
109
113
111
113
112
109
112
120
124
119
127
125
121
129
139

100
116
110

100
120
118

100
116
123

100
135
119

100
122
111

100
124
126

100
144
177

100
116
129

-2 2
+38
-1 7
+10
48

-6 1
+51
- 4
-1 3
62

-4 9
+55
-1 0
- 5
57

-2 4
+37
- 3
+ 2
38

-4 3
+87
-5 0
- 3
98

-5 4
+48
-1 4
- 2
75

-1 1
+42
- 4
- 8
44

-1 3
+68
-1 2
+53
128

- 6
+29
- 5
+20
47

6.1

9.1

8.1

3.9

17.5

16.0

3.7

14.7

5.0

1

2

24

2

143
111
99
107
90
94
102
92
77
85
124
111
115
145
156
123
142
153
135
i56
195
168
148
165

112
113
117
111
93
92
93
91
91
89
116
98
94
113
120
106
121
134
109
119
133
125
120
132

117
112
111
109
98
94
95
90
85
91
103
99
100
105
114
107
117
133
116
117
127
125
122
126

132
126
113
102
79
70
71
89
108
111
118
97
101
110
116
127
121
122
118
127
116
108
111
105

100
136
169

100
113
128

100
111
125

-4 1
+51
-1 8
+30
118

-1 9
+41
-2 5
+23
45

18.1

9.8




2

4

3

Number of com­
modities included.

106
104
103
101
99
100
101
96
94
95
98
96
96
96
98
106
119
120
114
121
118
116
127
137

1890.
1891.
1892.
1893.
1894.
1895.
1896.
1897.
1898.
1899.
1900.
1901.
1902.
1903.
1904.
1905.
1906.
1907.
1908.
1909.
1910.
1911.
1912.
1913.

100 Averages, 1890-1899.
1900-1909.
106
1910-1913.
125
Number of points
by which prices
rose ( + ) or fell
( —) in*—
1890-1896.
- 5
1896-1907.
+19
1907-1908.
- 6
1908-1913.
+23
Difference
between
43
highest and low­
est relative prices.
change
3.7 Average
from year to year.

56

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

C h a r t 6 .— IN D E X

NUMBERS OF THE PRICES OF 20 R A W M ATERIALS AND OF 20 PROD­
UCTS MANUFACTURED FROM THEM. (BASED ON TABLE 9.)




THE MAKING AND USING OF INDEX LUMBERS.

C h a r t 7.— IN D E X

57

NUMBERS OF THE PRICES OF WOOL, COTTON, HIDES, W H E A T, AND
PIG IRON IN TH E IR R A W , PA R T IA L LY MANUFACTURED, AND FINISHED FORMS.
(BASED ON TABLE 9.)




58

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

animal products, and farm crops. Table 10 presents index numbers
for these four groups. Fifty-seven commodities are included, all of
them raw materials or slightly manufactured products.1 Here the
striking feature is the capricious behavior of the prices of farm crops
under the influence of good and bad harvests. The sudden upward
jump in their prices in 1891, despite the depressed condition of busi­
ness, their advance in the dull year 1904, their fall in the year of
revival 1905, their failure to advance in the midst of the prosperity of
1906, their trifling decline during the great depression of 1908, and
their sharp rise in the face of reaction in 1911 are all opposed to the
general trend of other prices. The prices of animal products are
distinctly less affected by weather than the prices of vegetable crops,
but even they behave queerly at times, for example in 1893. Forestproduct prices are notable chiefly for maintaining a much higher
level of fluctuation in 1902-1913 than any of the other groups, a level
on which their fluctuations, when computed as percentages of the
much lower prices of 1890-1899, appear extremely violent. Finally,
the prices of minerals accord better with alternations of prosperity,
crisis, and depression than any of the other groups. And the
anomalies that do appear—the slight rise in three years (1896, 1903,
and 1913) when the tide of business was receding—would be removed
if the figures were compiled by months. For the trend of mineral
prices was downward in these years, but the fall was not so rapid as
the rise had been in the preceding years, so that the annual aver­
ages were left somewhat higher than before.2 An index number
composed largely of quotations for annual crops, then, would be
expected at irregular intervals to contradict capriciously the evidence
of index numbers in which most of the articles were mineral, forest, or
even animal products.
1 The lists of commodities are as follows:
Farm crops: Cotton, flaxseed, barley, com, oats, rye, wheat, hay, hops, beans, coffee, rice, pepper, tea,
onions, potatoes, cottonseed meal, and jute—18 articles.
Animal products: Hides, cattle, hogs, sheep, eggs, lard, milk, tallow, silk, and wool—10 articles.
Forest products: Hemlock, maple, oak, white pine, yellow pine, poplar and spruce lumber, together with
turpentine, tar, and rubber—10 articles.
Mineral products: Salt, anthracite coal, bituminous coal, coke, crude petroleum, copper ingots, lead (pig),
pig iron, bar iron, steel billets, quicksilver, silver bars, tin (pig), spelter, zinc, brick, cement,lime, and brim­
stone—19 articles.
2 Compare the monthly figures compiled by the Bureau of Labor Statistics for its group of “ Metals and
implements, ” Bulletin No. 149, p. 18. These figures are largely influenced by the relatively stable prices
of 11 different kinds of tools. Monthly data for the 19 mineral products of Table 10 would probably show
oven more of a decline between January and December in these years.




T11E MAKINCi AND ySIN O OF INDEX NUMBERS.
C H A R T 8.—IN D E X

59

NUMBERS OF THE PRICES OF 19 MINERAL PRODUCTS AND OF 18 FARM
CROPS. (BASED ON TABLE 10.)




GO
T able

BULLETIN OF THE BUREAU OF LABOR STATISTICS.
10 .-IN D E X NUMBERS MADE FROM PRICES OF MINERAL, FOREST, ANIMAL,
AND FARM PRODUCTS, B Y YEARS, 1890 TO 1913.
[Data from the Bulletin of the Bureau of Labor Statistics, No. 149.]
(Arithmetic means.

Average prices in 1890-1899= 100.)

Year.

Mineral
Animal
Forest
products. products. products.

Farm
crops.

Number of commodities included.................................................

19

10

10

1890..................................................................................................
1891.................... ............................................................................
1892..................................................................................................
1893..................................................................................................
1894..................................................................................................
1895..................................................................................................
1896..................................................................................................
1897..................................................................................................
1898..................................................................................................
1899..................................................................................................
1900..................................................................................................
1901..................................................................................................
1902..................................................................................................
1903..................................................................................................
1904..................................................................................................
1905..................................................................................................
1906..................................................................................................
1907..................................................................................................
1908..................................................................................................
1909..................................................................................................
1910..................................................................................................
1911..................................................................................................
1912..................................................................................................
1913..................................................................................................

119
111
105
98
87
91
92
88
92
117
120
113
119
124
115
123
135
137
118
121
120
120
132
136

107
105
99
98
95
96
94
95
99
112
121
113
123
137
142
149
163
169
151
164
181
172
168
169

106
108
109
116
94
95
82
88
97
105
111
112
128
117
113
121
128
135
126
144
152
131
146
150

119
126
110
105
101
92
76
83
92
96
105
114
120
116
124
116
116
125
124
130
134
151
158
135

Averages, 1890-1899........................................................................
1900-1909........................................................................
1910-1913........................................................................
Number of points by which prices rose ( + ) or fell ( —) in—
1890-1896...................................................................................
1896-1907...................................................................................
1907-1908...................................................................................
1908-1913...................................................................................

100
123
127

100
143
173

100
124
145

100
119
145

—27
+45
—19
+18

-1 3
+75
—18
+18

—24
+53
—9
+24

—43
+49
_ i
+U

Difference between highest and lowest relative prices................

50

87

70

82

Average change from year to year................................................

7.0

7.4

8.9

18

8.2

Fourth, there are characteristic differences between the price fluc­
tuations of manufactured commodities bought by consumers for
family use and the price fluctuations of manufactured commodities
bought by business men for industrial or commercial use. Such at
least is the story told by Table 11. The data employed here are
quotations for 28 articles from the Bureau of Labor Statistics*
list that rank distinctly as consumers’ goods and 28 that rank as
producers’ goods.1 Though consisting more largely of the erratic­
i The consumers' goods are bread, crackers, butter, cheese, salt fish, evaporated apples, prunes,
raisins, beef, mutton, pork, molasses, corn starch, sugar, vinegar, shoes, cotton textiles, woolen textiles,
candles, matches, quinine, furniture, earthenware, glassware, woodenware, table cutlery, soap, and
tobacco. The producers' goods are bags, cotton yarns, leather, linen shoe thread, worsted yarns, refined
petroleum, barbed wire, builders’ hardware, copper wire, lead pipe, nails, steel rails, tools, wood screws,
pine doors, plate glass, window glass, carbonate of lead, oxide of zinc, putty, rosin, shingles, muriatic acid,
sulphuric acid, malt, paper, proof spirit, and rope.
. It will be noticed that a large proportion of the consumers' goods are subject to very slight manufac­
turing processes, notably the foods. Hence the difference between the two index numbers can scarcely
be regarded as merely a fresh contrast between the fluctuations of finished goods and of intermediate
products.




61

THE M AKING AND USING OF INDEX NUMBERS.

ally fluctuating farm products, the consumers' goods are steadier in
price than the producers’ goods, because the demand for them is less
influenced by changes in business conditions.
T a b le

11.— IN DEX NUMBERS MADE FROM THE PRICES OF CONSUMERS' GOODS AND

PRODUCERS’ GOODS, B Y YEARS, 1890 TO 1913.
[Data from Bulletin of the Bureau of Labor Statistics, No. 149.]
(Arithmetic means.

Average prices in 1890-1899= 100.)

Year.

189
189
189
189
189
189
189
189
189
189
190
190
190
190
190
190
190
190
190
190
191
191
191
191

0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3

*.................................................

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

Averages, 1890-1899............................................................
1900-1909............................................................
1910-1913............................................................
Number of points by which prices rose ( + ) or fell ( —) in1890-1897........................................................................
1897-1907.......................................................................
1907-190 8
1908-191 3
.......................................
Difference between highest and lowest relative prices__
Average change from year to year.....................................

Consum­
ers’
goods.

Produc­
ers’
goods.

112
109
104
108
100
95
91
90
94
98
106
105
108
105
103
106
110
114
112
114
118
119
118
121

115
111
107
102
92
91
93
89
93
107
117
113
114
114
114
117
124
133
119
118
126
125
125
123

100
108
119

100
118
125

- 22
+ 24
- 2
+ 9

- 26
+ 44
- 14
+ 4

31

44

3.4

4.

Other groups of related prices having specific peculiarities of fluc­
tuation doubtless exist, but the analysis has been carried far enough
for the present purpose. That purpose is to show how the existence
of groups of prices which fluctuate in harmony with each other and
at variance with other groups affects index numbers in general and
in particular the six index numbers for the United States given in
Table 8. To apply the knowledge gained from the preceding analysis
to the explanation of the differences among these six index numbers
is not difficult when once the commodities included in each index
number have been classified on the basis of the groups which have
been examined.
First, the list of commodities used by the Bureau of Labor Sta­
tistics includes 29 quotations for iron and its products, 30 quotations
for cotton and its products, and 18 for wool and its products, besides




62

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

8 more quotations for fabrics made of wool and cotton together.

On
the other hand it has but 7 series for wheat and its products, 8 for
coal and its products, 3 for copper and its products, etc. The iron,
cotton, and wool groups together make up 85 series out of 242, or
35 per cent of the whole number. The same three groups furnish 36
(or 25 per cent) of the 145 series in the second index number in Table 8.
9.—IN D E X NUMBERS OF THE PRICES OF MANUFACTURED GOODS USED FOR
FAMILY CONSUMPTION AND FOR INDUSTRIAL PURPOSES. (BASED ON TABLE 11.)

Ch art

Similarly, cotton, wool, and wheat, or coal, or cattle, with their products,
make 20 per cent of the series in the third index number.
Does this large representation of three staples distort these index
numbers—particularly the bureau’s series where the disproportion is
greatest? Perhaps; but if so the distortion does not arise chiefly
from the undue influence assigned to the price fluctuations of raw
cotton, raw wool, and pig iron. For, contrary to the prevailing
impression, the similarity between the price fluctuations of finished
products and their raw materials is less than the similarity between




THE MAKING AND USING OF INDEX NUMBERS.

63

the price fluctuations of finished products made from different mate­
rials. Such at least is the testimony of Table 9. As babies from
different families are more like one another than they are like their
respective parents, so here the relative prices of cotton textiles,
woolen textiles, steel tools, bread, and shoes differ far less among
themselves than they differ severally from the relative prices of raw
cotton, raw wool, pig iron, wheat, and hides.1 Hence the inclusion,
of a large number of articles made from iron, cotton, and wool affects
an index number mainly by increasing the representation allotted
to manufactured goods. What materials those manufactured goods
are made from makes less difference in the index number than the
fact that they are manufactured. To replace iron, cotton, and
woolen products by copper, linen, and rubber products would change
the results somewhat, but a much greater change would come from
replacing the manufactured forms of iron, cotton, and wool by new
varieties of their raw forms.2
This similarity among the price fluctuations of manufactured goods
arises from the fact demonstrated by Table 9 that such articles are
relatively steady in price. Does knowledge of this steadiness assist
in explaining the differences among the six American index numbers
of Table 8 ? To answer we must find the proportions of raw and
manufactured commodities included in each index number. Classi­
fication along this line is rather uncertain in many cases, but the
results shown in the following schedule, if not strictly correct, are at
least uniform in their errors.
i A compilation of the differences among the relative prices in question taken seriatim for each of the 24
years 1890-1913 yields the following results:
Average differences between the relative prices of—
Raw cotton and cotton textiles...........................................................................20.7 points.
Raw wool and woolen textiles............................................................................. 8.9 points.
Pig iron and steel tools........................................................................................ 14.0 points.
Wheat and breads................................................................................................ 15.0 points.
Hides and shoes....................................................................................................31.6 points.
Average....................................................................................................... 18.0 points.
Cotton textiles and woolen textiles.....................................................................
Cotton textiles and steel tools................................•.............................................
Cotton textiles and bread.....................................................................................
Cotton textiles and shoes....................................................................................
Woolen textiles and steel tools............................................................................
Woolen textiles and bread...................................................................................
Woolen textiles and shoes...................................................................................
Steel tools and bread................. .........................................................................
Steel tools and shoes.............................................................................................
Bread and shoes...................................................................................................

5.3 points.
7.8 points.
6.9 points.
6.7 points.
6.1 points.
7.3 point#.
8.1 points.
9.4 points.
9.0 points.
4.7 points.

Average.......................................................................................................

7.2 points.

While the fluctuations in the prices of manufactured goods are generally slighter than those in the
prices of raw materials, they are nevertheless violent at times, as in the case of cotton yams and cotton
textiles during the Civil War. (See p. 52.)
2




64

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

Table 12.—NUMBER

AND PER CENT OF R A W AND MANUFACTURED COMMODITIES
INCLUDED IN THE SIX IN D E X NUMBERS OF TABLE 8.
Number of—

Percentage of—

Total
number
Manu­
Manu­
Raw
of com­
Raw
factured commod­
factured
modities. commod­ commod­
commod­
ities.
ities.
ities.
ities.

Index number.

First...............................................................................
Second...........................................................................
Third........... : ................................................................
Fourth...........................................................................
Fifth..............................................................................
Sixth..............................................................................

242
145
50
40
25
25

49
36
26
17
19
10

193
109
24
23
6
15

20
25
52
43
76
40

80
75
48
57
24
60

On this showing the Bureau of Labor Statistics series ought to be
the steadiest, and the second series the next steadiest— and so they are,
as the summaries at the bottom of the columns in Table 8 show.
With the smaller index numbers, however, the rule does not work
well, for the most variable of all—the sixth—has a larger per cent of
manufactured goods than the other three. Moreover, number four
is more variable than number three, though it has relatively more
manufactured goods. But the preceding studies of different groups
throw further light upon the matter.
It has been found that among manufactured commodities those
bought for family consumption are steadier in price than those bought
for business use. To take account of this factor the manufactured
goods in the several series are classified as primarily consumers’
goods, primarily producers’ goods, or as bought in large measure by
both classes of purchasers.
T able 13.—CLASSIFICATION OF THE MANUFACTURED COMMODITIES INCLUDED IN

THE SIX IN D E X NUMBERS OF TABLE 8.
Number of—

Index number.

First...........................
Second........................
Third.........................
Fourth..........
Fifth...........................
Sixth..........................

Per cent of—

Both con­
Both con­
Con­
Con­
Pro­
sumers’
Pro­
sumers’
Manu­
Manu­
ducers’ and pro­ factured
sumers’
sumers’
ducers'
and pro­
factured
com­
com­
com­
com­
ducers’
articles. modities. modities. ducers’
articles. modities. modities.
com­
com­
modities.
modities.
193
109
24
23
6
15

108
51
11
10
3
4

73
47
12
12
3
11

12
11
1
1

80
75
48
57
24
60

45
35
22
25
12
16

30
32
24
30
12
44

5
8
2
2

Here it does turn out that the two series (numbers four and six)
which are highly variable despite the inclusion of many manufactured
goods have relatively more of those manufactured goods which as
a group are most variable. So far as this factor counts, then, it
counts toward clearing up the contradiction pointed out in the
preceding paragraph. It also brings out a further reason for the
comparative stability of the first two series.



65

THE MAKING AND USING OF INDEX NUMBERS.

The one remaining form of analysis suggested above seems easy
to apply. In the schedule below, raw and slightly manufactured
commodities like those used in Table 10 are distributed among four
groups according as their constituents come chiefly from mines,
forests, animal sources, or cultivated fields. There is little doubt
about the classification here, but there is much doubt about the
significance of the results as applied to our six index numbers* The
figures in the schedule are either such small percentages of the whole
number of series that they can not exercise much influence upon the
results, or such small numbers that they can not claim to be typical
of their groups. Further, the second part of the schedule shows that
there is less difference among the six index numbers than appears at
first sight in the proportions of the raw and slightly manufactured
commodities which consist of mineral, forest, animal, and farm prod­
ucts. Hence it is not surprising that efforts to account for the
divergences in Table 8 by appealing to this schedule and to Table 10
accomplish little, especially for the smaller index numbers. This
much does appear regarding the first two series: Whenever mineral
products and farm crops move sharply in opposite directions the
Bureau of Labor Statistics’ index diverges from its mate in harmony
with mineral products, while the series of 145 commodities bends
toward the agricultural products—which is what should happen
according to the schedule.
14.—FARM, ANIMAL, FOREST, AND MINERAL PRODUCTS IN R A W OR SLIGHTLY
MANUFACTURED FORM, INCLUDED IN THE SIX IN D E X NUMBERS OF TABLE 8.

T able

Index
number.

First...........
Second.......
Third..........
Fourth.......
Fifth..........
Sixth..........

Number of—
Total
num­
Raw
ber
and
Ani­ Forest
of
slightly
com­
Farm
mal
prod­
prod­
modi­ manu­
crops.
fac­
ucts.
ties.
ucts.
tured
goods.
242
145
50
40
25
25

74
57
30
19
23
18

18
18
10
6
7
5

15
10
8
6
5
5

12
10
3
1
2
1

Per cent of the whole number consist­
ing of—
Raw
and
Min­ slightly
Farm
eral
crops.
prod­ manu­
fac­
ucts.
tured
goods.

Ani­
mal
prod­
ucts.

Forest
prod­
ucts.

7
12
20
15
28
20

6
7
16
15
20
20

5
7
6
3
8
4

29
19
9
6
9
7

30
39
60
48
92
72

Min­
eral
prod­
ucts.

12
13
18
15
36
28

Per cent of the raw and slightly
manufactured commodities con­
sisting of—
Indexnumber.
Farm
crops.

First......................................................................................................
Second...............................................^...................................................
Third...............................................................................................‘___
Fourth....................................................................................................
Fifth.......................................................................................................
Sixth......................................................................................................

94261°—Bull. 173--15----- 5



25
31
33
32
30
28

Animal
prod­
ucts.

Forest
prod­
ucts.

20
18
27
32
22
28

16
18
10
4
9
5

Mineral
prod­
ucts.
39
33
30
32
39
39

66

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

Two practical conclusions of moment to both the makers and the
users of index numbers are established by this section. ( 1) To make
an index number that measures the changes in wholesale prices at
large, samples must be drawn from all the various groups that behave
in peculiar ways. (2) In using an index number made by others, one
must study the list of commodities included critically with these
groups in mind to know what it really does measure.
The first conclusion seems to contradict a rule often practiced and
sometimes preached. Most of the middle-sized index numbers are
confined to raw materials and slightly manufactured goods. Most
of the small index numbers are confined to foods alone. The makers
of both sets argue that their series are more “ sensitive” and therefore
better measures of price changes than the larger series, which are
loaded down with a mass of miscellaneous manufactured goods.
And many users of index numbers seem to prefer a series like Sauer­
beck’s with only 45 commodities, or even one like the Annalist’s
with only 25 commodities, to one like that of the Bureau of Labor
Statistics with five or ten times the number.
Critics who take this stand usually assume tacitly that the purpose
of an index number is to serve as a “ business barometer,” or to
measure changes in “ the cost of living.” If these aims were always
clearly realized by the critics and clearly stated for their readers the
room left for differences of opinion would be narrow. In Table 8
the index number with 145 commodities shows itself a more sensi­
tive and on the whole more faithful barometer of changing business
conditions during the 24-year period from 1890 to 1913 than the
official series with 242 commodities,1 and the preceding analysis shows
that the sluggishness of the larger index number is due chiefly to its
proportion of manufactured goods. For this particular purpose, then,
a series modeled after Sauerbeck’s has strong claims to preference
over one including a larger number of commodities. Indeed, in the
light of the preceding discussion one might carry the process of exclu­
sion much further and throw out of the business barometer not only
manufactured goods but also all farm crops, on the ground that their
prices depend on the eccentricities of the weather, and most forest
products, on the ground that their prices are rising so fast as to
obscure the effects of bad times, etc. But clearly such exclusions,
while they might make the resulting figures more responsive to
changes in business conditions, would also make the figures less
acceptable as a measure of changes in prices as a whole. The slug­
gish movements of manufactured goods and of consumers’ commodi­
ties in particular, the capricious jumping of farm products, the rapidly
increasing dearness of lumber, etc., are all part and parcel of the fluc-




1Compare p. 111.

THE M AKING AND USING OF INDEX NUMBERS.

67

tuations which the price level is actually undergoing. Consequently,
an index number which pretends to measure changes in the general
level of prices can not logically reject authentic quotations from any
of these groups. Every restriction in the scope of the data implies a
limitation in the significance of the results.
Probably the most illuminating way of presenting an index num­
ber that aspires to cover the whole field of prices at wholesale would,
be to publish separate results for the groups that have characteristic
differences of price fluctuations, and then to publish also a grand total
including all the groups. The groups to be recognized and the distri­
bution of commodities among them is a difficult matter to decide.
Doubtless intensive research along the lines here followed would sug­
gest the desirability of further subdivisions and perhaps the realign­
ment of the whole classification. But, as matters stand, the most sig­
nificant arrangement seems to be ( 1) a division of all commodities
into raw and manufactured products; (2) the subdivision of raw
commodities into farm crops and animal, forest, and mineral products;
(3) the subdivision of manufactured products according as they are
bought mainly for personal consumption, mainly for business use, or
largely for both purposes. It would also be interesting in a supple­
mentary table to bring together index numbers for the leading raw
materials and the products manufactured from them.
This classification is based upon differences among the factors affect­
ing the supply of and the demand for commodities that belong to the
several groups—that is, upon differences among the factors which
determine prices. If we wish our index numbers to help toward an
understanding of changes in the price level, a classification along
these causal lines promises the most illuminating results; but it is
not the basis of classification usually adopted.
In most large index numbers the commodities are divided among
several classes, but these classifications seldom possess logical con­
sistency. Among the nine groups recognized by the Bureau of Labor
Statistics, for example, one group, “ Farm products,” emphasizes the
place of production; four groups, “ Food, etc.,” “ Fuel and lighting,”
“ Lumber and building materials,” and “ House-furnishing goods,”
emphasize the use to which commodities are put; three groups apply
a double criterion, use and physical character of the goods, namely,
“ Cloths and clothing” (which includes such articles as 2-bushel bags),
“ Metals and implements,” and “ Drugs and chemicals” ; the remain­
ing group is frankly styled “ Miscellaneous.” Such a classification is
not without usefulness, for there doubtless are readers especially
interested in the prices of, say, all things that are raised on farms,
and others who care especially about the prices of things used to fur­
nish houses, or things that can be classed together as drugs and
chemicals whether they are used chiefly as medicines or to make farm




68

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

fertilizers. But if a classification of this empirical character is main­
tained, it might with advantage be accompanied by a classification
that throws more light upon the workings of the complex system of
prices.
As for the small series made from the prices of foods alone or from
the prices of any single group of commodities, it is clear that, however
good for special uses they may be, they are untrustworthy as generalpurpose index numbers. Table 15 shows what differences are likely
to appear at any time between series confined to foods and series
covering a wider field. The general-purpose indexes are taken from
Table 8 ; two of the food indexes include the commodities quoted by
the Annalist index number and by the Gibson index number as now
constituted; the third food index is the bureau's own series for foods,
with decimals dropped and new arithmetic means for 1908-1913.1
T able

15.—IN D E X NUMBERS OF THE PRICES OF FOODS, AND GENERAL-PURPOSE
IN D E X NUMBERS, B Y YEARS, 1890 TO 1913.
[Data from BulletiiTof the Bureau of Labor Statistics, No. 149.]
(Arithmetic means. Average prices in 1890-1899= 100.)
General-purpose index
numbers from Table 8.
Year.

242 to 261
com­
modities.

Averages, 1890-1899......................................
1900-1909.......................................
1910-1913......................................
Number of points by which prices rose ( + )
or fell ( - ) i n 1890-1896.................................................
1896-1907.................................................
1907-1908.................................................
1908-1912.................................................
1912-1913.................................................
Difference between highest and lowest
relative prices............................................
Average change from year to year...............




25 com­
modities,
first list.

Index numbers of the prices of
foods.
25 com­
modities,
Annalist
list.

113
112
106
106
96
94
90
90
93
102
111
109
113
114
113
116
123
130
122
125
130
126
130
130

115
112
103
103
92
95
88
90
96
107
113
111
116
118
122
123
130
132
124
133
133
129
140
142

109
119
108
116
102
95
80
84
92
93
99
105
117
107
109
110
115
120
126
134
137
131
143
139

100
118
129

100
122
136

- 23
+ 40
- 8
+ 8
± 0

- 27
+ 44
- 8
+ 16
+ 2

40
4.0

22 com­
modities,
Gibson
list.

48 com­
modities,
Bureau of
Labor Sta­
tistics’ list.

109
121
108
110
98
94
81
87
96
96
100
106
118
107
115
114
111
121
128
127
137
134
147
139

112
116
104
110
100
95
84
88
94
98
104
106
111
107
107
109
113
118
122
125
129
127
135
131

100
114
138

100
115
139

100
112
131

- 29
+ 40
+ 6
+ 17
- 4

- 28
+ 40
+ 7
+ 19
- 8

- 28
+ 34
+ 4
+ 17
- 4

54
5.0

Compare the explanation given, pp. 42 to 44.

63
7.1

.

66
7.3

51
5.0

THE MAKING AND USING OF INDEX NUMBERS.

Ch a r t

69

10.—IN D E X NUMBERS OF THE PRICES OF 25 FOOD PRODUCTS AND OF 25 MIS­
CELLANEOUS COMMODITIES. (BASED ON TABLE 15.)




70

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

The three index numbers for foods agree better than might have
been expected in view of the dissimilarity of the lists of commodities
which they quote and the brevity of two of the lists.1 The bureau
series is rather steadier than the others, because of the larger propor­
tion of manufactured products included in it; but this series and that
of the Annalist invariably agree about the direction in which prices
are moving,2 and the Gibson figures agree with the other two series
in 19 years out of the 24. On the other hand, the three food indexes
often contradict the evidence of the two general-purpose index num­
bers in a striking fashion. Such contradictions occur in 1890-1891,
1892-1893, 1900-1901, 1902-1903, 1907-1908, and 1912-1913. these
differences are due chiefly to a contrast in the years mentioned between
business conditions and harvest conditions. They parallel the differ­
ences in Table 10 between the index numbers of mineral products and
those of farm crops, or farm crops and animal products taken together;
for the food indexes are made up almost wholly from the prices of vege­
table crops, food animals, and their derivatives.3 A food index num­
ber, then, is likely at any time to give a wrong impression regarding
the shifting of prices in general and is especially treacherous as a busi­
ness barometer. Nor can such an index when made from wholesale
prices be trusted to show changes in the “ cost of living” ; for living
expenses are made up of retail prices, and fluctuations in retail prices
do not follow closely those in the wholesale markets.
The second conclusion which this section establishes is that large
index numbers are more trustworthy for general purposes than small
ones, not only in so far as they include more groups of related prices,
but also in so far as they contain more numerous samples from each
group. What is characteristic in the behavior of the prices of farm
crops, of mineral products, of manufactured wares, of consumers7
goods, etc.—what is characteristic in the behavior of any group of
prices—is more likely to be brought out and to exercise its due effects
upon the final results when the group is represented by 10 or 20 sets
of quotations than when it is represented by only one or two sets.
i Of the 56 articles included altogether, only 11 are common to all three lists. The Gibson list has 8
commodities and the Annalist list has 4 commodities classified by the bureau with farm products instead
of with foods, while the bureau has 34 foods not quoted by Gibson and 27 not quoted by the Annalist.
Even the two short lists have only 15 articles in common, while Gibson has 7 articles not quoted b y the
Annalist, and the Annalist has 10 articles not quoted by Gibson.
For the bureau’s list see Bulletin No. 149, pp. 90-107.
The Annalist list runs—oats, cattle, fresh beef, salt beef, hogs, bacon, salt pork, lard, sheep, mutton,
butter (two kinds), cheese, coffee, sugar, wheat flour (two kinds), rye flour, com meal, rice, beans, potatoes,
prunes, evaporated apples, and codfish.
The Gibson list is—barley, com, oats, rye, wheat, cattle, hogs, sheep, butter, coffee, wheat flour (two
kinds), corn meal, bacon, fresh beef, salt beef, hams, mutton, sugar (two kinds), tea, and potatoes.
* Even in 1903-4 the bureau’s figures record a slight advance of prices in harmony with the Annalist
figures, though this advance is confined to the decimal columns and disappears when the decimals are
rounded off.
* The exceptions are salt and soda, and of these articles the Annalist and Gibson quote neither.




THE M AKING AND USING OF INDEX NUMBERS.

71

The basis of this contention is simple: In every group that has been
studied there are certain commodities whose prices seldom behave
in the typical way, and no commodities whose prices can be trusted
always to behave typically. Consequently, no care to include com­
modities belonging to all the important groups can guarantee accurate
results, unless care is also taken to get numerous representatives of
each group.
Even here the matter does not end. The different groups that have
been discussed, the other groups that might have been discussed,
and the commodities that are included within the several groups
differ widely in importance as elements in the system of prices. To
these differences, and to the methods of making them count in index
numbers, we must now turn.
7. PROBLEMS OF WEIGHTING.

It is customary to distinguish sharply between “ simple” and
weighted” index numbers. When an effort is made to ascertain
the relative importance of the various commodities included, and to
apply some plan by which each commodity shall exercise an influence
upon the final results proportionate to its relative importance, the
index number is said to be weighted. When, on the contrary, no such
effort is made, but every commodity is taken just as it comes and sup­
posedly allowed just the same chance to influence the result as every
other commodity, the index number is said to be unweighted, or
simple.
In unweighted series, however, it is seldom true that every com­
modity has just the same chance to influence the result as every other
commodity. For example, in Bradstreet’s index the influence of
every article upon the result varies as its price per pound happens to
be large or small.1 Again, the decisive objection to making index
numbers by merely adding the ordinary commercial quotations for
different articles is that these nominally simple series are in fact
viciously weighted series.2 Nor does the substitution of relative
prices for actual prices assure an equal chance to every article. For
instance, in its famous report of 1893, the Senate Committee on
Finance presented three wholesale-price index numbers— one simple
and two weighted; but in the simple series it included relative prices
for 25 different kinds of pocketknives, giving this trifling article
an influence upon the result more than eight times greater than that
given to wheat, corn, and coal put together. Finally, even if one
series of relative prices, and only one, be accorded each commodity,
it does not follow that equal percentages of change in the price of
“




i For details, see p. 101.

2 See

p. 34.

72

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

every article will always exercise equal influence upon the results.
For, as shown above, when the relative prices are computed upon
a fixed base and averaged by the use of arithmetic means, those
commodities that have a long period upward trend in price will
presently far outweigh in influence those commodities whose prices
are declining.1
Lack of attention to weighting, then, does not automatically secure
a fair field and no favor to every ocmmodity; on the contrary, it
results in what Walsh happily termed haphazard weighting.2 Indeed,
when it is desired to give each commodity an equal chance to influ­
ence the results, great care must be taken; practically a scheme of
equal weights must be devised. The real problem for the maker of
index numbers is whether he shall leave weighting to chance or seek
to rationalize it.
There are two excuses for neglect of weighting. First, as has been
shown in another connection, to collect satisfactory statistics showing
the relative importance of different commodities is extremely labori­
ous and extremely difficult.3 Second, there are high authorities who
hold that the results turn out much the same whether or not formal
weights are used.4 Certainly “ the weights are of much less impor­
tance in determining an index number of prices than the prices
themselves.” 5 But whether their importance is negligible is a ques­
tion best answered by a study of actual cases such as are shown in the
next table.6
1 See pp. 37 and 38.
M. Walsh, The Measurement of General Exchan^e-Value, pp. 81 and 82. Haphazard weighting is
not necessarily the worst weighting; indeed, it often is tetter than the weighting which results fromsome
systematic calculations. For example, Bradstreet’s plan of using actual prices per pound is certainly
systematic, but the weighting which this system involves is probably less defensible than the haphazard
weighting involved in most averages of the relative prices of commodities selected at random. See p. 101.
8 See p. 28. When the (then) Department of Labor started its present index number it canvassed the
subject of weighting, but decided to use a simple average, because of the “ impossibility of securing even
approximately accurate figures for annual consumption in the United States of the commodities included.”
(Bulletin of the Department of Labor No. 39, p. 234, March, 1912.) It did, however, allot two or more series
to certain commodities, and thus introduced a rough system of weights. Unfortunately the number of
series allotted to each commodity seems to have been determined quite as much by the ease of securing
quotations as by the importance of the articles. For criticism of the weighting which resulted, see pp. 61,
62, and 101.
* Compare A. L. Bowley, Elements of Statistics, 2d ed., pp. 113 and 220-224.
s Irving Fisher, The Purchasing Power of Money, revised edition, p. 406. For further details see the
papers by Edgeworth to which Fisher refers in his footnote.
* Details concerning the first three sets of simple and weighted averages can be found in the documents
referred to in the table. But the fourth set of comparisons is based upon hitherto unpublished data and
requires description.
The “ unweighted” index numbers in this set are arithmetic means of the relative prices given in the
bulletins of the Bureau of Labor Statistics for the commodities listed below. But where two or more series
of relative prices are shown in the bulletins for different grades of the same article, as in the case of cattle,
hogs, bacon, butter, com meal, pig iron, etc., they were replaced by a single average series for the article in
question, before the arithmetic means of the group were computed.
The “ weighted ” index numbers were made from these same relative prices in the following way: (1)
For each commodity included the Bureau of Labor Statistics made a careful estimate, based upon a critical
study of the best available sources of information, of the physical quantity of it entering into exchange in
the year 1909. By “ quantity entering into exchange” is meant the quantity bought and sold, irrespective
of the number of times it changed hands. (See pp. 77 and 78.) (2) These physical quantities were multi2 C.




THE M AKING AND USING OF INDEX NUMBERS.

73

The discrepancies here revealed between the averages with hap­
hazard and with systematic weights seldom amount to 10 per cent of
the results, except under the chaotic price conditions created by the
greenback standard in 1862-1873. In many kinds of statistics a 10
per cent margin of error is not accounted large. But in making wholesale-price index numbers for current years we may reasonably try to
get not two, but three, significant figures; and the third figure is
usually altered in appreciable degree by the substitution of systematic
for haphazard weights. Even the large Canadian series, with its 272
commodities, is shifted 9.5 points, or more than 7 per cent, in 1912 by
weighting.
plied by the average prices in 1909 of the respective commodities. (3) The resulting sums of money
were used as weights to multiply the relative prices of the respective commodities on the 1890-1899 base.
(4) The sums of the products were cast up for each year, and finally these sums were divided by the sums
of the weights, i. e., the value in exchange for 1909.
The average prices of the commodities in 1909 may be found in any of the recent wholesalc-price bulletins,
e. g., No. 149. The commodities included, and the estimated physical quantity of each entering into
exchange in 1909, are as follows:
Farm 'products: Barley, 75,300,538 bu.; cattle, 124,346,349 cwt.; corn, 460,778,251 bu.; cotton, 5,409,760,011
lbs.; flaxseed, 20,106,433 bu.; hay, 10,685,804 tons; hides, 922,243,894 lbs.; hogs, 76,438,923 cwt.; hops
48,076,921 lbs.; oats, 267,859,660 bu.; rye, 29,520,508 bu.; sheep, 11,498,090 cwt.; wheat, 683,416,528 bu.
Food, etc.: Beans, 8,468,385 cwt.; butter, 1,042,709,708 lbs.; cheese, 353,641,892 lbs.; coffee, 1,038,439,285
lbs.; eggs, 926,690,112 doz.; codfish, 684,692 cwt.; herring, 428,804 bbls.; mackerel, 190,565 bbls.; salmon,
18,431,003 doz. cans; buckwheat flour, 2,009,599 cwt.; rye flour, 1,594,346 bbls.; wheat flour, 107,306,408 bbls.;
currants, 32,163,998 lbs.; prunes, 138,795,607 lbs.; raisins, 12,438,044 boxes; glucose, 7,701,223 cwt.; lard,
1,243,572,129 lbs.; corn meal, 53,353,466 cwt.; bacon, 741,354,500 lbs.; beef, fresh, 4,209,196,748 lbs.; beef,
salt, 632,388 bbls.; hams, 789,861,744 lbs.; mutton, 495,458,067 lbs.; pork, salt, 4,760,690 bbls.; milk,
7,749,070,256 qts.; molasses, 55,689,983 gals.; rice, 1,042,538,693 lbs.; salt, 22,136,489 bbls.; soda, bicarbonate,
165,600,000 lbs.; pepper, 36,241,462 lbs.; sugar, raw, 6,316,033,669 lbs.; sugar, granulated, 7,366,818,210 lbs.;
tallow, 203,209,103 lbs.; vinegar, 98,403,927 gals.; potatoes, 397,491,062 bu.; onions, 4,972,947 cwt.; tea,
113,547,647 lbs.
Metals and implements: Bar iron, 2,166,529,067 lbs.; barbed wire, 6,471,300 cwt.; copper, ingot, 1,312,437,919
lbs.; copper wire, 278,964,000 lbs.; lead, pig, 732,152,538 lbs.; lead pipe, 1,058,280 cwt.; nails, wire, 13,916,097
kegs; pig iron, 9,896,248 tons; tin (pig), 94,248,471 lbs.; silver, 151,969,144 ozs.; spelter, 464,903,059 lbs.; steel
billets, 4,972,179 tons; steel rails, 3,025,009 tons; tin plate, 12,968,174 cwt.




74

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

T able

16.—COMPARISONS OF WEIGHTED AND UNWEIGHTED IN D E X NUMBERS.

[1. From the report of the Senate Committee on Finance, Mar. 3,1893.
(Arithmetic means.

Year.

1860...........................................
1861...........................................
1862...........................................
1863...........................................
1864...........................................
1865...........................................
1866...........................................
1867...........................................
1868...........................................
1869...........................................
1870...........................................
1871...........................................
1872...........................................
1873...........................................
1874...........................................
1875...........................................
1876...........................................
1877...........................................
1878...........................................
1879...........................................
1880...........................................
1881...........................................
1882...........................................
1883...........................................
1884...........................................
1885...........................................
1886...........................................
1887...........................................
1888:...........................
1889...........................................
1890...........................................
1891...........................................

All articles
averaged
according
Simple
to
impor­
arithmetic tance,
cer­
means, all
tain
ex­
articles.
penditures
being
uniform.
100.0
100.6
117.8
148.6
190.5
216.8
191.0
172.2
160.5
153.5
142.3
136.0
138.8
137.5
133.0
127.6
118.2
110.9
101.3
96.6
106.9
105.7
108.5
106.0
99.4
93.0
91.9
92.6
94.2
94.2
92.3
92.2

By years, 1860 to 1891.]

Prices in 1860=100.)

100.0
95.9
102.8
122.1
149.4
190.7
160.2
145.2
150.7
135.9
130.4
124.8
122.2
119.9
120.5
119.8
115.5
109.4
103.1
96.6
103.4
105.8
106.3
104.5
101.8
95.4
95.5
96.2
97.4
99.0
95.7
96.2

All articles
averaged Difference Difference Difference
according
between
between
between
to impor­ simple and simple and first and
tance: 68.6
first
second
second
per cent of weighted
weighted
weighted
total ex­
averages.
averages.
averages.
penditure.

100.0
94.1
104.1
132.2
172.1
232.2
187.7
165.8
173.9
152.3
144.4
136.1
132.4
129.0
129.9
128.9
122.6
113.6
104.6
95.0
104.9
108.4
109.1
106.6
102.6
93.3
93.4
94.5
96.2
98.5
93.7
94.4

4.7
15.0
26.5
41.1
26.1
30.8
27.0
9.8
17.6
11.9
11.2
16.6
17.6
12.5
7.8
2.7
1.5
1.8
3.5
.1
2.2
1.5
2.4
2.4
3.6
3.6
3.2
4.8
3.4
4.0

6.5
13.7
16.4
18.4
15.4
3.3
6.4
13.4
1.2
2.1
.1
6.4
8.5
3.1
1.3
4.4
2.7
3.3
1.6
2.0
2.7
.6
.6
3.2
.3
1.5
1.9
2.0
4.3
1.4
2.2

1.8
1.3
10.1
22.7
41.5
27.5
20.6
23.2
16.4
14.0
11.3
10.2
9.1
9.4
9.1
7.1
4.2
1.5
1.6
1.5
2.6
2.8
2.1
.8
2.1
2.1
1.7
1.2
.5
2.0
1.8

[2. From Bulletin of the Department of Labor, No. 27, March, 1900. January of the years, 1890 to 1899.]
(Arithmetic means. Averages of 9 quarterly quotations, January, 1890 to January, 1892=100.)

Year and month.

1890, January...........................
1891, January...........................
1892, January...........................
1893, January...........................
1894, January...........................
1895, January...........................
1896, January...........................
1897, January...........................
1898, January...........................
1899, January...........................




All articles All articles
averaged
averaged
according according Difference Difference Difference
to
impor­
between
between
to
impor­
between
All articles tance, cer­ tance,
com­
simple
simple
first and
simply
prising
and first and second
tain
ex­
second
averaged. penditures
68.6 per
weighted
weighted
weighted
being con­
cent of
averages.
averages.
averages.
total ex­
sidered
uniform. penditure.
102.0
100.6
96.5
97.2
89.6
81.7
85.2
82.0
83.3
86.5

100.1
102.2
100.0
103.4
97.5
93.5
92.8
90.3
91.0
91.0

100.2
103.2
100.1
105.0
96.4
90.5
89.5
85.9
86.8
86.8

1.9
1.6
3.5
6.2
7.9
8.8
7.6
8.3
7.7
4.5

1.8
2.6
3.6
7.8
6.8
5.8
4.3
3.9
3.5
.3

0.1
1.0
.1
1.6
1.1
3.0
3.3
4.4
4.2
4.2

75

THE MAKING AND USING OF INDEX NUMBERS.
T a b l e 1 6 .— COMP ARISON S

OF W EIGHTED AND UNWEIGHTED IN D E X N U M BERS-Conc.

[3. From Wholesale Prices, Canada, 1913. Report by R. H. Coats. By years, 1890 to 1913.]
(Arithmetic means.

Year.

Un­
Weighted weighted
index
index
number. number.
112.0
111.3
104.9
103.9
97.2
95.6
90.6
89.9
95.5
99.0
105.8
106.0

891
892
893
894
895
896
897
898
899
900
901

Average prices, 1890 to 1899=100.)

Differ-

110.3
108.5
102.8
102.5
97.2
95.6
92.5
92.2
96.1

1.7

2.8
2.1
1.4

1.9
2.3
.6

100.1

1.1

108.2
107.0

2.4

1.0

Year.

Un­
Weighted weighted
index
index
number. number.

1902
1903.
1904.
1905.
1906
1907.
1908.
1909.
1910.
1911.
1912.
1913.

109.6
109.7
110.6
113.8
120.1
129.2
125.1
126.3
128.0
131.1
143.9
139.6

Differ­
ences.

109.0
110.5
111.4
113.8
120.0
126.2
120.8
121.2
124.2
127.4
134.4
135.5

0.6
.8
.8
.1
3.0
4.3
5.1
3.8
3.7
9.5
4.1

[4. From new computations by the Bureau of Labor Statistics.1]
(Arithmetic means. Average prices in 1890 to 1899—100.)

37 food products.

13 farm products.

Year.

14 metallic products.

Weighted
Weighted
Weighted
by esti­
by esti­
by esti­
mated ex­ Dif­
mated ex­ Dif­
mated ex­ Dif­
Un­
Un­
Un­
pendi­
pendi­
fer­
fer­ weighted. pendi­
fer­
weighted.
tures
weighted.
tures
tures
upon each ences.
upon each ences.
upon each ences.
article in
article in
article in
1909.
1909.
1909.

1890.....................
1891.....................
1892.....................
1893.....................
1894.....................
1895.....................
1896.....................
1897.....................
1898.....................
1899.....................
1900.....................
1901.....................
1902.....................
1903.....................
1904.....................
1905.....................
1906.....................
1907.....................
1908.....................
1909.....................
1910.....................
1911.....................
1912.....................
1913.....................

113
124
112
106
96
93
78
84
97
99
109
117
130
120
130
125
122
139
135
150
161
166
173
152

109
117
105
107
94
95
86
93
97
98
109
115
129
120
128
123
124
136
135
154
165
150
164
161

4
7
7
1
2
2
8
9
0
1
0
2
1
0
2
2
2
3
0
4
4
16
9
9

114
116
105
112
99
95
83
87
93
98
108
110
114
110
113
110
115
120
122
124
129
128
137
133

114
114
103
111
97
94
86
90
96
96
100
102
108
104
110
109
106
112
119
126
127
125
137
127

0
2
2
1
2
1
3
3
3
2
8
8
6
6
3
1
9
8
3
2
2
3
0
6

128
118
110
102
88
88
93
82
83
124
124
114
114
114
105
116
131
138
103
109
111
111
120
119

i

131
116
107
98
84
88
91
80
81
124
123
113
114
113
102
113
130
140
108
107
108
103
114
115

3
2
3
4
4
0
2
2
2
0
1
1
0
1
3
3
1
2
5
2
3
8
6
4

1 See explanations in footnote, p. 72.

If rational weighting is worth striving after, then, by what criterion
shall the relative importance of the different commodities be judged ?
That depends upon the object of the investigation. If, for example,
the aim be to measure changes in the cost of living, and the data be
retail quotations of consumers’ commodities, then the proportionate
expenditures upon the different articles as representsd by collections
of family budgets make appropriate weights. If the aim be to study



76

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

changes in the money incomes of farmers, then the data should be
“ farm prices,” the list of commodities should be limited to farm
products, and the weights should be proportionate to the monetary
receipts from the several products. If the aim be to construct a
business barometer, the data should be prices from the most repre­
sentative wholesale markets, the list should be confined to com­
modities whose prices are most sensitive to changes in business pros­
pects and least liable to change from other causes, and the weights
may logically be adjusted to the relative importance of the com­
modities as objects of investment. If the aim be merely to find the
differences of price fluctuation characteristic of dissimilar groups of
commodities, or to study the influence of gold production or the issue
of irredeemable paper money upon the way in which prices change,
it may be appropriate to give identical weights to all the commodities.
If, on the other hand, the aim be to make a general-purpose index
number of wholesale prices, the question is less easy to answer.
One proposition, however, is clear. The prevalent practice of
weighting wholesale-price index numbers by figures drawn from family
budgets is to be deprecated. For family budgets do not show the
importance of wheat and cotton, of petroleum and spelter, of tar and
lime, of pig iron and hides, of brick and lumber; indeed, to apply
budget weights to half or more of the articles in any wholesale list is
impossible, or at best nonsensical. And to pretend that wholesaleprice index numbers when weighted on the basis of family expendi­
tures show fluctuations in the cost of living is to overtax the credulity
of those who know and to abuse the confidence of those who do not.
Allied to the family-budget method of weighting and yet vastly
better for wholesale-price index numbers is the “ aggregate expend­
iture” method.1 Here an attempt is made to ascertain the aggre­
gate sums of money laid out by the people of a whole country upon
the articles quoted and to adjust their weights upon this basis. Of
course the country as a whole buys raw materials, as single families
do not, and of course consumers’ commodities can be taken at their
aggregate values in wholesale markets. Similar in net effect is the
weighting on the basis of consumption practiced by the British
Board of Trade. For “ consumption is taken to mean any process
by which the commodity is substantially changed in character. In
other words, consumption in manufacture is recognized as well as
consumption by an individual.” 2 Somewhat different weights would
result if quantities or values produced were taken in place of quanti­
ties or values consumed. Mr. Walsh thinks it best to combine these
j 1 See G. H. Knibbs, Prices, Price Indexes, and Cost of Living in Australia. Commonwealth Bureau of
Census and Statistics, Labour and Industrial Branch, Report No. 1, pp. 11-14.
/ * Report on Wholesale and Retail Prices in the United Kingdom in 1902. London, 1903, p. 441. The
accuracy of the statistics upon which the Australian and British index numbers are based may be open
to question. Not the data, but the method is of interest here.




THE M AKING AND USING OF INDEX NUMBERS.

77

two criteria—that is, to take “ either the total product or the total
consumption according as the one or the other is the greater.” 1
Prof. Irving Fisher prefers “ an index number in which every article
or service is weighted according to the value of it exchanged at base
prices in the year whose level of prices it is desired to find.”2 On
this system the weight assigned to each article would be affected by
the number of times it changed hands on its way from producer to
final consumer. A variation of his plan is therefore represented by
the proposal to weight each article according to the quantity of it
which enters into the country’s commerce, irrespective of the fre­
quency with which it changes hands.
The practical consequences of adopting these different systems of
weighting may be illustrated by considering their application to
cotton, com, and coffee in the United States. Production weights
would give cotton much greater importance than consumption or
aggregate-expenditure weights, because so large a part of the Ameri­
can crop is exported and consumed abroad. Exchange weights
would be practically equivalent to production weights, because
practically all the cotton grown is sold by the planters and enters
into the commerce of the country, and very little cotton is imported.
On Prof. Fisher’s plan, however, the exchange weights would be some
multiple of the production weights, depending upon the average
number of American hands through which the cotton passed. In
the case of corn, production and consumption weights would sub­
stantially agree, for we import very little corn and export but a very
small percentage of the production. On the other hand, exchange
weights would be much less than either production or consumption
weights, because a large part of the corn crop is never sold, but
is consumed on the farms where it is grown. In the case of coffee,
production weights would be practically zero, while consumption
and exchange weights would correspond closely.
We are helped toward a choice among these rivals by common
agreement upon a slightly different point. In arranging any system
of weights except Prof. Fisher’s, double counting is to be avoided so
far as possible. For example, if cotton is counted at its full impor­
tance as a raw material, then cotton yarns and later cotton fabrics
made of the yarn can not be counted at their full importance with­
out assigning triple weight to the raw cotton which is represented at
these two successive stages of manufacture. Now, if this sensible
observation be applied to cases like those of corn, hay, etc., it casts
the die in favor of exchange weights. For if these articles, which
are used largely by the original producers in making things quite
i C. M. Walsh, The Measurement of General Exchange-Value. New York, 1901, p. 95.
* Irving Fisher, The Purchasing Power of Money, revised edition. New York, 1913, pp. 217 and 218.




78

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

different from corn and hay (for instance, pork and beef) are counted
at the full amount produced or consumed, and if their products (the
pork and beef) are also counted at the full amount produced or con­
sumed, there will be a great deal of double counting. Not all but much
of this duplication can be eliminated by counting only the amount
of corn and hay sold by the producers and letting the rest of these
articles produced and consumed get their proper representation under
the captions of pork, beef, etc.1
If for this reason exchange appears a rather better criterion of
importance than production, consumption, or a combination of the
two, it remains only to decide whether the number of times a thing
is exchanged should be recognized. Prof. Irving Fisher had good
cause to propose multiple counting, for he wanted an index num­
ber of prices for constructing the “ equation of exchange,” a mathe­
matical expression of the necessary equivalence between the total
volume of business done in a country and the total volume of pay­
ments effected by means of money and credit instruments. Of
course the oftener an article is sold and paid for the more important
it is as a factor in this equation. But it does not follow that the eco­
nomic importance of an article is greatly changed by reorganizing the
chain of business enterprises that deal in it. “ Integration of indus­
try,” as expressed in our trusts, does not make pig iron less signifi­
cant as an item in the country’s economic life, except in the sense
that it reduces the average number of transfers of ownership. The
quantity of the article that enters into exchange, then, irrespective
of the number of turnovers, is probably the most satisfactory gauge of
importance to apply in making general-purpose index numbers.
But anyone experienced in the search for statistical information will
need no warning that in the working out of weights along this line
many puzzling cases will arise in which consistency will be difficult to
maintain, to say nothing of wide gaps and many weak places in the
existing data being revealed.
Three interesting questions remain: Should the weights be sums
of money or physical quantities? Should the weights be changed
from year to year or kept constant % Should the weights be adjusted
to the importance of the commodities as such, or should there be
taken into account also the importance of the commodities as repre­
senting certain types of price fluctuations %
When relative prices are being used the weights should be reduced
to a common denominator. As multipliers, of course, weights may
be regarded as merely abstract numbers; but in studying the weights
i Of course, this same end might be attained without surrendering the production or consumption basis
if the rule against double counting of raw materials and products were made broad enough to include corn,
for example, as the raw material of pork; but needless to say there is little likelihood that the common
meaning of terms will be stretched to such an extent.




THE M AKING AND USING OF INDEX NUMBERS.

79

themselves it is necessary to have some common standard by which
the relative importance assigned to various commodities can be
accurately compared. The only common denominator for all com­
modities that is significant for economic ends and capable of quanti­
tative expression is money value. But it is ill advised to weight by
money values when actual prices are being used, for the common
denominator is already present in the quotations themselves. These
price quotations are best multiplied by the physical quantities of
the goods produced, exchanged, or consumed, as the case may be.
The argument in favor of changing the weights at frequent inter­
vals is that the relative importance of commodities is continually
varying. Hence a system of fixed weights applied to prices over a
long period of years is certain to be inaccurate for most of these
years, however accurately it was adjusted to conditions prevailing
when it was devised. The rejoinder is that an index number is pri­
marily a device for measuring changes in prices; if the weights are
revised it becomes a measure of two sets of changes, and no one can
tell what part of the net results is due to variations in prices and what
to variations in weights.1 Practically, then, the compiler must
choose between two evils—inaccurate weights and ambiguous price
measures. Sometimes he can minimize the first evil by collecting
data showing the average importance of his commodities over a
period of years, for these averages are less likely to go awry than
figures for a single year. And when he makes chain index numbers
he is free to revise his weights as often as he likes, since such series do
not profess to yield accurate comparisons except between successive
years. In other cases the least objectionable compromise is probably
to revise the scheme of weights, say, once a decade, and to show the
effect of this change by computing overlapping results for a few years
with both the old and new weights.2 A further practical reason in
favor of this compromise is found in the heavy expense of time and
labor required for frequent revisions of the weights.
To the third question, whether weights should be adjusted to the
importance of the commodities as such, or whether there should also
be taken into account the importance of these commodities as repre­
sentatives of certain types of price fluctuations, little* attention has
been paid. But the preceding section shows that this neglected
problem is both important and difficult. The prices of raw materials
behave differently from the prices of manufactured goods; among
the raw materials the prices of farm crops, of forest, animal, and
mineral products behave differently; there are also differences of
behavior between the prices of manufactured goods bought by pro­
1 See the criticism of index numbers made from import-export values, subsec. 3, p. 31.
2 Compare G. H. Knibbs, Prices, Price Indexes, and Cost of Living in Australia. Commonwealth
Bureau of Census and Statistics, Labour and Industrial Branch, Report No. 1, pp. xxiv and xlix.




80

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

ducers and by consumers, etc. An accurate measure of changes in
the level of all wholesale prices is not obtained unless all of the differ­
ent types of fluctuation, doubtless including types not yet definitely
recognized, are represented in accordance with the relative importance
of the commodities belonging to each.
How can such representation be attained ? If all the commodities
bought and sold could be included in the index number, it would
suffice to weight each by the criterion of its own individual importance.
Since that is out of the question, it is theoretically necessary to draw
from each part of the whole system of prices samples sufficient to
determine its characteristic fluctuations, and then to make sure that
each part of the whole system counts for the proper amount in de­
termining the final result. On this plan commodities would be
weighted simply as commodities in making the subtotals for each
recognized group, and these subtotals would be weighted again in
making up the grand totals.
Perhaps it is a counsel of perfection to urge such refinements in
systems of weighting. Certainly the difficulties to be encountered
are very great. Statistical knowledge is not complete enough to sup­
ply accurate data for weighting all the different parts of the system of
prices that are known to have characteristic peculiarities of fluctua­
tion. Nor have these different types and the commodities exhibiting
each been adequately studied. And puzzling difficulties are raised
by overlapping among the types—there are commodities that belong
in two places at once. But here is certainly a promising lead for
future efforts to improve present measurements of changes in the
price level. Even now it might be feasible by taking pains to secure
rough justice as between raw and manufactured commodities, and as
between raw vegetable, animal, forest, and mineral products. One
modest step in the right direction can readily be taken by any com­
piler of index numbers: He can make clear that his results do not
measure changes in the general level of wholesale prices accurately
when they are obtained without an effort to represent each part of the
field according to its due importance.
8. AVERAGES AND AGGREGATES.

Among all the problems involved in the making of index numbers
the one that has been discussed most thoroughly is the best form of
average to strike. Most of these discussions have come from men
interested in the mathematical side of statistics rather than in the
problem of ascertaining what changes have actually occurred in
prices. The practical makers of index numbers, on the contrary,
have seldom troubled themselves greatly about theoretical refine­
ments of method. Indeed, the two problems of finding out how
prices have actually changed, and finding the best method of me as-




THE M AKING AND USING OF INDEX NUMBERS.

81

uring changes, appeal to two types of interest, which are seldom
strongly developed in the same mind. The mathematical statistician
is likely to know little and care less about the field work of collecting
price quotations. To the practical statistician this field work is of
overshadowing importance, and the subsequent manipulation of his
data is a matter of secondary interest. Hence, a study of index
numbers as they are made need not carry one into long mathematical
flights.1
First, it should be recalled that certain compilers of index numbers
do not strike averages at all. The old form of the Economist index
and Gibson's present index, for example, are sums of relative prices.
More important are the series which dispense with the use of relative
prices for each commodity, and give results in the form of sums of
actual prices, or such sums thrown back into a series of relative num­
bers. These cases are still exceptional, however, and most index
numbers are made by finding some sort of average from the relative
prices of the commodities included.
The sort of average struck is almost always the arithmetic mean—
that is, the sum of the relative prices divided by their number.
Occasionally medians are used—that is, the midmost relative prices
which divide the whole number of cases into two equal groups. In one
famous investigation,2 geometric means were employed—that is, all
the relative prices for a given date were multiplied together and the
7&th roots of the products were extracted, n standing for the number
of commodities included. But Jevons has had few imitators. The
other standard forms of averages—the mode and the harmonic mean—
have been discussed frequently, but used seldom, in making index
numbers.3
For the geometric mean two great merits are claimed. First, unlike
the arithmetic mean, it is not in danger of distortion from the asym­
metrical distribution of price fluctuations. Chart 2 shows that in a
large collection of percentage variations from the prices of the pre­
ceding year, the extreme cases of rise run about twice as far up the
scale as the extreme cases of fall run down. Such a distribution is
characteristic of relative prices in general. Indeed, the case cited is
distinctly moderate; most collections of variations covering many
1 The best systematic discussions of averaging for thepurpose inhand are to be found in Prof. Edgeworth’s
papers referred to in footnote on p. 6; Irving Fisher’s The Purchasing Power of Money, revised edition,
1913, pp. 385-429; and C. M. Walsh’s The Measurement of General Exchange-Value, 1901.
2 W . S. Jevons, “ A serious fall in the value of gold ascertained,” 1863. Reprinted in his Investigations in
Currency and Finance, 1884, pp. 13-118.
* Concerning the properties of these averages see, for example, F. 2i2ek, Statistical Averages (translated
by W . M. Persons), and G. U. Yule, Introduction to the Theory of Statistics, pp. 120-123,128-129.
The “ crude mode” is that relative price which occurs most frequently in the data under examination, e. g.,
in Chart 2 it is “ no change.” The true mode is “ the value of the variable corresponding to the maximum
of the ideal frequency-curve which gives the closest possible fit to the actual distribution.” “ The harmonic
mean of a series of quantities is the reciprocal of the arithmetic mean of their reciprocals,”

94261°—Bull. 173—15----- 6




82

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

years would show a greater difference. Of course there is no limit to
the possible percentage of rise, while the possible percentage of fall
can not exceed 100. If the price of one article be doubled and the
price of another be halved, their relative prices will be 200 and 50.
Then an arithmetic mean will show a net rise of 25 per cent; for
200 + 50 = ^25.

But a geometric mean will show that the price level

has not changed; for V^OOX 50 = 100. Second, geometric means can
be shifted from one base period to another easily and without
inaccuracy. Suppose, for example, that the price of wheat falls
from $1 per bushel in 1913 to 50 cents in 1914, while the price of
com remains unchanged at 40 cents. Then the relative prices are—
(1) On the basis, prices in 1913 = 100:
1913
Wheat......................................................
Com.........................................................

100
100

1914
50
100

(2) On the basis, prices in 1914 = 100:
1913
Wheat.....................................................
Com.........................................................

200
100

1914
100
100

The arithmetic and geometric means are
(1) On the basis of prices in 1913:

1913..................
1914...................

Arithmetic means.

Geometric means.

(100+100)-4-2= 100
(50+100) -*-2= 75

V ioox 100= 100.00
V 50X100= 70.71+

(2) On the basis of prices in 1914:

1913...................
1914...................

Arithmetic means.

Geometric means.

(200+100)-4-2*= 150
(100+100)-5-2*= 100

7200X100=141.427100X100=100.00

Here the arithmetic means can not, but the geometric means can,
be shifted from the 1913 base to the 1914 base or vice versa by
simply dividing the index number for one year by that for the other.
That is, 100^75 = 133J, not 150; but 10070.71 = 141.42.1 By




*See the discussion of shifting bases, pp. 39-44.

THE M AKING AND USING OF INDEX NUMBERS.

83

shifting the base in this simple fashion geometric means can be made
to give direct comparisons between the price levels at any two dates
covered by the investigation, whereas with arithmetic means com­
parisons must always be made in terms of prices at the original base
period.
The chief objection to geometric means in an index number
intended for general use is that this form of average is unfamiliar and
therefore more likely to be misinterpreted than arithmetic means.
Further, geometric means do not have any direct bearing upon
changes in the purchasing power of money as do arithmetic means
and weighted aggregates of actual prices.1 Finally, geometric means
are more laborious to compute than arithmetic means or medians.
Instead of adding the relative prices just as they stand and dividing
the sums by their number, the computer must convert the relative
prices of every commodity into their logarithms, add these loga­
rithms, divide the sums by the number of logarithms, and look up
the natural numbers corresponding to the quotients.2 Statisticians
are the more loath to incur the extra labor of this process since the
special merits of the geometric mean are shared in part by certain
other forms of index numbers. Like geometric means, sums of
actual prices, or relatives made directly from them, can be shifted to
any base desired. Like geometric means, again, medians are not
more affected by cases of exceptionally great advances in price than
1This point is more fully explained on pp. 88 and 89.
2If relative prices are not needed for any other purpose, it is quicker to compute the geometric mean from
the logarithms of the successive actual prices and then to find the ratios between the results. But even
that is a somewhat longer process than calculating relative prices, casting them up, and dividing by their
number.
That geometric means can be computed either with or without the use of relative prices is readily shown.

Let po, px
V'o, p’x stand for the actual prices of n commodities in the two years o and x.

Po’
Then the relative prices of these articles in the year x on the basis of actual prices in the year o are
V
Jh.
Po * p V

V
1
p*

yo

The geometric mean of these relatives is

# ) © ' ( ? )
But this expression is equal to

7(0 00 • • • 00
7(p»)

(p'o) ■ ■

• (?>:)

And the latter expression, of course, is the ratio between the geometric means of the actual prices in the
two years.




84

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

by cases of exceptionally great declines. Hence in practice most
makers of index numbers who distrust arithmetic means abandon
relative prices altogether or use medians instead of taking to geo­
metric means.
Medians, indeed, have several distinguished champions among
theoretical writers.1 Of all averages they are the easiest to compute,
for a quick arrangement of the materials followed by simple count­
ing of the items takes the place of casting sums and dividing by the
proper number. And Prof. Edgeworth has recently argued that the
median is safer than the arithmetic mean when, as in the case of
index numbers, the items to be averaged are samples drawn from a
larger field. For, according to the theory of probabilities, the prob­
able error of the median can not in any case be much greater than
that of the arithmetic mean and in other cases it may be very much
less.2
But medians have their drawbacks. (1) They are not perfectly
reversible; that is, they can not always be shifted from one base to
another by simple division. (2) The median may not answer pre­
cisely to its definition when several of the items to be averaged have
identical values. For example, in Table 2 of this bulletin it often
happens that the median falls in a large group of precisely identical
figures, so that it ceases to be true that half of the cases are above
and half below the median.3 (3) Medians of different groups can not
be combined, averaged, or otherwise manipulated with ease as can
arithmetic means. For example, in making up its index number the
Bureau of Labor Statistics can add the sums used for making arith­
metic means of the relative prices of farm products, foods, cloths and
clothing, etc., and from the sum of these sums strike the grand aver­
age for all commodities. It could not handle medians in this con­
venient fashion; instead of combining the sums from the groups it
would have to combine the single commodities. Similarly, a reader
who finds arithmetic means of two groups in different sources can
compute the arithmetic mean of these means, provided the number
of items in each group be stated, with no greater error than that
arising from the dropping of fractions in the published data; but he
can not approximate except in the vaguest way the median of two
1 Compare, for example, F. Y . Edgeworth, “ Index numbers,” Dictionary of Political Economy, Vol. II,
p. 386; Irving Fisher, The Purchasing Power of Money, revised edition, p. 425; A. L. Bowley, Elements of
Statistics, second edition, p. 224. Walsh, however, prefers the geometric mean. See his Measurement of
General Exchange-Value.
2See his paper “ On the use of analytical geometry to represent certain kinds of statistics,’ * Journal of
the Royal Statistical Society, June, 1914, Vol. L X X V II, p. 733.
3 In 1891, for instance, 232 commodities are represented. One-half the list is 116. But 82 commodities
rose in price, while 106 fell. The remaining 44 did not change. Here not only the median but also the
6th decil fell in the group “ no change.” Of course the median here does not divide the whole number of
casesinto two equal parts any more than the 6th decil does.




THE M AKING AND USING OF INDEX NUMBERS.

85

medians.1 (4) When the number of items to be averaged is small,
medians are erratic in their behavior. For in such groups there is
often a considerable interval between the midmost relative price and
the relative prices standing next above it and next below. No change
in any of the items, large or small, can alter the position of the median
unless it shifts an item from the upper half of the list to the lower half,
or vice versa. But any change of this character, large or small, will
make the median jump over the whole interval between its former
position and that of the next highest or next lowest relative price,
unless the change happens to place a new item within these limits.
In large groups such erratic jumps are less likely to occur, because the
intervals between the median and its nearest neighbors are usually
slight.
Most of the advantages and defects of arithmetic means have been
mentioned incidentally, but it is well to list them all together. ( 1)
Arithmetic means, then, stand next to medians in ease of computation,
and even ahead of medians when the items are to be averaged first in
small and later in large groups. (2) They are perfectly definite in
meaning. (3) Their familiarity to all readers is a great advantage
in work intended for wide reading. (4) They are more represent­
ative averages than medians, being affected by any change in any of
the items in the group. (5) They can themselves be averaged and ma­
nipulated algebraically in various other ways.2 On the other side of
the score it must be said (6) that arithmetic means are liable to distor­
tion from the occurrence of one or two extremely high relative prices;
(7) that arithmetic means of relative prices can not consistently be
shifted from one base to another without recomputation in full,3
and (8) that they may conceivably give contradictory results con­
1 It is a convenient feature of arithmetic means computed from relatives based on average prices over a
period of years that the mean of these means for the base period must be 100—again barring discrepancies
caused by dropping fractions. For example, the arithmetic means of the Bureau of Labor Statistics index
numbers for the ten-year period 1890-1899 would always add up to 1,000.0, had all the fractions been kept
and had all commodities been quoted in every year of the decade. If medians made from these figures add
up to 1,000.0 in 1890-1899, it is accidental.
2 See, for example, G. U. Yule, Introduction to the Theory of Statistics, pp. 114-116.
s See subsec. 5, “ Base Periods.”




86

BULLETIN OF TH E BUREAU OF LABOR STATISTICS.

cerning the direction in which prices are moving, according as rela­
tive prices are computed on one base or on another.1
Concerning the numerical value of the three averages under dis­
cussion, it can be proved that the geometric mean is always less than
the arithmetic. On the other hand, the median may be either above
or below the arithmetic mean, and likewise either above or below the
geometric mean. For example, if the relative prices of the 145 com­
modities represented in the second index number of Table 8 be aver­
aged in these three ways, the results are as follows for 1913:
Geometric mean, 125.7; median, 126.9; arithmetic mean, 131.3.
A fuller study of the relations between medians and arithmetic
means is provided for by the following table.2 In the chain index
the two forms of average never quite coincide; the median is smaller
in 20 cases and larger in 3; it is also steadier than the arithmetic mean
in the sense that it indicates an average annual change of 2.22 per
i Take, for example, the following data:

Wheat, per bushel.....................................
Corn, per bushel........................................

1913

1914

$0.50
.48

$1.00
.24

1913

1914

Then compute index numbers on the basis 1913=100:

Wheat, relative prices................................
Corn, relative prices...................................

100
100

200
50

Index numbers................................

200
100

250
125

Also, compute index numbers on the basis 1914=100:
1913

1914

Wheat, relative prices.................................
Corn, relative prices.....................................

50
200

100
100

Index numbers................................

250
125

200
100

Thus it appears that prices were 25 per cent higher in 1913 than in 1914 and also that they were 25 per cent
higher in 1914 than in 1913. Much stress is often laid upon illustrations of this sort; but they are not seri­
ously damaging to the good repute of arithmetic means when properly interpreted. What they really say
is: The arithmetic mean variation of prices from 1913 to 1914 may conceivably be upward in percentages of
prices in 1913, and at the same time be downward in percentages of prices in 1914. No real inconsistencyis
involved in that statement to one who can keep the meanings of the two results in mind. It should be
added that cases in which such apparent inconsistency occurs, while common in theoretical discussions,
seldom if ever occur in the practical computation of wholesale-price index numbers. In retail-price indexes
they are not unknown. An example has been pointed out in the British Board of Trade’s reports upon
cost of living of the working classes. See the reviews by J. M. Keynes in the Economic Joumal, September
and December, 1908.
2 For numerical examples of geometric and arithmetic means computed from the same data, see F. Y .
Edgeworth, “ A defense of index numbers,” Economic Journal, Vol. V I (1896), p. 137, and A. W . Flux,
“ Modes of constructing index numbers,” Quarterly Journal of Economics, Vol. X X I (1907), p. 627.




87

THE M AKING AND USING OF INDEX NUMBERS.

cent from prices in the preceding year, as against 3.64 per cent for the
arithmetic mean. In the fixed-base series for 1890-1913, including
145 commodities, the median is likewise steadier than the arithmetic
mean, showing a smaller percentage of change except during the
middle nineties, when the price level was at its lowest. The second
series for these years illustrates the erratic character of the median
when used to average a small group of variations. Here the median
is greater than the arithmetic mean in 13 years, just the same in one
year, and less in 10 years. Moreover, it shows a greater average
change from one year to the next than the arithmetic mean. Finally,
the figures for prices during the period of irredeemable paper money
(1862-1878, inclusive) show how far arithmetic means may depart
from the medians when a few commodities attain very high relative
prices. The maximum difference occurs in July, 1864, when the
arithmetic mean exceeds the median by 42 points, or more than 20
per cent. This excessive difference is due to the high prices of cotton,
tar, and other southern products. It is precisely in cases such as
this that the median is distinctly safer to trust than the arithmetic
mean.
T able 17.—COMPARISONS OF MEDIANS AND ARITHMETIC MEANS AS AVERAGES OF

RELATIVE PRICES.
[Data from Bulletin of the Bureau of Labor Statistics, No. 149.J

Chain index number
(prices in preceding
year= 100).1
Year.
Medians.

1890...........................................
1891............................................
1892............................................
1893............................................
1894............................................
1895............................................
1896............................................
1897............................................
1898............................................
1899............................................
1900............................................
1901............................................
1902............................................
1903............................................
1904............................................
1905............................................
1906............................................
1907...........................................
1908............................................
1909............................................
1910............................................
1911............................................
1912............................................
1913............................................

db 0
-3 .1
± o
-7 .1
-2 .4
-1 .2
± o
+1.8
+5.5
+7.5
-1 .5
+2.2
+1.3
± o
+ .7
+5.1
+3.9
- 3 .8
± 0
+1.5
- .9
+1.0
+ .5

Arithmetic
means.

- 0.2
- 4.4
- .2
- 8.7
- 1.5
— 2.8
+ .2
+ 4.8
+10.4
+ 9.4
- 1.1
+ 4.6
+ 1.2
- .1
+ 2.9
+ 5.8
+ 6.0
- 5.6
+ 3.2
+ 4.1
- 1.9
+ 3.4
+ 1.2

Averages, 1890-1899..................
1900-1909..................
1910-1913..................
Average change from one year
to the next............................
1 Compare Tables 2 and 6.




2.22

3.64

Relative prices of 145
commodities (aver­
age prices in 18901899*=-100).3

Relative prices of 25
commodities (aver­
age prices in 18901899==100).8

Medians.

Arithmetic
means.

Medians.

112
111
107
104
96
94
90
91
94
100
109
107
110
,111
112
114
119
129
119
121
124
125
127
127

114
113
106
105
96
93
89
89
93
103
111
110
114
114
114
116
122
130
121
124
131
130
134
131

116
109
106
102
90
94
89
92
99
108
117
112
115
112
124
126
131
133
125
130
126
131
136
127

115
112
103
103
92
95
88
90
96
107
113
111
116
118
122
123
130
133
124
133
133
129
140
142

100
115
126

100
118
132

101
123
130

100
122
136

3.61

* Compare Table 8, second series.

4.13

Arithmetic
means.

5.70

* Compare Table 8, fifth series.

5.09

88
T

able

BULLETIN OF THE BUREAU OF LABOR STATISTICS.
1 7 . —COMPARISONS

OF MEDIANS AND ARITHMETIC MEANS AS AVERAGES OF
R E LA TIV E PRICES—Concluded.

[From W . C. Mitchell, Gold Prices and Wages under the Greenback Standard, pp. 59,60.]

92 commodities at wholesale (prices in 1860=100).

Year.

1860, January.
A p ril...
July----October.
1861, January.
A p ril...
July----October.
1862, January.
A p ril...
July----October.
1863, January,
A p ril...
July....
October.
1864, January
A p ril...
July—
October.
1865, January
A pril...
July----October..
1866, January
A p ril...
July----October.

Me­ Arith­
dians. metic

100

102

100

102
100
102
100
98
95
103
115

100

100
100
96
96
97
100
100
100
111

125
137
134
135
156
169
194
200
216
190
158
175
182
173
181
173

112

120
126
142
160
155
155
179
197
236
239
248
206
183
205
199
186
191
188

Year.

1867, January.
A pril....
July----October.
1868, January.
April__
J u ly ....
October.
1869, January.
A pril....
July----October.
1870, January.
April__
J u ly ....
October.
1871, January.
April....
July----October.
1872, January.
April....
July----October.
1873, January
April...,
July----October.

Me­ Arith­
metic
dians. means.
169
166
150
162
158
162
154
159
159
159
158
153
147
140
132
135
133
131
130
129
133
140
130
133
135
137
130
131

179
175
170
172
171
176
165
166
165
165
158
157
152
146
145
143
142
140
137
139
141
145
139
143
142
144
140
140

Year.

1874, January
April__
July......
October.
1875, January
A pril....
July......
October.
1876, January
April__
July......
October.
1877, January
April__
July......
October.
1878, January
April__
July......
October.
1879, January
April__
July......
October.
1880, January
April....
July----October.

Arith­
Me­ metic
dians. means.
130
129
130
130
127
125
121
120
117
115
110
108
114
108
100
102
99
96
90
94
88
84
85
95
ias
107
102
101

140
141
138
138
138
132
129
127
122
122
118
117
121
118
114
110
107
105
99
102
100
99
98
103
114
116
110
111

Average change from one quarter to the next: Medians, 5.66 points; arithmetic means, 5.65 points.

Wise choice of the average to use in making an index number,
then, involves careful consideration of the materials to be dealt with
and of the purpose in view. ( 1) If that purpose be to measure the
a v era g e r a tio o f ch a n g e in prices, the geometric mean is the best,
indeed, in strictness, it is the only proper average to employ. For,
alone among our averages, the geometric mean always allows equal
influence to equal ratios of change in price, quite irrespective of the
previous levels of the prices in question, the amounts of money rep­
resented by the changes themselves, or any other factor. As has
been said already, in a geometric mean the doubling of one price is
precisely offset by the halving of another price—though if the two
prices were originally the same the rise amounts in money to twice
the fall. And further changes of 10 per cent from the two new prices
will again be precisely equal in their influence upon a geometric
mean, although 10 per cent of the price that has doubled represents
a sum of money four times as great as 10 per cent of the price that
has been halved. (2) But these same examples show that geometric
means are not proper averages for measuring alterations in the
amount of money that goods cost. And as a rule our interest does
center in the money cost of goods rather than in the average ratio




THE M AKING AND USING OP INDEX NUMBERS.

89

of changes in price. For example, when we are investigating the
increased cost of living, the doubling of one item in the family budget
may well be twice as important as its halving; and when we are
studying the “ relation of prices to the currency, a large upward
variation should count for more than a small downward variation,
for it requires more currency.” 1 For such purposes the arithmetic
Chart 11—A COMPARISON OP MEDIANS AND ARITHMETIC MEANS OF 145 COMMODI­

TIES. (BASED ON TABLE 17.)

mean is the logical average to use. (3) Frequently, however, the
very fact that an article has advanced greatly in price cuts down its
market, so that the increase in money cost represented by the arith­
metic mean exists on paper rather than in fact.2 When such cases
of extreme advance are numerous among the relative prices to be
averaged, the median may give more significant results than the
arithmetic mean. (4) When the number of commodities included
1 Irving Fisher, The Purchasing Power of Money, revised edition, p. 426, note 2.
2 Such cases might he met b y reducing the weight allowed the article in question; "but we have seen that
revising weights blurs the meaning of the index number, b y making it impossible to say how far the final
results measure the change in prices and how far they measure the change in weights. See p. 79.




90

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

in the index number is small, however, medians are likely to prove
highly erratic, representing less the general trend of prices than the
peculiarities of the data from which they are made. (5) If the index
number is designed for the public at large, the familiarity of arith­
metic means is an argument in their favor; but it counts for nothing
in the case of figures intended for specialists. (6) Often the useful­
ness of a new index number may be enhanced without detriment to
its special purpose by throwing it into a form directly comparable
with that of index numbers already in existence. Then, of course,
not only the form of average but also the base period employed in
making the existing series has special claims for imitation. ( 7) Finally,
the desirability of making index numbers that can be shifted from one
base to another deserves far more consideration than is commonly
accorded it. On this count the score is in favor of the geometric
mean. If geometric means were invariably used, all index numbers
could readily be compared with one another, whatever the bases on
which they were originally computed. And that would be a great
gain to all students of prices.
No single form of average made from relative prices, then, is with­
out its merits and its defects. Can we not escape the necessity of
relying upon any one of them by giving up the use of relative
prices and falling back upon aggregates of actual prices ?
Index numbers made on this latter plan practically compel the
compiler to adopt a systematic scheme of weighting. For the hap­
hazard weighting involved in merely adding up the raw quotations
for different commodities in terms of their ordinary commercial
units is far more dangerous than the haphazard weighting involved
in using the same materials after reduction to relative prices.1 It
is also true that sums in dollars and cents are likely to run in amounts
awkward for comparison; but these sums can quickly be turned
into a series of relatives on the scale of 100. The same device meets
the objection that the introduction of new commodities, necessary
at intervals in any large index number that is kept up to date, dis­
turbs a sum of actual prices more than it disturbs an average of relative
prices. This statement is valid because the quotations for new com­
modities, however adjusted, are just so much added to the old sum;
while the relative prices of new commodities may be either above
or below the old average, and often exercise but a trifling net effect
upon its value. But by noting the ratios between the sums of actual
money which include and which exclude the new commodities, and
by using these ratios to adjust successive aggregates, the compiler
i See the example from Hunt’s Merchants’ Magazine given in subsection 4. However, a very rough
system of weights based upon guesswork may give quite as good results as the haphazard weighting of
relative prices. Prof. Irving Fisher suggests to the writer a “ lazy man’s index number,” made by adding
actual prices for ordinary commercial units, with their decimal points shifted forward or backward, or
left unchanged, according to the estimated importance of each article.




TH E M AKING AND USING OF INDEX NUMBERS,

91

meets this difficulty quite as well as if he were averaging relatives
from the start.
The technical difficulties attending the construction of index num­
bers made of actual prices, then, can be surmounted. Offsetting these
difficulties are numerous and substantial advantages. Aggregates of
money prices weighted according to the importance of the several arti­
cles are as easy to understand as arithmetic means of relative prices.
They are less laborious to compute than any other form of weighted
series, for no relative prices are used; the original quotations are
multiplied directly by the physical quantities used as weights, and
the products added together. They are not tied to a single base
period; but from them relative prices can quickly be made upon
the chain system or any fixed base that is desired, and these relative
prices themselves can be shifted about at will as readily as geometric
means.1 Hence they are capable of giving direct comparisons
between prices on any two dates in which an investigator happens
to be interested. Hence, also, they can be compared with any index
numbers covering the same years, on whatever base the latter are
computed. Their meaning is perfectly definite—which is not always
true of medians. They can not be made to give apparently incon­
sistent results like arithmetic means. When published as sums of
money, they can be added, subtracted, multiplied, divided, or aver­
aged in any way that is convenient. When weighted on a sound
system, they can not be unduly distorted by a very great advance
in the price of a few articles, and yet, unlike medians, they allow
every change in the price of every article to influence the result. In
fact, they combine most of the merits and few of the defects charac­
teristic of the various methods of averaging relative prices.
i The legitimacy of shifting these relatives b y the “ short” method is best shown by the use of symbols.
Let po, Px, py ^represent the money prices of the two commodities p and pf in three years
P'oyP'xtP'ylo, x, and y.
Then the sums of these actual prices will be—
p o+ p 'o in the year o.
p z + p 'x in the year x.
p y + p 'y in the year y.
Relative prices in the year x computed from these sums will be—
the basis of prices in the year o, and
the basis of prices in the year y.
Relative prices in the year y will be—
" [ " / on the basis of prices in the year o.
Now the relative price in the year x, computed on the basis of prices in the year o, can be turned into
the relative price for the year x on the basis of prices in the year y, by dividing the relative for the year
x on the basis of prices in the year o by the relative for the year y on the basis of prices in the year o. For
p x + p 'x . p y + v 'y j p x + p 'x
p o + p 'o ’ p o + p ’o p y + p 'y

The reason why ordinary arithmetic means of relative prices can not be consistently shifted to another
base by this simple method is explained in subsec. 5, p. 39.




92

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

But the main issue has still to be faced. Averages of relative
prices and aggregates of actual prices are different kinds of things.
Which kind do we want in an index number ? Do we wish to know
how certain sample prices have changed on the average, or do we
wish to know how the total cost of a sample bill of goods has changed ?
This is practically the same question we considered on pages 88 and
89 in discussing how best to average relative prices. And the answer
given there is valid here. If our interest really lies in measuring
average ratios of change, then geometric means are best. But ( 1)
the unfamiliarity of this average outside technical circles is itself an
objection to measuring average changes in an index number designed
for wide use, and (2) a measure of change in the money cost of goods
probably serves more uses than a measure of average ratios of change
in prices. Now, the weighted aggregate of prices is the best measure
of change in the money cost of goods; it is better in several ways
than the simple arithmetic mean of relative prices, and in addition
it has all the merits of the latter form of average. For the relatives
which can be computed from these aggregates with little trouble are
identical with arithmetic means of relative prices, when the latter are
weighted by the money values of the physical quantities used as
weights for the corresponding actual prices.
This identity of the variations of a weighted aggregate of actual
prices and the arithmetic-mean variations of similarly weighted rela­
tive prices can readily be demonstrated. Suppose that we have
collected the price quotations and the weights to be used in an index
number, and have decided what period to make the base for com­
parisons. Then if we want an aggregate of actual prices, we merely
multiply the quotations of each commodity at each date by the
physical quantities used as weights, and add these products. To
measure the variations of these aggregates in terms of prices at the
base period, we have only to divide the aggregate for each period by
the aggregate for the base period. But if we plan to make a weighted
arithmetic mean of price variations, we begin by turning the quota­
tions into relative prices. That is, we divide the actual price of each
commodity at each date by its price in the base period. Then we
weight these relatives, not by physical quantities as in the first case,
but by the money values of the physical quantities at the prices of
the base year. But in this step the prices of the base year, which
were just used as divisors to get relative prices, are used again as
factors by which the relative prices are multiplied. Hence our results
are the same as if we had neither multiplied nor divided by the prices
of the base year; in other words, the same as if we had multiplied the
quotations of each commodity in each year by the physical quantities
used as weights. But that is just what we did when we set out to
make an aggregate of actual prices. So far, then, the two processes




TH E M AKING AND USING OF INDEX NUMBERS.

93

are identical in their outcome. And the remaining steps are also the
same. The products must be added, and the sums divided by the
physical quantities used as weights times the actual prices of the base
year. Therefore, to make relative prices from aggregates of actual
prices is a shorter way of getting the same results as are obtained by
making similarly weighted arithmetic means of relative prices.1
In addition to the advantages peculiar to themselves, then, aggre­
gates of actual prices can readily be given all the advantages claimed
for weighted arithmetic means of relative prices. This combination
of qualities makes them the most desirable type of general-purpose
index numbers.2
V.— A COMPARISON OF THE LEADING AMERICAN INDEX
NUMBERS FOR THE YEARS 1890 TO 1913.

Most of the threads running through the preceding sections can
be woven into a comparison of the best-known index numbers cur­
rently published in the United States—a comparison having intrin­
sic interest of its own, as well as making a fitting summary of the
introduction to this bulletin. Much repetition of conclusions already
stated will be necessary, but repetition makes the essence and the
usefulness of summaries.
1 The explanation given in the text may be put in the form of algebraic formulae for readers willing to
study symbols.
Let Vo, Vx\
P'o, V'x [represent the prices of the commodities from which an index number is to be made in the
VUf PnI kase year o and in some other year designated by the subscript x.
Let q, q', and q» respectively represent the physical quantities of these commodities to be used as
weights.
Then an unweighted arithmetic mean of relative prices is represented by the following formula, in
which n stands for the number of commodities included:
/

vn

P'o

vn

JPJLjJPJiji. . . ._L*
Vo

_________ _0
n

A weighted aggregate of prices reduced to relatives is represented by the following formula:
px q+p'x

pn
x

po q+p'o q'+ . • . pnQ qn
A weighted arithmetic mean of relative prices is represented by the following formula:

(? V o )+ P 0 <*' P'o) + • • • % « » P*0)

_______________ ____ f_0_____
Po q + p'oq' + . . . . pn
o qn

But in the numerator of this fraction, po, p'o, and p® cancel out. Then formula (3) becomes identical
with formula (2). That is, the weighted aggregate of prices gives the same results when turned into
relatives as the weighted arithmetic mean of relative prices, and gives them with less work.
2 Because of the disadvantages of calculating index numbers from relatives the Bureau of Labor Statis­
tics has discontinued that method and is now constructing its wage index numbers from weighted aver­
ages of money wages, hours of labor index numbers from weighted averages of actual hours worked, and
priee index numbers from weighted aggregates of money prices. The quantities entering into exchange
in the census year 1909 have been taken as the weight for wholesale prices. In making relatives from
these aggregates, the fixed base 1890-1899 has been discarded in favor of the last completed year.




94

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

1. ANALYSIS OF THE SIMILARITIES AND DIFFERENCES.

Seven index numbers are available for the analytical study pro­
posed: Dun’s, Bradstreet’s, the Annalist’s, Gibson’s, made from the
original list and from the present list of commodities, and two forms
of the series compiled by the Bureau of Labor Statistics.
The first step toward comparing index numbers is to throw them
into similar form and establish them upon a common base. Both
forms of the bureau’s series 1 and the Annalist series are arithmetic
means of relative prices on the base, average prices in 1890-1899=
100. So, likewise, is the first Gibson series, which is made by the
writer from the bureau’s relative prices for the 50 commodities included
in Gibson’s original list.2 These four series accordingly are compara­
ble without more ado.
Not so the remaining three series. Gibson’s present index num­
ber is a sum (not an average) of relative prices, originally computed
on the 1890-1899 basis, but shifted for splicing upon Dun’s series as
it stood in 1907. This shift was effected by multiplying the sums of
relative prices by a number which made the Gibson figures in Jan­
uary, 1907, almost the same as the Dun figures. Hence the Gibson
series can be put back on the 1890-1899 basis by dividing the pub­
lished results by this multiplier.3 Dun’s and Bradstreet’s series are
sums of actual prices, and accordingly have no base of their own,
but may be shifted to any base desired—in this case 1890-1899=
100. Dun’s figures for this decade average $84.32. By dividing the
published figures by this sum and multiplying the results by 100 we
can make a new series strictly comparable with the rest of our mate­
rial. Shifting Bradstreet’s series is less satisfactory, because it does
not begin until 1892. The best that can be done is to equate Brad­
street’s average for 1892-1899 with the bureau’s average for these
years—that is, to put $6.7785=97.1—and then to apply the rule of
three.4
1For the differences between these two forms, see pp. 46 and 47, and footnote.
This list is given in the second footnote on p. 47. The original designer of this series, Dr. J. P. Norton,
confined himself to commodities quoted by the bureau, but he changed the form of the bureau’s relatives to
make them correspond with Dun’s figures in 1907. (See the reference in the next footnote.) It is, of course,
easy to use his list and the bureau’s relatives to make an index number covering the years 1890 to 1913
on the 1890-1899 basis.
W xD
s This multiplier was obtained from the formula— Jj— in which W =the Dun weight for foods, 0.50;
2

D =the average of Dun’s results for 1890-1899, namely, $84.3,* and C=the number of foods included in the
list, namely, 22. This formula has the value JL.9159. To divide the published results by this multiplier
restores the original sums of relative prices. T o get arithmetic means instead of sums, we must divide by
1.9159X22= 42.1+. SeeJ. P. Norton, “ A revised index number for measuring the rise in prices,” Quarterly
Journal of Economics, August, 1910, vol. 24, pp. 753,754.
4 No violence is done by this procedure to Bradstreet’s series, but the comparison is not quite satisfactory,
because our other series were not worked out on the basis, prices in 1892-1899*97.1, and would probably
have shown somewhat different results if they had been. The only way to make a perfect comparison
with Bradstreet’s figures would be to recompute all the relative prices that enter into the Bureau of Labor
Statistics’ index number and the other index numbers here derived from it on the basis of 1892-1899= 100.
In other words, we are here practically applying the short method of shifting a base, which has been shown
to involve inaccuracies. See Sec. IV, subsec. 5,




95

THE M AKING AND USING OF INDEX NUMBERS.

These seven series, then, all in comparable form, are assembled in
Table 18.1 The second and third sections of the table facilitate
certain detailed comparisons of greatest interest.
A cursory examination shows that these seven series, made by
five independent organizations, have a marked family resemblance—
as was found to be the case with the six index numbers made from
the Bureau of Labor Statistics data and presented in Table 8 . They
all say that prices fell heavily in 1890-1896, that prices rose in 18971900, that they wavered rather uncertainly in 1901-1904, that they
rose sharply again in 1905-1907, and once more in 1908-1913. They
all agree that the general level about which these oscillations clustered
was distinctly higher in 1910-1913 than in 1900-1910, and higher in
1900-1910 than in 1890-1899. About the major facts of price fluctu­
ations, in short, the testimony from different sources is so unanimous
that one can scarcely doubt its validity.
i The annual averages, made from the original figures published by Dun, Bradstreet’s, and Gibson, run
as follows:

Year.

1890.
1891.
1892.
1893.
1894.
1895.
1896.
1897.
1898.
1899.
1900.
1901.
1902.
1903.
1904.

Dnn.o

$90.9
92.2
90.0
92.4
84.7
81.3
76.0
74.0
78.9
82.8
93.4
95.9
100.4
99.0
100.2

Brad­ Gibson.
street’s.

$7.78
7.53
6.68

6.43
5.91
6.12
6.57
7.21
7.88
7.57
7.88
7.94
7.92

$43.4
50.8
45.3
46.0
43.4
42.0
34.0
34.7
38.7
41.6
44.2
44.5
53.5
49.0
48.3

Year.

190 5
190 6
190 7
190 8
1 9 0 9 ....:....
191 0
191 1
191 2
191 3
191 4
Averages:
1890-1899
1900-1909
1910-1913

Dun.®

Brad­
street’s. Gibson.

$100.6 $8.10
105.3
111.8
109.9
117.8
119.2
116.8
124.4
120.9
122.2

8.42
8.90
8.01
8.52
8.99
8.71
9.19
9.21

84.32
103.43
120.33

66.78
8.11

$47.3
49.8
50.9
54.2
59.2
59.3
56.9
62.6
58.1
42.9
51.0
59.2

» Mr. Douglas R. Little courteously supplied Dun’s figures for 1907-1914 in advance of their publication
in Dun’s Review.
6 Average of 1892-1899.




1 8 .— A

COMPARISON OF THE LEADING AMERICAN IN D E X NUMBERS FOR THE
Y EA R S 1890 TO 1913.

Year.

Number of commodities.........
1890...........................................
1891..........................................
1892..........................................
1893..........................................
1894...........................................
1895...........................................

1899..........................................
1900..........................................
1902..........................................

1906..........................................
1907..........................................
1908..........................................
1909..........................................
1910...........................................

Averages, 1890-1899.................
1900-1909.................
1910-1913.................

BradDun’s
street’s
index
index
number, number,
calculated calculated
on base, on base,
average average
prices for prices
for
1890-1899 1892-1899
= 100.
=97.1.

310?
108
109
107
110
100
96
90
88
94
98
111
114
119
117
119
119
125
133
130
140
141
139
148
143
100
123
148

96

Gibson
index
number,
original
list,
average
prices for
1890-1899
= 100.

Bureau
of Labor
Statistics’
index
number,
average
prices for
1890-1899
= 100.

Revised
.Bureau
of Labor
Statistics’
index
number,1
average
prices for
1890-1899
= 100.

Gibson
index
number,
present
form,
calculated
on base,
average
prices for
1890-1899
= 100.

Annalist
index
number,
average
prices for
1890-1899
=100.

50

242±

145

22

25

111
108
96
92
85
88
94
103
113
108
113
114
113
116
121
127
115
122
129
125
132
132

114
114
105
105
94
94
87
89
95
103
112
109
116
115
116
118
123
132
125
132
135
129
138
138

113
112
106
106
96
94
90
90
93
102
111
109
113
114
113
116
123
130
122
125
130
126
130
130

114
113
106
105
96
93
89
89
93
103
111
110
114
114
114
116
122
130
121
124
131
130
134
131

103
121
108
109
103
100
81
82
92
99
105
106
127
116
115
112
118
121
129
141
141
135
149
138

3 97
116
130

100
120
135

100
118
129

100
118
132

100
119
141

2 109
119
108
116
102
95
280
84
92
93
299
105
117
107
109
110
115
120
126
134
to

T able

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

CO

96

2 131
2 143
2 140
100
114
138

1 For explanation of the reasons for and methods of revising the bureau’s index number, see pp. 46 and 47.
2 As computed by the Annalist.
The remaining figures in this column were computed from the
Bureau of Labor Statistics’ relative prices for the articles on the Annalist list. The results of this com­
putation agree with the Annalist’s results except in 1913, when there is a discrepancy—139 as against 140.
a Average of 1892-1899.




97

THE M AKING AND USING OF INDEX NUMBERS.
T a b l e 1 8 .— A

COMPARISON OF THE LEADING AMERICAN IN D E X NUMBERS FOR THE
Y EA R S 1890 TO 1912—Continued.

Number o f points by which the other index numbers were greater (+) or less ( —) than
the Bureau o f Labor Statistics* series, in each year from 1890 to 1913.

Year.

1890..........................................
1891..........................................
1892..........................................
1893..........................................
1894..........................................
1895 ........................................
1896..........................................
1897 ......................................
1898 ......................................
1899 ......................................
1900
...................................
1901
...................................
1902
...................................
1903
...................................
1904
...................................
1905..........................................
1906..........................................
1907..........................................
1908-.........................................
1909..........................................
1910..........................................
1911..........................................
1912..........................................
1913........................................

Gibson
Revised index
BradDun's street’s
Gibson Bureau number,
index
index
index of Labor
Annalist
number, number,
number, Statis­ present
index
calcu­
form, number,
calcu­
original
tics’
calcu­ average
lated on lated on
index lated
list,
base,
base, average number, base,on prices for
average average
prices for average average 1890-1899
prices for pricesfor
for
1890-1899 1892-1899 1890-1899 prices
for =100.
=100. 1890-1899 prices
=100.
= 100. 1890-1899
=97,1.
=100.
+
+
+
4±
—
+
—
±
+
+
+
+
+
+
+
+
+
+
+
+
+

5
3
1
4
4
2
0
2
1
4
0
5
6
3
6
3
2

3
8
15
11
13
18
13

132
Arithmetic sums....................
Algebraic s u m s ................... +104
Average differences com­
puted from the—
5.5
Arithmetic sums.............
+ 4.3
Maximum differences............
Minimum differences.............

+ 18
± o

+
+
±
—
—
—
+
+
+
—
±
±
±
±
—
—
—
+
+

2

3
7
3
1
1
2
2

42
— 12

+
+
—
—
—
±
—
—
+
+
+
±
+
+
+
+
±
+
+
+
+
+
+
+

1
2
1
1
2
0
3
1
2
1
1
0
3
1
3
2

0
2

3
7
5
3
8
8

60
+ 44

+
+
±
—
db
—
—
—
±
+
±
+
+
±
+
±
—
±
+
+
+
+

1
1
0
1
0
1
1
1
0
1
0
1
1
o
1
0
1
0
1
1
1
4
4
1

— 10
+ 9
+ 2
+ 3
+ 7
+ 6
— 9
— 8
— 1
— 3
— 6
— 3
+ 14
+ 2
+ 2
— 4
— 5
— 9
+ 7
+ 16
+ 11
+ 9
+ 19
+ 8

—
+
+
+
+
+
—
—
—
—
—
+
—
—
—
—
+
+
+
+
+
+

4
7
2
10
6
1
10
6
1
9
12
4
4
7
4
6
8
10
.4
9
7
5
13
10

11
12
6
11
9
8
10
8
3
10
14
9
14
10
10
9
10
13
15
19
12
14
19
13

23
9

173
+ 57

159
— 3

269

+

7.2
2.4

—

1.9
.5

+

2.5
1.8

+

1.0
.4

+

±

7
0

+
±

8
0

+
±

4
0

+ 19
- 1

94261°—Bull. 173—15----- 7




5
2
0
2
5
2
1
1
2
1
0
0
0
0

Mini­
Maxi­
mum
mum
differ­
differ­
ence be­ ence be­
tween
tween
any two any two
index
index
num­
num­
bers.
bers.

—

6.6
.1

+ 13
+ 1

11.2
19
3

1

1
1
1
1
5

0.2
1
±0

98

BULLETIN OP THE BUREAU OF LABOR STATISTICS.

T able 18.—A COMPARISON OF TH E LEADING AMERICAN IN D E X NUMBERS FOR THE
Y EA R S 1890 TO 1912—Concluded.
Number o f points by which each index number rose ( +) or fell ( —) in each successive year .

BradDun's street’s
index index
number number
calcu- calcu­
lated
on
on
base,
average
prices prices
for
for
18901899
==100. =97.1.

Year.

Number of commodities.......
1890-189 1
1891-189 2
1892-189 3
1893-189 4
1894-189 5
1895-189 6
1896-189 7
1897-189 8
1898-189 9
1899-190 0
1900-190 1
1901-190 2
1902-190 3
1903-190 4
1904-190 5
1905-190 6
1906-190 7
1907-190 8
1908-190 9
1909-1910 .
1910-191 1
1911-191 2
1912-191 3

310*
+

- 2

+ 5
-

2

+
±

2
0

+6
+ 8
- 3

50

242±

+6
+ 9
+10
- 5
+ 5
+ 1

- 1

+ 3
+ 5

+6

1

- 9

-

-11

-1 0

±

0

± o

-

7

+2
+6
+8
+ 9
- 3
+ 7
-

1

+1

+2

+ 5

+ 9

6

±0
2

-

-

4

±0

+ 3

+ 9
+ 9

2

-

+ 4
+1
- 1
+ 3

+ 7
+ 7

145
-

1

-

7

- 1
- 9
-

3

- 4
± 0
+ 4
+10

+8

- 1
+ 4
± 0
± 0
+ 2
+ 6
+ 8
- 9

5

-1 2
+ 7
+ 7
- 4
+ 7
± 0

-1 8
+43
- 3
+18
- 5

1 -2 6
+42
-1 2
+17
± 0
247

51

40

45

4.7

2 5. 6

4.9

4.0

35

221

17

+10
1
- 2
+

+ 9
-

Net rise in 1890-1913.............

- 3
-1 2
- 4
- 7
+ 3

Anna­
list
index
num­
ber,
prices
for
18901899

=

100.

Maxi­
mum
differ­
ence
be­
tween
any
two
index
num­
bers.

Mini­
mum
differ^
ence
be­
tween
any
two
index
num­
bers.

= 100. = 100.

±0

1

+ 3
-1 0
- 4
6
2
+6
+ 4
+13
+ 3

Net rise ( + ) or fall (—):
1890-1896.........................
1896-1907.........................
1907-190 8
1908-191 2
1912-1913.........................
Difference between highest
and lowest points..............
Average change from year to
year....................................

96

Re­ Gibson
vised index
Gibson Bureau
Bureau
num­
of
index Labor
of
ber,
numStatis­
Labor
present
ber*
tics'
Statis­ form,
origi­ index
calcu­
tics'
nal
num­ index lated
list,
ber,
num­
on
average average ber,
base,
prices prices average average
for
for
prices prices
18901890for
for
1899
18901899
= 100. = 100. 18901899
1899

-

7

+ 7
+ 3

- 6
±0

+ 9

-2 7
+45
- 7
+13
± 0

8
+ 3
-

+ 5
- 4
+ 4
± 0

+40

- 8
+8
±

0

+ 3
+ 7

1
+ 4

-

3

+41
- 9
+13
- 3

22

25

+18
-1 3
+ 1

+10

6

-

- 3
-1 9
+ 1
+10
+ 7

+6

+ 1
+21
-1 1

- 1
- 3

+6
+
8
+12
+ 3

±

-

0
6

+14

-11
+8

-1 4
- 7
-1 5
+ 4
+ 8
+ 1
+ 6
+ 6
+12
-1 0

+2
+ 1
+ 5
+ 5

+6
+8
+ 3

6

-

+12

-1 1

-2 2
+40
+8

-2 9
+40

+6

+20
-1 1

+17

4.1,

7.9

7.1

17

35

31

63

21892-1913.

The man who thinks that index numbers do well if they get within
10 per cent of the truth might be satisfied with this showing. But
the man who hopes for three significant digits1would be disappointed

if he had to accept these seven series as similar in meaning and equal
in authority. For the detailed differences among them are neither
few nor trifling. Indeed these differences are distinctly greater
than those found among the six index numbers made from the
bureau’s data and presented in Table 8 . For example, ( 1) the net
change in the price level between 1890 and 1913 is made twice as
great by two series as it is made by two others; (2) the maximum
i Or for two significant digits when the index number is less than 100.




THE M AKING AND USING OF INDEX NUMBERS.

99

difference between any two series for a given year averages over 11
points and varies irregularly between the wide limits of 3 and 19
points; (3) in a year of such decided business character as 1908 two
of the series show a rise of 6 to 8 points, while four indicate a fall of
Chart 12.—DUN’ S, BRADSTREET’ S, AND THE BUREAU OF LABOR STATISTICS’ IN D E X

NUMBERS REDUCED TO A COMMON BASIS. (BASED ON TABLE 18.)

7 to 12 points; (4) indeed the seven series all agree about the direc­
tion of price changes in only 12 cases out of 23; (5) regarding the
degree of these changes from one year to the next they show discrep­
ancies ranging all the way from 2 to 20 points and averaging nearly




100

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

10 points for the whole period; (6) the seven series also differ strik­
ingly in respect to steadiness, the least steady making the average
change in prices from one year to the next almost twice as great as
the steadiest series makes it; (7) certain of the series reflect changes
in business conditions with marked regularity, others are quite
unreliable business barometers, etc.
To show that these series differ in many details, however, means
little. The significant problem is whether these differences are due
to the inherent difficulty of measuring changes in the price level, to
the crudity of the general method of measurement in vogue, or to
technical differences in the construction of the particular index
numbers in question.
The way to attack this problem is shown by the preceding sections.
The seven series may be analyzed with respect to the ultimate sources
of information drawn upon, the adequacy of the original quotations
of each commodity, the numbers and kinds of commodities included,
the weights employed, the use made of relative prices, and the kinds
of average struck. At each step the question is whether the observed
differences among the index numbers accord with the differences
found to be characteristic of the various methods considered. If
most of the differences can be accounted for in this way, considerable
confidence may be felt in the possibility of measuring approximately
the variations in prices by index numbers.
The sources of information, the frequency of the quotations, and
the forms of average used, are in part so little known and in part so
similar that they give us no help in explaining the discrepancies
among the results.1 On the contrary, a marked influence can be
i Concerning the sources of information drawn upon by the compilers of Dun’s, Bradstreet’s, Gibson’s,
and the Annalist’s index numbers, the published information is slight. Despite the meager information,
however, there seems little reason to doubt that all these authorities use market prices at wholesale. The
Bureau of Labor Statistics alone states in full the source of every set of quotations, except those obtained
from private hous-es. The differences among the results therefore can not be ascribed to differences in
the nature of the ultimate data.
Dun’s Review uses quotations for Jan. 1 and July 1 in the years 1890 to 1900; from 1901 forward it
uses quotations on the 1st day of every month. Bradstreet’s uses quarterly quotations from 1892 to
1896, and monthly quotations thereafter. The Bureau of Labor Statistics in 1913 secured weekly quota­
tions for 44 commodities and monthly quotations for 208. In 1890-1901, however, it was obliged to content
itself with only one quotation a year for 16 commodities and with quarterly quotations for 4 more. (See
Bulletins No. 39, p. 215, and No. 149, p. 28.) The Gibson index number for 1890-1908 is made from quota­
tions collected by the Bureau of Labor Statistics; from January, 1909, onward it is based upon an independ­
ent collection of monthly figures. (See J. P. Norton, “ A revised index number for measuring the rise in
prices,” Quarterly Journal of Economics, Aug., 1910, vol. 24, p. 758, note.) As explained above, the index
number made from the list of 50 articles originally included by Gibson is compiled throughout from the
bureau’s data. Finally, the Annalist also takes its data from the Bureau of Labor Statistics for earlier
years (1890, 1896, 1910, 1911), but seemingly makes an independent compilation by months for 1912, and
by weeks for 1913 and 1914. (The Annalist figures for 1891-1895, 1897-1899, and 1901-1909 have been filled
in by the present writer from computations based upon the bureau’s relative prices.) All these authori­
ties, then, have more frequent and therefore more representative quotations for later years. For the earlier
years the Bureau of Labor Statistics’ collection seems rather the fullest on the whole. But it is not easy
to show just how the results are affected by differences in the frequency of quotations.
Also, the form of average used may be set aside as of no moment in explaining the discrepancies among
the results. Five of the series are arithmetic means of relative prices. The other two are relative figures
based upon sums of actual prices, and of course these sums bear the same ratios to each other as arithmetic
means made from them would bear, provided the weights were adjusted to that end. (See pp. 92 and 93.)




THE MAKING AND USING OF INDEX NUMBERS.

101

traced with confidence to differences in methods of weighting and in
the numbers and kinds of commodities included.
Dun’s index number is said to be weighted by per capita consump­
tion, and the weights for the separate commodities are so arranged
that foods count for 50 per cent of the total, textiles for 18 per cent,
minerals for 16 per cent, and other commodities for 16 per cent.1
Gibson’s index number in its present form is also said by the pub­
lisher to be weighted according to Dun’s method.2 Bradstreet’s series
has a curious combination of rational and irrational weights. The ra­
tional element consists in the inclusion of several quotations for impor­
tant articles like pig iron, coal, lumber, and hog products, and only one
quotation for articles like lemons, tea, and flax. The irrational ele­
ment results from the reduction of all the original quotations to prices
per pound. On April 1, 1897, these prices per pound ranged from
$0.0008 for soft coal and coke to $0.52 for quicksilver and $0.83 for
rubber. Recognition of the excessive influence upon the results
accorded to these high-priced articles presently led the computers
to drop them from the index number; but they seem to have retained
articles like alcohol and Australian wool which in 1897 cost $0.33
and $0.49 per pound—400 and 600 times as much as soft coal and
coke. Haphazard weighting preponderates also in the two series
from the Bureau of Labor Statistics, for the representation accorded
to different commodities has not been thoroughly worked out on any
logical plan. It is true that in the original figures certain highly
important articles are represented by two or more series—for instance,
coal, iron, cattle, and leather; but so also are certain articles of slight
moment, such as window glass, glassware, saws, sheetings, etc.3 In
the two remaining index numbers, the Annalist series and the original
form of Gibson’s index number, no formal weights are applied; but
the lists of commodities have been carefully studied and the most
important articles allotted two or three sets of quotations.
The constitution of the seven series with respect to the numbers
and kinds of commodities included can best be represented in tabular
form. The analysis, given in the next table, can not be applied to
Dun’s index number for lack of information about the commodities
and weights used, and it can not be strictly applied to Gibson’s pres­
ent series because we know the commodities but not the weights
allotted each. In the case of Bradstreet’s index number the percent­
ages of the total are computed on the basis of the prices per pound of
96 commodities published for April 1,1897. This basis is not wholly
i See J. P. Norton’s article in the Quarterly Journal of Economics, Aug., 1910, vol. 24, p. 754.
* It must be these weights that make Gibson’s figures as recomputed from the published index number
for Table 18 differ from the series made from Gibson’s present list and presented in Table 15. The latter
figures are unweighted arithmetic means of the relative prices prepared by the Bureau of Labor Statistics
for thft 22 articles which Gibson now includes.
* Compare pp. 61,62, and 72, note 3.




102

BULLETIN OF TH E BUREAU OF LABOR STATISTICS.

satisfactory, because the relative price per pound of different com­
modities, and therefore their relative influence upon the result, has
doubtless changed considerably from year to year. But the error
arising from using these figures for a single date is less than the error
that would arise if we merely counted the number of Bradstreet’s
commodities in the several classes.1 In dealing with the remaining
series counting the number of commodities in each class is satisfactory,
since there are no weights to be considered aside from the number of
forms or products by which each article is represented.
T able 19.—ANALYSIS OF THE COMMODITIES INCLUDED IN THE LEADING AMERICAN

IN D E X NUMBERS.

I. Division into raw, slightly manufactured, and manufactured products .

Index number.

1. Bureau of Labor Statistics,
original...........................
2. Bureau of Labor Statistics,
revised............................
3. Bradstreet’s.......................
4. Gibson, original.................
5. Annalist.............................
6. Gibson, present.................

Total
number
of
commodi­
ties.

Number of commodities
classified as—

Raw.

Percentage of the total.

Slightly
Manu-*
manu­
factured. factored.

Raw.

Slightly
Manu­
manu­
factured. factured.

242

49

25

168

20

10

70

145
96
50
25
22

36
40
26
8
11

21
22
4
5
2

88
34
20
12
9

25
2 36
52
32
50

14

61
255
40
48
41

29

8
20
9

2. Subdivision o f the manufactured and slightly manufactured goods.
Number of these commodities
classified as—

Index number.

1. Bureau of Labor Statistics,
original...........................
2. Bureau of Labor Statistics,
revised............................
3. Bradstreet’s.......................
4. Gibson, original.................
5. Annalist.............................
6. Gibson, present.................

Number
of com­
modities.

Con­
sumers’
goods.

Pro­
ducers’
goods.

Both
consum­
ers’ and
pro­
ducers’
goods.

Percentage of the total.

Con­
sumers’
goods.

193

108

45

109
56
24
17
11

51
21
11
17
11

35

226

22

68

Pro­
ducers’
goods.

Both
consum­
ers’ and
pro­
ducers’
goods.

30
32

226
24

8

2 12

2

50

1 Bradstreet’s now publishes quotations of 106 commodities, bases its index number on quotations of
96, and does not tell which 10 are omitted. Its prices per pound, published for only a short while in 1897,
include 98 articles, among them rubber and quicksilver, which are known to have been dropped from the
index number at a later date. Accordingly the quotations for the remaining 96 articles have been ac­
cepted as the basis of this analysis. Their prices per pound sum up to $5.9154, whereas Bradstreet’s revised
index number for this date is $6.0460—a difference of about 2 per cent.
2Percentage of the total weights on Apr. 1, 1897, not of the number of commodities included.




103

THE M AKING AND USING OF INDEX NUMBERS,

T able 19.— ANALYSIS OF THE COMMODITIES INCLUDED IN THE LEADING AMERICAN

IN D E X NUMBERS—Concluded.
3. Subdivision of the raw materials and slightly manufactured goods•

Index number.

1. Bureau of Labor Statistics,
original...........................
2. Bureau of Labor Statistics,
revised............................
3. Bradstreet’s.......................
4. Gibson, original.................
5. Annalist.............................
6. Gibson, present.................

Num­
ber of
com­
modi­
ties.

Number of these commodities
classified as—

Percentage of the total.

Mineral Farm Animal Forest Mineral
Farm Animal Forest
prod­ prod­ prod­
prod­ prod­ crops.
crops. prod­
ucts.
ucts.
ucts.
ucts.
ucts.
ucts.

74

18

15

12

29

7

6

5

12

57
62
30
13
13

18
24
10
6
8

10
15
8
7
5

10
6
3

19
17
9

12
114
20
24
36

7
125
16
28
23

7
il
6

13
15

18

1 Percentage of the total weights on Apr. 1, 1897, not of the number of commodities included.

What light do these facts about weights and the numbers and
kinds of commodities included shed upon the differences among the
seven index numbers ?
To begin with, the present Gibson and the Annalist index num­
bers are confined to one kind of commodities—foods, or rather
foods and the staples from which foods are prepared. The other
index numbers include besides foods an equal or greater number of
textile materials and fabrics, minerals, building materials, fuels, drugs,
etc. The constitution of the seven series in this respect is as follows: 1

Index number.

1.
2.
3.
4.
5.
6.
7.

Bureau of Labor Statistics, original...........................................................
B 11re9.11 of Labor Statistics, revised.................................................................
Bradstreet’s.........................
Gibson, original..................
Dun’s...................................
Gibson, present...................
Annalist..............................

Whole
number Number
of com­ of foods.
modities.
242
145
96
50
310?
22
25

58
40
37
21
?
22
25

Per cent
of foods.

24
28
129
42
1 50
100
100

* Weights allotted foods. Bradstreet’s weights as of Apr. 1, 1897.

Now it has been shown above that food index numbers differ widely
and capriciously from miscellaneous-list index numbers, because the
prices of agricultural products are largely dependent upon the yield
of each season’s harvests, while the prices of most other articles are
less dependent upon weather conditions than Upon the activity or
depression of business.2 Hence, if index numbers are sufficiently accu­
1 Foods are here taken in the rather liberal sense implied by the present Gibson and the Annalist index
numbers. Hence the number of foods credited to the Bureau of Labor Statistics is greater than the number
of articles which it so classifies in its own index number.
2 See pp. 68-70.




104

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

rate to charge their very differences with meaning, the seven series
under analysis should fall into three groups. (1) The two index
numbers composed exclusively of foods should resemble each other
rather closely and should differ rather widely from the three series
in which foods count for less than a third of the total. (2) These
three series, in turn, should resemble each other closely and differ,
not only from the food indexes pure and simple, but also, though in less
measure, from the two series in which foods count for approximately
half of the total. (3) The latter, Dun’s index number and the index
number made from Gibson’s original list, should be hybrids, stand­
ing intermediate between the two pure stocks, Dun’s inclining rather
toward the food index numbers and Gibson’s toward the misceUaneous-list group.
These expectations are put to the test in the next table and hand­
somely realized. The best simple criterion of relationships among
the index numbers is the average number of points by which their
results differ for each of the 24 years for which data are available.
On this basis it appears that the two forms of the Bureau of Labor Sta­
tistics’ series and Bradstreet’s index number come very close together—
the greatest average difference is only 2 points. On the other hand,
the two food index numbers agree much better with each other than
they agree with any of the other series—though the average differ­
ence between them is 3.9 points—distinctly larger than the differ­
ences among the miscellaneous-list series. Presumably, this greater
difference arises from the relatively small number of articles included
by both the Annalist and Gibson’s present list, 25 and 22, respectively.
Finally, it also turns out not only that Dim’s index number and the
series made from Gibson’s original list stand between the two extreme
groups, but also that of the two the Gibson series bears a distinctly
greater resemblance to the miscellaneous-list group and Dun’s index
number a rather closer resemblance to the food group.1
i The influence of the food factor can be traced in the detailed differences among the series as well as in
the average differences. For instance, it is the peculiar harvest conditions of 1891,1893, 1901, and 1908
that force the food index numbers up when the miscellaneous-list series fall because of business depres­
sion; and it is harvest conditions of an opposite sort that check the rise of the food index numbers in 1905
and 1910 when the miscellaneous-list series respond buoyantly to the increasing activity of trade. In all
of these cases Dun’s index number, and in less measure that made from Gibson’s original list, move in par­
tial sympathy with the food series. Again, the food index numbers change more from one year to the next
than the other series, because raw materials in general and farm crops and animal products in particular
are more variable in their prices than manufactured goods and raw mineral and forest products. In addi­
tion, their high percentages of raw materials and especially of agricultural products account also for the
relatively high levels upon which the food index numbers fluctuate in the later years covered by the tables;
for it has been found that these classes of commodities have risen more in price since 1890-1899 than those
with which they are contrasted. (See Tables 9 and 10.)




THE MAKING AND USING OF INDEX NUMBERS.

105

TABLE 20.—DEGREES OF KINSHIP AMONG THE SEVEN AMERICAN IN D E X NUMBERS
OF TABLE 18 AS SHOWN B Y THE AVERAGE NUMBER OF POINTS B Y WHICH T H E Y
D IFFE R IN THE YEARS 1890 TO 1913.
I. Average differences between the original form o f the Bureau o f Labor Statistics index
number and—
Points.
Bureau of Labor Statis­
tics, revised....................
Bradstreet’s ......................

Points.

Points.
Gibson, original..............

1 .0

2.5
5.5

6 .6

7.2

Gibson, present form___

1.9

2 . Average differences between the revised form o f the Bureau o f Labor Statistics index
number and—

Bureau of Labor Statis­
tics, original...................
Bradstreet’s......................

Points.

Points.

Points.
Gibson, original..............
1 .0
2 .0

6.3

2 .0

5.3

Gibson, present form___

6 .8

3. Average differences between Bradstreet's index number and—
Points.

Points.
Bureau of Labor Statis­
tics, original...................
Bureau of Labor Statis­
tics, revised....................

Gibson, original..............
1.9

3.5
6 .6

Points.
Annalist..........................
Gibson, present form___

6.7
7.0

2 .0

4. Average differences between the index number made from Gibson*s original list and—
Points.

Points.
Bureau of Labor Statis­
tics, original.................
Bureau of Labor Statis­
tics, revised....................
Bradstreet’s......................

2.5

Dun’s..............................

4.1

Points.
Annalist..........................
Gibson, present form___

5.5
5.9

2 .0

3.5

5 . Average differences between Dun*s index number and—
Points.

Points.
Bureau of Labor Statis­
tics, original...................
Bureau of Labor Statis­
tics, revised....................
Bradstreet’s ......................

Gibson, original..............

4.1

5.5

Points.
6 .1

Gibson, present form.. 1.

4.5

5.3
6 .6

6. Average differences between the Annalist index number and—
Points.

Points.
Bureau of Labor Statis­
tics, original...................
Bureau of Labor Statis­
tics, revised....................
Bradstreet’s ......................

6 .1
6 .6

Gibson, original..............

Points.
Gibson, present form___

3.9

5.5

6.3
6.7

7. Average differences between the present form o f Gibson*s index number and—
Points.
Bureau of Labor Statis­
tics, original...................
Bureau of Labor Statis­
tics, revised........ ...........
Bradstreet’s......................




7.2
6 .8

7.0

Dun’s..............................
Gibson, original..............

Points.

Points.

4.5
5.9

3.9

106

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

Gibson’s present series, then, and the Annalist index number may
be set aside as different in kind from the miscellaneous-list series.
They do not aim to measure the same thing as the latter, and there­
fore the wide and frequent discrepancies between the two groups are
not disquieting. Quite the contrary, the series differ from the
miscellaneous-list series in precisely the ways that the previous
sections would lead one to expect. This fact is highly reassuring;
for it means that in different parts of the business field there really
are general trends among the apparently random variations of
prices, and that existing index numbers have measured these divergent
trends with approximate accuracy. Otherwise such close consistency
would hardly exist among the results.
It is equally reassuring to find that most of the small discrepancies
among the three miscellaneous-list series are also consistent with
what has already been learned about the price fluctuations of
different kinds of commodities. Indeed it is curious that two such
dissimilar kinds of weighting as are used in Bradstreet’s index
and in the two series drawn from the Bureau of Labor Statistics
should not have produced wide discrepancies. These three series
never contradict one another flatly about the direction in which
prices are moving. The nearest approach to disagreement occurs in
the five years (1893, 1897, 1903, 1904, and 1913) when one or two
fail to change while another moves up or down a trifle. In no year
are the two bureau series more than 4 points apart, and their average
difference is only 1 point. Similarly, Bradstreet’s is never more
than 7 points out with the original bureau index, and never more than
6 points out with the revised series. Its average differences from
them are 1.9 and 2 points, respectively. Bradstreet’s is sometimes
above and sometimes below the two bureau series, so that its average
differences from them computed from algebraic sums of the plus and
minus quantities are only five-tenths and nine-tenths of 1 pbint, re­
spectively. The corresponding average difference between the two
bureau series is four-tenths of 1 point.1
1 It is interesting to compare these differences with those which separate the index numbers worked
out above for different parts of the system of prices.

Difference.
Index number.
Bureau of Labor Statistics, original, and Bureau of Labor Statistics,
revised..........................................................................................................
Bureau of Labor Statistics, original, and Bradstreet’s.................................
Bureau of Labor Statistics, revised, and Bradstreet’s ..................................
49 raw materials and 183 to 193 manufactured articles................................
20 raw materials and 20 of their products..........
products.
5 raw materials and 5 groups of their products.
Mineral and farm products.................................
Mineral and animal products.
Mineral and forest products............
Farm and animal products............
Farm and forest products...............
Animal and forest products............
Producers’ and consumers’ goods .

Average.

Maxi­
mum.

1.0
1.9

2.0
5.9
9.1
14.0
1 0 .1

9.0
18.6
8.9
19.6
15.8
6.7

18
21

28
31
32
61
20

47
41
19

N ote.—For the figures from which these differences are computed see Tables 18,9,10, and 11.




Mini­
mum.

TH E M AKING AND USING OF INDEX NUMBERS,

107

The discrepancies that do occur arise chiefly from the fact that
while a given change in business conditions affects all three series in
the same way it usually causes a wider fluctuation in Bradstreet’s
index than in the revised bureau series, and a wider fluctuation in
the latter than in the bureau’s original series. This difference in
steadiness is just what should follow from the constitution of these
three index numbers with reference to their proportions of raw
materials and manufactured products. To the reader who remembers
that raw materials fluctuate much more widely in price than goods
manufactured from them, the following schedule tells its own story:

Index number.

Bureau of Labor Statistics, original..........................................................................
Bureau of Labor Statistics, revised......................- ...................................................
Bradstreet’s.................................................................................................................

Average
change
from year
to year.

Percentage
of raw
materials.

Points.
4.0
4.1
5.6

25
36

20

The only thing that is difficult to explain, indeed, is the general
level on which the three index numbers fluctuate in 1900-1913. We
should expect Bradstreet’s to stand a little higher than the two bureau
indexes because of its larger proportion of raw materials and smaller
proportion of minerals. In fact it stands a shade lower, and the slight
weight it assigns to the rapidly rising prices of forest products seems
hardly sufficient to account for this result, since these products count
for only 5 and 7 per cent of the totals in the two bureau series.
The preceding comparison of index numbers on the 1890-1899
basis may be supplemented by a similar comparison on the chain
basis, that is, prices in the preceding year equal 100. Table 21 sup­
plies chain index numbers for this purpose. Only three of the seven
series can be included; for among the relative-price index numbers
yearly percentages of rise or fall in price have been computed and
averaged only for the official list of the Bureau of Labor Statistics.
Of course chain figures can easily be made from the aggregate actual
prices given by Bradstreet’s and Dun.
In part the results merely show in a slightly different form the
differences among the three series brought out by Table 18 and
explained in the preceding commentary. It may be noted, however,
that the use of a finer scale, including one place in the decimal col­
umn, makes the agreement between the bureau’s series and Bradstreet’s even closer than it has hitherto appeared. In Table 21 these
two series invariably agree about the direction in which prices
changed from one year to the next, though Bradstreet’s index num­
ber, maintaining its greater sensitiveness, makes the degree of change
rather larger on the average. On the other hand, Dun’s series contra­
dicts the other two about the direction of change in 7 years out of 23.



108

BULLETIN OF THE BUREAU OF LABOR STATISTICS.
Table 21.—CHAIN INDEX NUMBERS.
(Percentage of rise ( + ) or fall ( —) from prices in the preceding year.)
Chain index numbers made
from data supplied by—
Year.

Bureau
of Labor
Statis­
tics.

1891
1892
1893
1894
1895
1896
1897
1898,
1899
1900
1901
1902

-

Year.
Bradstreet’s.

0.2

- 4.4
-

.2

- 3.1

- 8.7
- 1.5

-

2.8

-

-

+ .2
+ 4,
+ 10.
+ 9.
- 1.
+ 4.

Chain index numbers made
from data supplied by—

11.2

- 3.7
+
+
+
+
+

8.1

3.4
7.4
9.7
9.3
3.9
4.0

Bureau
of Labor
Statis­
tics.

Dun.

+
+
-

6.9
7.4
2.7
8.4
4.0
6.4
2.7
+ 6.6
+ 5.0
+12.7
+ 2.7
+ 4.7

1903
1904
1905.
1906.
1907.
1908.
1909.
1910.
1911
1912.
1913.

+
+
+
+
+
+
+
+

1 .2
.1

2.9
5.8
6 .0

5.6
3.2
4.1
1.9
3.4
1 .2

Bradstreet’s.

+ .8
- .2
+ 2.3
+ 3.9
+ 5.8
- 10.1
+ 6.3
+ 5.5
- 3.1
+ 5.4
+ .2

Dun.

+
+
+
+
+
+
+
-

1.4
1 .1

.4
4.7
6 .1
1 .8

7.2
1 .2
2 .0

6.5
2 .8

The new point of chief interest in this table is that the chain index
numbers differ less than the corresponding series computed on a fixed
base. A comparison covering the years 1892-1913 works out as
follows:
Difference.
Average.

Maximum.

Minimum.

Index number.
Fixedbase
series.
Bureau of Labor Statistics and Bradstreet’s.......................................................
Bureau of Labor Statistics and Dun’s .......
Bradstreet’s and Dun’s...............................

Chain
index
number.

1.9
5.4

2.5

6 .6

2 .8

1 .8

Fixedbase
series.

7
18
18

Chain
index
number.

Fixedbase
series.

Chain
index
number.

5
5
8

For the closer agreement among the chain-index form of these
figures there are two reasons, one of them arising from the trend of
price fluctuations in the particular period covered, and the other of
more general significance. In a majority of the years 1890 to 1913
the price level was higher than its average in 1890-1899. Hence the
majority of the year-to-year changes are percentages of a larger
number than that upon which the fixed-base relatives are computed.
Therefore the percentages are themselves smaller numbers. And of
course the differences between smaller numbers are themselves
smaller, other things being equal. The second reason has been
brought out before. Variations of prices over an interval of only
one year are more highly concentrated about a central tendency than
variations over an interval of several years.1 Hence averages made




* See pp. 21-23.

TH E M AKING AND USING OF INDEX NUMBERS.

109

from the yearly variations are less likely to be distorted by differences
in the samples used than averages made from variations computed
on a fixed base.
2. CRITICAL VALUATION.

A just evaluation of our seven American index numbers is not easy
to make. For a comparison has little meaning unless it deals with
all the important points at which the series differ. And since no one
series is superior to the others at all points a verdict can not be
rendered in a single sentence.
In the publication of actual prices, the Bureau of Labor Statistics
and Bradstreet’s stand foremost. The contribution they have thus
made to the knowledge of prices possesses great and permanent value
over and above the value attaching to their index numbers. For, it
is well to repeat, all efforts to improve index numbers, all investiga­
tions into the causes and consequences of price fluctuations, and all
possibility of making our pecuniary institutions better instruments
of public welfare depend for their realization in large measure upon
the possession of systematic and long-sustained records of actual
prices. And much of this invaluable material would be lost if it were
not recorded month by month and year by year.
Critical users of statistics justly feel greater confidence in figures
which they can test than in figures which they must accept upon
faith. Hence the compilers of index numbers who do not publish
their original quotations inevitably compromise somewhat the repu­
tation of their series. They compromise this reputation still further
when they fail to explain in full just what commodities they include,
and just what methods of compilation they adopt.1 In the latter
respect the Annalist index number shares first honors with the Bureau
of Labor Statistics’ series. Anyone who chooses to take the trouble can
find what commodities are used, and how the final results are worked
up from the raw material. Bradstreet’s index number suffers a bit in
comparison because readers are not told which 96 commodities out of
the 106 of which prices are published are included in the index number,
and because the method of reducing prices by the yard, the dozen,
the bushel, the gallon, etc., to prices per pound is not fully explained.
Dun’s index number is more mysterious still, because neither the list
of commodities nor the weights applied to each commodity are dis­
closed. And Gibson’s present series also stands partly in the shadow
because, while the list of commodities is known, the publishers state
merely that these articles are weighted by Dun’s system.
With reference to weighting, Bradstreet’s index number takes low
rank, for the plan of reducing all quotations to prices per pound grossly
misrepresents the relative importance of many articles. That figures




1 Compare

p. 29.

110

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

made thus should give results in close agreement with the Bureau of
Labor Statistics’ series is a remarkable demonstration of the ability
of index numbers to extract substantial truth even from unprom­
ising materials. The agreement is all the more remarkable since the
bureau’s series is also badly weighted, though in a different way and
in less degree.1 The revised bureau series is scarcely better than the
original in this respect. It is better in substituting a single set of
relatives for the articles of minor importance to which the original
accorded several sets (for example, shirtings, sheetings, tools, win­
dow glass, etc.), but worse in cutting down the representation ac­
corded to great staples (for example, pork, coal, pig iron, and leather).2
The Annalist index number follows the sensible, though rudimentary,
plan of including two or three varieties of the most important articles,
and only one of the less important. The like can be said in favor of
Gibson’s indox number, both in its original and its present form, and
in addition Gibson uses the Dun system of weights. The latter
system is, in theory, the nearest approach to a satisfactory plan of
weighting made by any American index number at present. Whether
the practice is as good as the theory is doubtful, to say the least, for
anyone familiar with the deficiencies of American statistics of con­
sumption must wonder whence the compilers derived their estimates
of the quantities of 310 commodities “ annually consumed by each in­
habitant.” Moreover, what little is known concerning the actual
weights is not unobjectionable. Fifty per cent of the total is too large
a weight to allow to foods in a wholesale-price series. Even in the
great collection of budgets of workingmen’s families made by the Com­
missioner of Labor in 1901 the average expenditure for food was less
than 45 per cent of total family expenditure;3 and in wholesale mar­
kets, of course, many commodities that are never directly consumed
by families have great importance.
Dun’s index number is supposed to stand first in number of com­
modities included, but lack of definite information makes it impossi­
ble to judge whether its list is well balanced, The bureau’s list also
is long and contains samples of many different kinds of goods, manu­
factured as well as raw, consumed for all sorts of purposes and produced
under all sorts of conditions; but the representation accorded to
different parts of the whole system of prices is certainly far from
equitable. Bradstreet’s list, while less than half as long as the
bureau’s, seems better chosen. It is particularly strong in raw ma­
terials and rather weak in manufactured goods. The same remarks
apply to Gibson’s original list, though it suffers in comparison by
1Compare p. 72, note.
2 See the list of commodities used in this index number, p. 47, footnote.
* Eighteenth Annual Report of the Commissioner of Labor, 1903, p. 66. The data represented 25,440
families and 124,108 persons, both natives and immigrants.




I ll

TH E M AKING AND USING OF INDEX NUMBERS.

being only about half the length of Bradstreet’s. Finally, the pres­
ent Gibson index number and the Annalist series are confined to
foodstuffs, and make no pretense of representing prices at large.
In the form of presenting results, Bradstreet’s set an admirable
example, which was wisely followed by Dun’s. Their sums of actual
prices can readily be turned into relatives on any base desired, and
hence can be made to yield direct comparisons between any two
dates. The other series, as averages of relative prices on the 18901899 basis, can not be properly shifted without a detailed recompu­
tation of the relative prices of each commodity, and force readers to
make all their comparisons in terms of what prices were in the decade
used as base.
It is interesting, finally, to test the reliability of the several index
numbers as “ business barometers.” Monthly figures would be much
better than our yearly averages for this purpose; but, since they are
not to be had for most of the series during most of the period cov­
ered, we must do the best we can with the rougher gauge. In 11 of
the 23 cases of changes from one year to the next the seven index
numbers disagree as to whether prices rose, fell, or remained constant.
In the following schedule these 11 years are represented by columns
in which each index number is credited with plus one when its change
accords with the character of the alteration in business conditions,
debited with minus one in cases of disagreement, and marked zero
when it recognizes no change in the price level.1 The net scores
made by casting up the plus and minus entries indicate roughly the
relative faithfulness with which these series have reflected changes
in business conditions in the past. Of the index numbers regularly
published, Bradstreet’s makes much the best showing. Even the
scores against it in 1895 and 1903, and its failure to show the reaction
in business conditions in 1913, would be wiped out were the data by
quarters and months used in place of the annual averages.
Index number.

1891

1. Bradstreet’s .................... 2+1
2. Bureau of Labor Statis­
tics, revised...................
+1
3. Gibson, original..............
0
4. Bureau of Labor Statis­
tics, original................... +1
-1
5. Annalist..........................
6. Dun’s ..............................
-1
-1
7. Gibson, present..............

1893

1895

1897

1901

1903

1904

1905

1908

1910

Net
1913 score.

+1

-1

+1

+1

-1

+1

+1

+1

+1

0

+6

+1
0

-1
0

0
+1

+1
+1

0
+1

0
-1

+1
+1

+1
+1

+1
+1

+1
0

+6
+5

0
-1
-1
-1

-1
-1
-1
-1

0
+1
-1
+1

+1
-1
-1
-1

-1
+1
+1
+1

+1
-1
-1
+1

+1
+1
0
-1

+1
-1
+1
-1

+1
+1
+1
0

0
+1
+1
+1

+4
-1
-2
—2

1 For a description of American business conditions in this period, see W . C. Mitchell, Business Cycles,
Chapter III (Summary, p. 88).
2 Based on Bradstreet’s original figures for 1890 and 1891, figures which are not used in the index number
as currently published.




112

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

Each of these seven series, then, has its special uses, its merits, and
its defects. Choice among them should be made in accordance with the
particular purpose for which an index number happens to be wanted.
But it seems feasible to construct an American series which would
present a stronger combination of good qualities as a general-purpose
index number than any now existing. The original quotations might
be collected from the records of the Bureau of Labor Statistics and
Bradstreet’s, a list of commodities more complete than Bradstreet’s
and better balanced than the bureau’s might be drawn up, the use of
actual prices might be adopted from Bradstreet’s and Dim’s, the
several commodities might be weighted by physical quantities after
Dun’s fashion, but with the use of a criterion more appropriate to
wholesale prices, and the whole process of construction might be set
forth with the frankness characteristic of the Annalist and the
bureau. Such a series might differ little from the figures now avail­
able; but, however it might turn out, its results would merit greater
confidence than can properly be felt in any of the present index
numbers as a measure of changes in the general level of wholesale
prices.
VI.— CONCLUSIONS.

1. Variations in the level of wholesale prices from one year to the
next are capable of being measured with a close approximation to
accuracy, for these variations are highly concentrated about a
central tendency. There are two American chain index numbers
which never differ by more than 5.3 per cent, and differ on the average
by less than 2 per cent, although they were compiled from start to
finish quite independently of each other, based upon dissimilar sets
of price variations, constructed by unlike methods, and extended
over 22 years of violent fluctuations. Moreover, these moderate
differences are not inscrutable results of dependence upon chance for
the samples used for analysis, but for the most part arise from known
causes, and harmonize with the outcome of investigations into the
dissimilarities of variation characteristic of different parts of the
system of prices.
2. Variations in prices that have been cumulating] through several
or many years show much less concentration about a central tend­
ency than variations from one year to the next. Hence, index num­
bers on a fixed base become less trustworthy the greater the time
elapsing between the base period and the year under consideration.
Hence, also, most of the entries in a fixed-base series are less trust­
worthy than chain index numbers with a one-year interval made from
the same data. Nevertheless, the discrepancies observed between
the two series just referred to (Bradstreet’s and the Bureau of Labor
Statistics’ index numbers) never exceed 7 points in the scale of relative




TH E M AKING AND USING OF INDEX NUMBERS.

113

prices, and average less than 2 points, even when compared in the
fixed-base form. And, to repeat, the discrepancies themselves are
of the character which an investigator would predict, if he were
familiar both with the data used in these two series and with the
fluctuations characteristic of various groups of commodities.
3. The choice of methods to be employed in making an index
number should be guided by the purpose for which the results are
to be used. These purposes are so numerous and so diverse that it is
impossible to make a single series well adapted to them all. Probably
the time is near when certain uses will be so standardized that,
several divergent types of index numbers will be regularly compiled
to serve the needs of various groups of users. Even now we have
special index numbers of the prices of foods, of farm products, of
metals, etc. To this list there might well be added a series especially
designed to throw changes in business conditions into high relief, and
assist in the bettering of business forecasts. Most of the currently
published index numbers, however, are what may be called generalpurpose series, which undertake to measure changes in the whole­
sale price level at large.
4. The best form for these general-purpose series is a weighted ag­
gregate of actual prices
5. The more commodities that can be included in such an index
number the better, provided that the system of weighting is sound.
Certainly, each of the following classes of commodities should be
represented, and represented as fully as is feasible: Raw mineral,
forest, animal, and farm products, and manufactured products in
various stages of elaboration, bought for family consumption and for
business use.
6. Probably the best weights to apply are the average physical
quantities of the commodities bought and sold over a period of years
without reference to the number of times their ownership is changed.
These weights should be applied directly to the actual prices of each
commodity in making up the totals for the several groups that have
been mentioned, and then, if the necessary data can be secured, the
totals for the several groups should be weighted again in making up
the grand totals for “ all commodities.”
7. In presenting such an index number, it is well to publish the
aggregate actual prices, both for the several groups and for the grand
totals. But it is highly desirable to publish also relatives made from
these actual prices on a percentage scale, since comparisons can be
made more easily from such figures than from the aggregates of actual
prices, which are likely to run in awkward quantities. Indeed, several
sets of these relatives, computed on the basis of actual prices at
different times, can readily be provided for readers interested in

94261°—Bull. 173—15------8




114

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

knowing how prices have changed with reference to recent or to past
years. Among the relatives of greatest significance is the set which
shows the annual percentage of rise or fall as compared with prices in
the preceding year. In such chain index numbers it is usually possible
to include some commodities for which quotations are lacking in cer­
tain of the years covered by the whole investigation.
8.
While index numbers are a most convenient concentrated ex­
tract of price variations, they are far from being a competent represen­
tation of all the facts which they summarize. Most “ consumers of
.statistics” lack the time to go back of the finished products to the
data from which they are made. But the increase of knowledge con­
cerning the causes and consequences of price variations depends much
more upon intensive study of the ultimate data than upon the manipu­
lation of averages or aggregates. Upon the extension of knowledge
in this field depend in turn large issues of public welfare. Hence it is
highly important to collect and to publish in full the actual prices of
as many commodities as possible, even though some of the quotations
may not now be available for use in making an index number.




PART II.—INDEX NUMBERS OF WHOLESALE PRICES
IN THE UNITED STATES AND FOREIGN COUN­
TRIES.
UNITED STATES.
INDEX NUMBERS OF THE UNITED STATES BUREAU OF LABOR STA­
TISTICS.
PUBLICATION.

An index number is published in connection with the reports on
wholesale prices issued by the Bureau of Labor Statistics of the United
States Department of Labor at Washington. These reports are
issued in bulletin form and appear annually.
HISTORY.

The publication of this index number was begun in 1902. Prior to
that time the Department of Labor, now the Bureau of Labor Sta­
tistics, had conducted an inquiry into the course of wholesale prices
from 1890 to 1899, the results of which were published in March, 1900.1
The purpose of this inquiry was to continue, so far as practicable, the
investigation made for the Senate Committee on Finance for the years
1840 to 1891 under the direction of Roland P. Falkner, statistician to
the committee.2 In the report of the Department of Labor alluded
to, the index numbers appearing in the Senate Finance Committee’s
report were brought down to 1899, important changes with respect to
the base period and the method of weighting being adopted. In 1902,
however, when the material for the new report on wholesale prices was
being assembled, it was found that many articles included in the report
of the Senate Committee on Finance were either no longer manu­
factured or had ceased to be important factors in the market. On
the other hand, a number of articles not shown in that report had be­
come of such importance as to render necessary their inclusion in the
new report. These facts necessitated the computation of a new series
of index numbers based on the revised list of commodities. It
was found, however, that prices of such commodities could be
obtained for a period dating back to 1890, so that the new series of
index numbers, as published in the 1902 report,3 covered the 12 years
from 1890 to 1901, inclusive. This series has been continued in sub­
sequent wholesale-price reports.
i Bulletin of the Department of Labor, No. 27.
* Report from the Committee on Finance of the United States Senate on Wholesale Prices, Wages,
and Transportation. Mar. 3,1893. 52d Congress, 2d session, Report No. 1394.
•Bulletin of the Department of Labor, No. 89, March, 1902.




116

BULLETIN OF THE BUREAU OF LABOR STATISTICS,
SOURCE OP QUOTATIONS.

The commodities included in the reports have been selected, not only
with regard to their representative character, but also with regard to their
availability in the future in the continuation of the price record. Stand­
ard trade journals, reports of boards of trade, chambers of commerce,
produce exchanges, and leading manufacturers or their selling agents
are the usual sources from which the price quotations are obtained.
It has been the aim to secure the quotations for the various com­
modities from their primary markets. At present about one-half of
the quoted prices are those in the New York market. For grains, live
stock, etc., Chicago prices are quoted; for fish, New York and Boston
prices; for pig iron, Pittsburgh prices; for tar, Wilmington, N. C.,
prices; etc. The prices for textiles are those prevailing in the general
distributing markets, such as New York, Boston, and Philadelphia;
and where no market is mentioned it should be understood that the
prices are for the general market.1
BASE PERIOD.

In the compilation of the bureau’s index numbers it was recognized
that in reducing a series of actual prices to relative prices a base must
first be chosen that represents, approximately at least, prices when
business conditions are normal. This may be either a single quota­
tion, the average price for one year, or the average for two or more
years. If the price for a single year is chosen, it is essential that the
year be a normal one, for if prices are high in the year chosen for the
base any subsequent fall will be unduly magnified, while on the other
hand, if prices are low any subsequent rise will be unduly magnified.
For the reason that all commodities probably never present a normal
condition as regards prices in any one year, it was decided that an
average price for a number of years would better reflect average con­
ditions and form a broader and more satisfactory base than would the
price for any single year. The period chosen as this base was that
from 1890 to 1899—a period of 10 years. In the cases of a few articles
for which prices for the entire 10-year period could not be obtained,
the average for such years prior to 1899 as were available was chosen
as the base.
The relative prices included in the series have been calculated in
the usual manner and represent the percentage which each monthly
or yearly price is of the average price for the base period 1890-1899.
The average price of every article for the base period is represented
by 100, and the relative prices for each month or year show the
percentage of rise and fall, from month to month or from year to
year, of the prices of each single commodity, of each group of com­
modities, and of all commodities in terms of the average prices in
1890-1899.
i Bulletin of the United States Bureau of Labor Statistics, No 149, p. 27.




INDEX NUMBERS— UNITED STATES.

117

PRICES: HOW SHOWN AND COMPUTED.

Weekly prices have been secured in the case of all articles which are
subject to frequent fluctuations in price, such as butter, cheese, eggs,
grain, live stock, meats, etc. In the case of articles whose prices are
more stable, monthly prices have been taken. The majority of the
weekly quotations show the price on Tuesday, but if for any reason
the price was not obtainable on the particular day stated, the first
available price thereafter has been taken. The quotations from
trade and other journals, when credited to the first of each month,
are not in all instances the price for the exact day stated. The prices
are, however, the earliest prices quoted for the month in the journal
to which the article is credited.
In many localities the price of bread per loaf is not affected by
changes in the price of flour, but the weight of the loaf is changed
from time to time. For this reason the relative prices of bread are
computed on the price per pound and not per loaf.
The average price for the year is obtained by dividing the sum of
the quotations for a* given commodity by the number of quotations
shown. For example, the sum of the 52 Tuesday prices of cotton
for 1913 was $6.6520. This total divided by 52 gives $0.12792 as
the average price for the year. When a range is shown the mean
price for each date is found, and this is used in computing the yearly
average as above described. It should be understood that, in order
to secure for any commodity a strictly scientific average price for the
year, one must know the quantity marketed and the price for which
each unit of quantity was sold. It is manifestly impossible to secure
such detail, and even if it were possible the labor and cost involved
in such a compilation would be prohibitive. It is believed that the
method adopted here, which is also that used in the construction of
other index numbers, secures results which are quite satisfactory for
all practical purposes.1
Net cash prices are shown for textiles and all articles whose list
prices jare subject to large and varying discounts. In the case of a
few articles, however, the prices of which are subject to a small dis­
count for cash, no deduction has been made. All rebates have been
deducted.
NUMBER AND CLASS OF COMMODITIES.

In the record of prices from 1890 to 1913, 234 series of quotations
have been presented for the entire period and an additional 38 for
some- portion of the period. The number of commodities included
in the report for 1913 was 252, classified as follows: Farm products,
20 articles; food, etc., 54 articles; cloths and clothing, 63 articles;
fuel and lighting, 13 articles; metals and implements, 38 articles;
lumber and building materials, 28 articles; drugs and chemicals, 9
* Bulletin of the United States Bureau of Labor Statistics, No. 149, p. 29.




118

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

articles; house-furnishing goods, 14 articles; and miscellaneous, 13 arti­
cles. It was recognized by the bureau that, in thfe computation of an
index number of this character, it is important thap the greatest care be
exercised in the choice of commodities, in order mat a simple average
of their relative prices shall show a general price level, and it has
been the aim to select only important and representative articles in
each group. The use of a large number of articles, carefully selected,
minimizes the effect on the general price level of an unusual change
in the price of any one article or of a few articles.
DESCRIPTION AND GROUPING OF COMMODITIES.

The following list, compiled from the 1913 report (pp. 33 to 82),
shows the grouping and description of the articles:

Farm 'products (20 artkles).
Barley: Choice to fancy malting, by samples.
Oattle:
Steers, choice to prime, corn fed.
Steers, good to choice, com fed.
Corn: Contract grades, cash.
Cotton: Upland, middling.
Flaxseed: No. 1, cash.
Hay: Timothy, No. 1.
Hides: Green, salted, packers’ , heavy native steers.
Hogs:
Heavy (range of quotations for selected 260 to 300 pounds packing and fair to
fancy heavy shipping).
Light (range of quotations for common to choice, light bacon, and fair to fancy
selected butcher’s).
Hops: New York State, prime to choice.
Horses: Draft, choice to extra.
Mules: Medium to extra (16 hands from January to second week in February, 16
to 16J hands from third week in February to December).
Oats: Contract grades, cash.
Poultry: Live fowls (by freight), choice.
Rye: No. 2, cash.
Sheep:
Native wethers, poor to prime.
Western wethers, plain to prime.
Tobacco: Burley, dark red, good leaf..
Wheat:
No. 1, northern spring, cash.1 .
_A
x
No 2 red winter ca5i
prices combined to form 1 series of quotations.

Foods, etc. (54 articles).

Beans: Medium, choice.
Bread:
Crackers, oyster, puff, in boxes.
Crackers, soda, in boxes, containing 3 dozen each.
Loaf (weight before baking, 16 ounces).
Loaf (weight before baking, 15 to 15J ounces).




INDEX NUMBERS— UNITED STATES,

119

Butter:
Creamery, Elgin.
Creamery, extra.
Dairy, New York State, tubs, finest.
Canned goods:
Com, Republic No. 2, fancy.
Peas, Kepublic No. 2, sifted.
Tomatoes, Arlington, standard, New Jersey, No. 3.
Cheese: New York State, full cream, large, colored, fancy.
Coffee: Rio No. 7, Brazil grades.
Eggs: New laid, State, Pennsylvania and near-by, range of hennery and fresh-gathered
white.
Fish:
Cod, whole fish, dry, bank, large.
Herring, pickled, Newfoundland split, large No. 1.
Mackerel, salt, large No. 3.
Salmon, canned, Columbia River, 1-pound tails (Chinook fancy).
Flour:
Buckwheat.
Rye (range, in jute and in wood).
Wheat, spring patents (range, in sacks and barrels).
Wheat, winter straights (range, in sacks and barrels).
Fruit:
Apples, evaporated, choice.
Currants, uncleaned, in barrels.
Prunes, California, 60s to 70s, in 50-pound boxes.
Raisins, California, London layer.
Glucose: 42 degrees mixing.
Lard: Prime contract.
Meal:
Com, fine white.
Com, fine yellow.
Meat:
Bacon, rough sides (short rib), smoked, loose.
Bacon, short clear sides, smoked, loose.
Beef, fresh, carcass, good native sides.
Beef, fresh, native sides.
Beef, salt, extra, mess.
Hams, smoked, loose.
Mutton, dressed.
Pork, salt, mess, old to new.
Milk: Fresh.
Molasses: New Orleans, open kettle.
Poultry: Fresh-killed dressed fowls, western, dry packed, dry picked, fancy, 48 to
55 pounds to the dozen.
Rice: Domestic, choice, head.
Salt: American, medium. .
Soda: Bicarbonate of, American.
Spices: Pepper, black, Singapore.
Starch: Com, for culinary purposes, Sunbeam, 48 1-pound packages in box.
Sugar:
96 degrees centrifugal.
Granulated, in barrels.
Tallow.




120

BULLETIN OF THE BUBEAU OP LABOB STATISTICS.

Tea: Formosa, fine.
Vegetables, fresh:
Cabbage, white (range of domestic-grown cabbage from native and Danish seed).
Onions.
Potatoes, white, ordinary to fancy.
Vinegar: Cider, Monarch, 40-grain, in 45-gallon barrels.
Cloths and clothing {68 articles).

Bags: 2-bushel, Amoskeag.
Blankets:
All wool, 11-4, 5 pounds to the pair.
Cotton, 2 pounds to the pair, 54 by 74 inches.
Boots and shoes:
Men’s seamless Creedmores (split ties), first quality standard screw or pegged.
Men’s vici calf shoes, blucher bal., vici calf top, single sole.
Men’s vici kid shoes, Goodyear welt.
Women’s solid grain shoes, leather, polish or polka.
Broadcloths: First quality black, 54-inch, made from X X X wool.
Calico: American standard prints, 64 by 64, 7 yards to the pound.
Carpets:
Brussels, 5-frame, Bigelow.
Wilton, 5-frame, Bigelow.
Cotton flannels:
2f yards to the pound.
3J yards to the pound.
Cotton thread: 6-cord, 200-yard spools, J. & P. Coats.
Cotton yarns:
Carded, white, mule-spun, northern, cones, 10/1.
Carded, white, mule-spun, northern, cones, 22/1.
Denims: Amoskeag.
Drillings:
Brown, Pepperell.
30-inch, Massachusetts D standard, 2.85 yards per pound.
Flannels: White, 4-4 Ballard Vale, No. 3.
Ginghams:
Amoskeag.
Lancaster.
Horse blankets: All wool, 6 pounds each.
Hosiery:
Men’s cotton half hose, seamless, fast black, 20 to 22 ounce, 160 needles, single
thread, carded yarn.
Women’s cotton hose, high-spliced heel, double sole, full fashioned, combed
peeler yarn.
Women’s cotton hose, seamless, 26-ounce, 176 needles, single thread, carded yarn.
Leather:
Chrome calf, dull or bright finish, B grade (range of prices).
Harness, oak, 17 pounds and up, No. 1.
Sole, hemlock, Buenos Aires and Montevideo, middles, No. 1.
Sole, oak, scoured backs, heavy No. 1.
Linen shoe thread: 10s, Barbour.
Overcoatings:
Covert cloth, all wool, double and twist, 14-ounce.
Soft-faced, black, plain twill, 24-ounce.




INDEX NUMBERS— UNITED STATES.

121

Print cloths: 28-inch, 64 by 64.
Sheetings:
Bleached, 10-4 Pepperell.
Bleached, 10-4, Wamsutta S. T.
Brown, 4-4 Indian Head.
Brown, 4-4, Pepperell R.
Brown, 4-4 Ware Shoals L. L., 4 yards to the pound.
Shirtings:
Bleached, 4-4, Fruit of the Loom.
Bleached, 4-4, Rough Rider.
Bleached, 36-inch, Lonsdale.
Bleached, 4-4, Wamsutta
Silk:
Raw, Italian, classical.
Raw, Japan, Kansai, No. 1.
Suitings:
Clay worsted, diagonal, 12-ounce, Washington mills.
Clay worsted, diagonal, 16-ounce, Washington mills.
Serge, 11-ounce, Fulton mills, 3192.
Wool dyed blue, 55-56 inch, 15-ounce.
Tickings: Amoskeag, A.C.A.
Trouserings: Fancy worsted, worsted warp and filling, worsted back, 16£ to 17£
ounce.
Underwear.
Shirts and drawers, merino, natural color, full-fashioned, 50 per cent wool, 24gauge.
Union suits, merino, natural color, 40per cent wool, circular, 24-gauge, lightweight.
Women’s dress goods:
Cashmere, all wool, 8-9 twill, 35-inch, Atlantic Mills, 3602.
Cashmere, cotton warp, 9-twill, 4-4, Atlantic Mills, F.
Cashmere, cotton warp, 4-4, Hamilton.
Panama cloth, all wool, 54-inch.
Poplar cloth, cotton warp and worsted filling, 4-4.

Sicilian cloth, cotton warp, 50-inch.

Wool:
Ohio, fine fleece (X and X X grade), scoured.
Ohio, medium fleece (one-fourth and three-eighths grade), scoured.
Worsted yarns:
2-40s, Australian fine.
2-32s, crossbred stock, white, in skeins.

Fuel and lighting (18 articles).
Candles: Adamantine, 6s, 14-ounce.
Coal:
Anthracite, broken.
Anthracite, chestnut.
Anthracite, egg.
Anthracite, stove.
Bituminous, Georges Creek, at the mine.
Bituminous, Georges Creek, f. o. b. New York harbor.
Bituminous, Pittsburgh (Youghiogheny), lump.
Coke: Connellsville, furnace.
Matches: Globe, No. 1, in cases.




122

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

Petroleum:
Crude.
Refinad, for export, in barrels, cargo lots, S. W., 110 degrees fire test.
Refined 150 degrees fire test, water-white, in barrels (jobbing lots).
Metals and implements ($8 articles).

Augers: Extra, 1-inch.
Axes: Essex, pattern handled.
Bar iron:
Best refined.
Common to best refined, from mill.
Barb wire: Galvanized.
Butts: Loose pin, wrought steel, 3J by 3£ inches.
Chisels: Extra, socket firmer, 1-inch.
Copper:
Ingot, electrolytic.
Sheet, hot-rolled (base sizes).
Wire, bare, No. 8, B. & S. gauge and heavier (base sizes).
Door knobs: Steel, bronze-plated. *
Files: 8-inch mill bastard, Nicholson.
Hammers: Maydole, No. 1£.
Lead: Pig, desilverized.
Lead pipe.
Locks: Common mortise, knob lock, 3$-inch.
Nails:
Cut, 8-penny, fence and common.
Wire, 8-penny, fence and common.
Pig iron:
Bessemer.
Foundry No. 1.
Foundry No. 2, northern.
Gray forge, southern, coke.
Planes: Sargent 414, jack plane.
Quicksilver: Jobbing lots.
Saws:
Cross-cut, Disston No. 2, 6-foot, Champion tooth.
Hand, Disston No. 7, 26-inch.
Shovels: Ames, No. 2, cast-steel, long-handle, round-point, back-strap, black.
Silver: Bar, fine.
Spelter (pig zinc): Western.
Steel billets.
Steel rails.
Steel sheets: Black, No. 27,*box annealed, cold-rolled, United States standard.
Tin: Pig.
Tin plate, coke.
Trowels: Johnson’s, brick, lOJ-inch.
Vises: Solid box, 50-pound.
Wood screws: 1-inch, No. 10, fiathead.
Zinc: Sheet, ordinary numbers and sizes, packed in 600-pound casks.

Lumber and building materials {28 articles).
Brick: Common red, domestic building.
Carbonate of lead: American, in oil.




INDEX NUMBERS— UNITED STATES.

128

Cement:
Portland, domestic.
Rosendale.
Doors: Western white-pine, 2 feet 8 inches by 6 feet 8 inches, If inches thick, 4-panel
No. 2 0 . G.
Hemlock: Base price, Pennsylvania and West Virginia stock.
Lime: Rockport, common.
Linseed oil: Raw, in barrels.
Maple: Hard and soft, 1-inch (4-4), firsts and seconds.
Oak:
White, plain, mixed, rock, mountain or West Virginia stock, 1-inch (4-4) firsts
and seconds.
White, quartered, Indiana, firsts and seconds, 6 inches and up wide, 10 to 16
feet long.
Oxide of zinc: American, extra dry.
Pine:
White, boards, No. 2 bam, 10 inches wide, rough.
White, boards, upper, 1-inch (4-4), rough or dressed.
Yellow, flooring, long-leaf, B, heart-face, rift sawn, 1-& by 2£-face (counted 1
by 3), D. & M.
Yellow, siding, long-leaf, boards, heart-face, 1-inch and lj-inch.
Plate glass:
Polished, glazing, area 5 to 10 square feet.
Polished, glazing, area 3 to 5 square feet.
Poplar: Yellow, 1-inch, firsts and seconds, 7 to 17 inches and up wide, rough.
Putty: Commercial (bulk).
Rosin: Common to good, strained.
Shingles:
Cypress, best all heart, 5 inches wide, 16 inches long.
Red cedas, clear, random width, 16 inches long.
Spruce: 6 to 9 inch, cargoes, eastern.
Tar: Pine.
Turpentine: Spirits of, southern, barrels.
Window glass:
American, single, AA, 25-inch.
American, single, B, 25-inch.

Drugs and chemicals (9 articles).

Alcohol:
Grain, 190 proof, U. S. P.
Wood, refined, 95 per cent.
Alum: Lump.
Brimstone: Crude, domestic.
Glycerin: Refined, chemically pure, in bulk.
Muriatic acid: 20 degrees.
Opium: Natural, in cases.
Quinine: American, in 100-ounce tins.
Sulphuric acid: 66 degrees.

House-furnishing goods (14 articles).

Earthenware:
Plates, cream-colored, 7-inch.
Plates, white granite, 7-inch.
Teacups and saucers, white granite, with handles.




124

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

Furniture:
Bedroom sets, 3 pieces, iron bedstead, hardwood dresser and washstand.
Chairs, bedroom, maple, cane seat.
Chairs, kitchen, common spindle.
Tables, kitchen, 3J-foot.
Glassware:
Nappies, 4-inch, common.
Pitchers, one-half gallon, common.
Tumblers, table, one-third pint, common.
Table cutlery:
Carvers, stag handles, No. 016, 8-inch.
Knives and forks, cocobolo handles, metal bolsters, No. 210.
Woodenware:
Pails, oak-grained, 3-hoop, wire ears.
. Tubs, oak-grained, 3 in nest, C, 19, 21, and 23 inches in diameter.

Miscellaneous (IS articles).

Cottonseed meal.
Cottonseed oil: Prime, summer, yellow.
Jute: Raw M-double triangle, shipment, medium grades.
Malt: Western grade, standard.
Paper:
News, wood, roll, contract.
Manila, wrapping, No. 1 jute.
Proof spirit: Finished goods basis (whisky).
Rope: Pure manila (base sizes, ^-inch and larger to Mar. 10; thereafter f-inch and
larger, three-strand).
Rubber: Para Island, fine.
Soap: Castile, mottled, pure.
Starch: Laundry, 50-pound boxes, in bulk.
Tobacco:
Plug, Climax, 12 pieces to the pound.
Smoking, granulated, Seal of North Carolina, 1-ounce bags.
SUBSTITUTIONS AND ADDITIONS.

Since the issuance of the first wholesale price report covering the
years 1890 to 1901 a number of changes have been made necessary
in the character of the articles included. Certain articles no longer
commercially important or for which satisfactory price quotations
could no longer be obtained have had to be discontinued and other
articles substituted therefor. Thus material changes in the descrip­
tion of 3 articles became necessary in 1902, of 2 articles in 1903, of 1
article in 1904, of 4 articles in 1905, of 6 articles in 1906, of 3 articles
in 1907, of 19 articles in 1908, of 1 article in 1909, of 2 articles in 1910,
of 4 articles in 1911, of 4 articles in 1912, and of 16 articles in 1913.
For 7 of these articles the trade journals no longer supply satisfac­
tory quotations; the manufacture of the particular grade of 15
previously quoted has been discontinued by the establishments
heretofore furnishing quotations, and for 43 articles the substituted
descriptions more nearly represent the present demands of the
trade. In making these substitutions articles were supplied corre­
sponding as closely as possible to those which were previously used.




INDEX NUMBERS---- UNITED STATES.

125

In explanation of the method adopted for computing the relative
prices of articles substituted for others the statement is made in the
report for 1913 (p. 31) that in any year where it was found necessary
to introduce an article to replace another the relative price assigned
to the new article for that year was identical with the relative price
already ascertained for its predecessor in the same year. In other
words, it is assumed that if the price of a commodity in any year,
as represented by the price of a particular grade of the commodity
in question, be correctly expressed by the relative number assigned
it, then its price in the same year may continue to be expressed by
assigning this same relative number to a new grade of the commodity
when substituted for the former grade.
In 1908 a number of articles (11) were added to the list of those
previously included. For such articles no relative price based on
the 1890-1899 period could be computed, owing to the impossibility
of obtaining satisfactory prices for those years. However, it was
deemed necessary to include these new commodities in the several
group relatives for 1908. This was accomplished by dividing the
1908 price of each article in the group, both old and new, by its 1907
price and then computing the simple average of the percentages thus
obtained. This last result, which represents the group index for 1908
expressed as a percentage of the 1907 group index, was then multi­
plied by the relative price of the group in 1907 to produce the group
relative for 1908. Similarly, in succeeding years, the relative prices
for individual articles in the current year, computed on the prices of
the preceding year as a base, have been averaged and the result mul­
tiplied by the group relative for the preceding year to give the cor­
responding group relative for the current year. The general index
number for all commodities in 1908 and in subsequent years has
been computed in the same manner as explained above.
INTERPOLATION.

Prices have not been interpolated for periods when price quotations
were lacking for any of the commodities for which index numbers have
been computed, although whenever new commodities have been intro­
duced or substituted for other commodities, the assumption is tacitly
made that the price of the newly introduced article has changed
by the same percentage as the group as a whole and that the price
of every substituted commodity has varied exactly as the price of the
old article varied up to the time when the substitution was made.
WEIGHTING.

In compiling the present series of index numbers weighting in
its technical sense has not been attempted. Instead, it has been
thought best to use simply a large number of representative staple
articles, selecting them in such a manner as to make them, to a large




126

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

extent, weight themselves. Upon a casual examination it may
seem that by this method a comparatively unimportant commodity—
such, for instance, as tea—is given the same weight or importance
as one of the more important commodities, such as wheat. A
closer examination, however, will disclose the fact that tea enters
into no other commodity under consideration, while wheat is quoted
in the raw state and enters into the two descriptions of wheat flour,
the two descriptions of crackers, and the two descriptions of loaf bread.
This method is the one employed by Mr. Sauerbeck in his compilation
of English prices.
TESTING.

No formal comparison of the index number compiled by the Bureau
of Labor Statistics with other index numbers has been made in any of
the wholesale-price reports. Great care is exercised in the selection
and compilation of the data on which the index is based and the per­
centage changes in prices recorded are therefore quite accurate.
TABLES OF RESULTS.

The following table, which has been reproduced from the 1913
report,1 shows the movement in wholesale prices for the period from
1890 to 1913 in each of the nine principal groups of commodities and
the monthly variations from January to December, 1913:
RELATIVE PRICES OF COMMODITIES, B Y GROUPS, 1890 TO 1913, AND JANUARY TO
DECEMBER, 1913.
(Base period, 1890-1899=100.)

Year.

Farm
prod­
ucts.

Food,
etc.

Cloths
and
cloth­
ing.

Metals Lumber
Drugs
Fuel
and
and
and
and
building chemi­
imple­
lighting. ments.
mate­
cals.
rials.

Housefumish- Miscel­
ing
laneous.
goods.

All
com­
modi­
ties.

1890..........
1891..........
1892..........
1893..........
1894..........

110.0
121.5
111.7
107.9
95.9

112.4
115.7
103.6
110.2
99.8

113.5
111.3
109.0
107.2
96.1

104.7
102.7
101.1
100.0
92.4

119.2
111.7
106.0
100.7
90.7

111.0
108.4
102.8
101.9
96.3

110.2
103.6
102.9
100.5
89.8

111.1
110.2
106.5
104.9
100.1

110.3
109.4
106.2
105.9
99.8

112.9
111.7
106.1
105.6
96.1

1895..........
1896..........
1897..........
1898..........
1899..........

93.3
78.3
85.2
96.1
100.0

94.6
83.8
87.7
94.4
98.3

92.7
91.3
91.1
93.4
96.7

98.1
104.3
96.4
95.4
105.0

92.0
93.7
86.6
86.4
114.7

94.1
93.4
90.4
95.8
105.8

87.9
92.6
94.4
106.6
111.3

96.5
94.0
89.8
92.0
95.1

94.5
91.4
92.1
92.4
97.7

93.6
90.4
89.7
93.4
101.7

1900..........
1901..........
1902..........
1903..........
1904..........

109.5
116.9
130.5
118.8
126.2

104.2
105.9
111.3
107.1
107.2

106.8
101.0
102.0
106.6
109.8

120.9
119.5
134.3
149.3
132.6

120.5
111.9
117.2
117.6
109.6

115.7
116.7
118.8
121.4
122.7

115.7
115.2
114.2
112.6
110.0

106.1
110.9
112.2
113.0
111.7

109.8
107.4
114.1
113.6
111.7

110.5
108.5
112.9
113.6
113.0

1905..........
1906..........
1907..........
1908..........
1909..........

124.2
123.6
137.1
133.1
153.1

108.7
112.6
117.8
120.6
124.7

112.0
120.0
126.7
116.9
119.6

128.8
131.9
135.0
130.8
129.3

122.5
135.2
143.4
125.4
124.8

127.7
140.1
146.9
133.1
138.4

109.1
101.2
109.6
110.4
112.4

109.1
111.0
118.5
114.0
111.7

112.8
121.1
127.1
119.9
125.9

115.9
122.5
129.5
122.8
126.5

1910..........
1911..........
1912..........
1913..........

164.6
162.0
171.3
165.8

128.7
131.3
139.5
137.1

123.7
119.6
120.7
123.7

125.4
122.4
133.9
142.2

128.5
119.4
126.1
127.5

153.2
151.4
148.2
151.8

117.0
120.3
122.9
124.1

111.6
111.1
113.7
118.1

133.1
131.2
133.2
137.1

131.6
129.2
133.6
135.2

» Bulletin of the United States Bureau of Labor Statistics, No. 149, p. 11.




127

INDEX NUMBERS— UNITED STATES.

R E LA TIV E PRICES OF COMMODITIES, B Y GROUPS, 1890 TO 1913, AND JANUARY TO
DECEMBER, 1913—Concluded.

Year.

1913.
January ..
February.
March —
A p ril.......
May.........
June.........
J u ly..
.
A ugust.. .
September
October...
November
December.

Farm
prod­
ucts.

Food,
etc.

160.4
162.3
166.3
167.8
163.1
162.7
162.7
164.6
168.6
168.9
169.7
171.8

132.7
133.1
132.4
132.9
132.5
133.4
135.4
136.4
141.2
144.1
143.3
144.6

Drugs HouseMetals Lumber
Cloths
and
Fuel
and
and
furnish- Miscel­
and
and
laneous.
chemi­
cloth­ lighting. imple­ building
ing
mate­
cals.
goods.
ments.
ing.
rials.

124.2
124.7
124.7
124.6
124.1
123.6
123.7
122.9
123.5
123.5
123.7
123.2

144.3
144.3
142.8
138.9
138.5
139.9
141.0
142.7
143.9
143.2
142.7
143.6

132.8
132.1
130.4
129.4
129.1
127.7
126.1
126.2
126.7
125.5
123.3
120.5

153.1
154.1
154.8
154.7
153.2
152.3
151.2
150.3
150.7
148.9
148.9
149.7

123.0
124.1
123.5
124.5
124.9
124.7
123.5
123.1
123.7
124.7
124.6
124.8

117.5
117.5
118.3
118.3
118.3
118.3
118.3
118.3
118.3
118.3
118.3
118.3

134.9
134.5
134.3
135.5
135.6
136.4
138.6
138.7
140.3
139.6
138.8
137.6

All
com­
modi­
ties.

134.9
135,3
135.1
135.0
134.3
134.1
134.3
134.4
136.1
136.3
135.8
135.7

In order to follow the movement in the two great classes of com­
modities—-raw and manufactured—the following table, which shows
the relative prices by years, 1890 to 1913, and by months, January
to December, 1913, has been prepared:1
RELA TIV E PRICES OF R A W AND MANUFACTURED COMMODITIES, B Y YEARS, 1890
TO 1913, AND B Y MONTHS, JANUARY TO DECEMBER, 1913.
(Base period, 1890-1899=100.)

Year or month.

189
189
189
189
189
189
189
189

0
1
2
3
4
5
6
7

190 0
190 1
1902...................................................
190 3
190 4
190 5
190 6
190 7
190 8
190 9
191 0
191 1
191 2
191 3
1913
January.............................................
February...........................................
March................................................
Ajpril............................ , ...................
May....................................................
June...................................................
July....................................................
August...............................................
September.........................................
October................... ..........................
November.........................................
December..........................................

Manufac­
Raw com­ tured
com­
modities.
modities.
115.0
116.3
107.9
104.4
93.2
91.7
84.0
87.6
94.0
105.9
111.9
111.4
122.4
122.7
119.7
121.2
126.5
133.4
125.5
136.8
139.7
139.9
149.5
147.8

112.3
110.6
105.6
105.9
96.8
94.0
91.9
90.1
93.3
100.7
110.2
107.8
110.6
111.5
111.3
114.6
121.6
128.6
122.2
123.9
129.6
126.6
129.7
132.0

112.9
111.7
106.1
105.6
96.1
93.6
90.4
89.7
93.4
101.7
110.5
108.5
112.9
113.6
113.0
115.9
122.5
129.5
122.8
126.5
131.6
129.2
133.6
135.2

145.2
145.6

132.4
132.7
132.6
132.8
131.9
131.8
131.8
131.5
131.9
131.7
131.3
131.1

134.9
135.3
135.1
135.0
134.3
134.1
134.3
134.4
136.1
136.3
135.8
135.7

U 5.4

143.7
143.6
143.0
144.2
146.5
153.2
155.2
154.1
154.4

i Bulletin of the United States Bureau of Labor Statistics, No. 149, pp. 13 and 14.




All com­
modities.

128

BULLETIN OF THE BUKEAU OF LABOK STATISTICS.

INDEX NUMBERS OF THE UNITED STATES SENATE COMMITTEE ON
FINANCE.1
PUBLICATION.

The Committee on Finance of the United States Senate published
in 1893 an exhaustive report in which the course of wholesale prices in
the United States was shown by means of index numbers for the 52year period from 1840 to 1891. The report was of a special character,
involving an extensive research, and the price data contained therein
have not been continued except in a modified form for subsequent
years.
HISTORY.

A Senate resolution of March 3, 1891, authorized the Committee
on Finance “ to ascertain in every practicable way, and to report
from time to time to the Senate, the effect of the tariff laws upon the
imports and exports, the growth, development, production, and
prices of agricultural and manufactured articles at home and abroad.”
Pursuant to this resolution the committee undertook to ascertain
through accurate and adequate statistics of prices and wages the
changes which had taken place in the condition, as shown by the
relative purchasing power of their earnings, of the great mass of
people in the country for the preceding 50 years. The report of the
committee submitted on March 3, 1893, contained a mass of statistics
relating to wholesale prices compiled by the statistician of the com­
mittee, Roland P. Falkner. (See Report on Wholesale Prices, Wages,
and Transportation, Part 1, Appendix A.)
A continuation of this series of prices has been published in Bulletin
No. 27 issued by the United States Department of Labor (now the
Bureau of Labor Statistics) bringing the data down to 1899. In this
latter series, however, two important changes of method were intro­
duced. The first was in adopting as a basis the average price for
the nine quarters—January, 1890, to January, 1892, inclusive—in
place of the single-date basis, and the second in departing from the
simple average method of allowing to each article equal weight, and
instead combining the index numbers of similar articles to form one
index number, to be used as one article only in calculating the index
numbers for groups and for all commodities.
Another presentation of the data for the years 1860-1880 in some­
what different form (by quarters) is contained in “ Gold Prices and
Wages under the Greenback Standard,” by Wesley C. Mitchell.3
SOURCE OF QUOTATIONS.

The wholesale price quotations included in the report were col­
lected mainly by the United States Department of Labor through its
corps of agents and experts. In some cases experts employed
i Report from the Committee on Finance of the United States Senate on Wholesale Prices, Wages and
Transportation. Mar. 3,1893. 52d Congress, 2d session, Report No. 1394.
* University of California Publications in Economics, vol. 1, Mar. 27,1908.




INDEX NUMBERS— UNITED STATES.

129

directly by the committee furnished the data for the tabulation. As
a rule, the prices were obtained first hand; that is, from records of
actual sales. In the selection of articles for quotations the committee
frequently consulted the representatives of leading industries.
“ The greatest care was exercised to secure absolutely accurate
statements, and the books of merchants and manufacturers were
ransacked in order to obtain figures worthy of every confidence.” 1
BASE PERIOD.

It is explained that the year 1840 was not used as a base because
a statement based on that year “ would have rendered comparatively
useless for purposes of comparison all the articles the quotations for
which begin later than 1840.” 2 For this reason the year 1860, which
would include most of the figures presented, was considered prefer­
able. Moreover, it was believed that “ the year 1860 represents a
period in our industrial development midway between the older
methods of production that prevailed before the war and those which
have come into use since that period. It is also a period of compara­
tively normal prices. The markets of the country had recovered
from the crisis of 1857 and the disturbances of trade caused by the
war had not yet taken place.” 2
Also, a single year, 1860, rather than the average for a period of
years, was taken because “ it was not always practicable to secure for
the articles in question the average prices that would have covered the
period immediately prior to 1860, while in the following year some
prices already manifested the disturbances due to the unsettled state
of national affairs” ; 1860 possessed all the aspects of a nornial
year. “ Its price varies little from that of 1859 or of 1858 on the one
hand and of 1861 on the other. It is therefore quite as proper a
basis of comparison as would be an average of these four years.” 2
PRICES: H O W SHO W N AND COMPUTED.

As a rule the prices used were actual prices obtained at certain
dates. In a few cases average prices for the year were used, when
such prices were considered representative. The index numbers were
calculated on the basis of the January prices in each year where the
prices were quoted by quarters. An exception was made to this rule
in the cases of those articles for which the January price was not the
representative price for the year, as for fresh vegetables, in which
cases the most appropriate month was selected.
NUMBER AND CLASS OF COMMODITIES.

In all there were 230 series of quotations presented, covering the
prices not only of food products and raw materials but also of a very
1 Report from the Committee on Finance of the United States Senate on Wholesale Prices, Wages, and
Transportation, Pt. I, p. 29.
2Idem, p. 28.

94261°—Bull. 173—15----- 9



180

BULLETIN OF TH E BUREAU OF LABOR STATISTICS.

large number of manufactured articles. While all series of quota­
tions did not cover the entire period, owing to the difficulty of ob­
taining for the earlier years prices of articles in use during the later
years, prices for 85 articles quoted in 1891 were secured as far back
as 1840, and for 223 articles as far back as 1860. Those articles
which are articles of luxury only and whose price had increased so
immoderately that they could not be said to enter into consumption
in the same degree as formerly were omitted.
DESCRIPTION AND GROUPING OF COMMODITIES.

The 223 articles were grouped as follows:
Food (53).
Cloths and clothing (28).
Fuel and lighting (10).
Metals and implements (54).
Lumber and building materials (35).
Drugs and chemicals (18).
House-furnishing goods (15).
Miscellaneous (10).
Following is an enumeration of the articles appearing on pages 30
to 52 of Part I of the report:
Food.

Beans.
Bread:
A ship bread.
B ship bread.
Boston crackers (two quotations).
Navy ship bread.
Oyster crackers.
Ship biscuits.
Soda crackers.
Butter.
Cheese.
Coffee, Rio, fair.
Eggs.
Fish:
Cod.
Mackerel, salt, shore, No. 1.
Mackerel, salt, shore, No. 2.
Mackerel, salt, shore, No. 3.
Flour, wheat.
Flour, rye.
Fruit:
Apples, dried.
Currants, Zante.
Raisins.
Lard.
Lard, pure leaf.
Meal, corn, yellow, kiln-dried.




INDEX NUMBERS— UNITED STATES.

Meat:
Bacon, clear.
Beef, loins.
Beef, salt, mess.
Beef, ribs.
Ham, sugar-cured.
Lamb.
Mutton.
Pork, salt, mess.
Milk, fresh.
Molasses:
New Orleans, prime.
Porto Rico, best.
Bice, Carolina, prime.
Salt:
Ashton’s.
Ashton’s Liverpool, fine.
Coarse, solar.
Fine, boiled.
Turk’s Island.
Spices:
Nutmegs.
Pepper, whole, Sumatra.
Starch, corn (two quotations).
Sugar:
Brown.
Cut.
Fair refining.
Refined, crushed, and granulated.
Tallow, prime, city, in hogsheads.
Vegetables:
Fresh, potatoes, white (two quotations).
Cloths and clothing.

Blankets, 11-4, 5 pounds to the pair:
Cotton warp, cotton and wool filling.
Cotton warp, all-wool filling.
Broadcloths:
First quality, black, 54-inch, made from X X X wool.
Second quality, black, 54-inch, made from X X wool.
Calico, Cocheco prints.
Carpets:
Brussels, 5-frame, Bigelow.
Ingrain, 2-ply, Lowell.
Wilton, 5-frame, Bigelow.
Cassimeres, all-wool:
3-4, 7-ounce, Harris double and twist.
3 -4 ,12-ounce, Harris double and twist.
3 -4 ,12-ounce, Harris silk mixed.
Fancy, 3-4, light weight.
Checks, black and white, all-wool, 3-4, 7-ounce, Harris.
Cotton, upland, middling.




131

132

BULLETIN

OF THE BUREAU OF LABOR STATISTICS.

Deniins, Amoskeag.
Drilling, 30-inch, Pepperell.
Hides, dry, Buenos Aires.
Horse blankets, 6 pounds, all-wool.
Leather, harness.
Print cloths:
28-incIi, 64 by 64, Metacomet.
28-inch, 7 yards to the pound, standard.
Shawls, standard, 72 by 144 inches, weighing 42 ounces, made of X X Ohio fleece,
wool.
Sheetings, brown, 4-4, Atlantic A.
Shirtings, bleaches, 4-4, New York mills.
Sole leather, first quality, medium weight, Buenos Aires.
Tickings, Amoskeag, A. C. A.
Wool, Ohio, medium fleece, scoured.
Wool, Ohio, fine fleece, scoured.
Candles, best adamantine.
Coal, anthracite:
Chestnut.
EggF. lump.
Grate.
Pea.
S. lump.
Stove.
Coal, bituminous.
Matches, 8-card.

Fuel and lighting.

Metals and implements.

Anvils, domestic.
Bar iron, best refined, rolled.
Butts, loose, joint, cast, 3 by 3 inch .
Copper, ingot.
Copper, sheet.
Door knobs, mineral.
Iron rods, for making common wood screws.
Iron wire, market, No. 10.
Lead, drop shot.
Lead:
Pig (two quotations).
Pipe.
Locks:
Common mortise.
Common rim.
Meat cutters, Hale’s, No. 12.
Nails, cut.
Pig iron, No. 1, anthracite, foundry.
Pocket knives:
Redwood, iron-lined handle, 2£-incli, 1 blade.
Standard, black horn, brass G. S. handle, 3J-inch, pen, 2 blades.
Standard—
Cocoa, brass G. S. handle, 3J-inch, 2 blades.
Cocoa, brass G. S. handle, 3f-inch, 2 blades.
Cocoa, iron-lined handle, 2f-inch, 1 blade.
Cocoa, iron-lined handle, 3|-inch, 1 blade (two quotations).




INDEX NUMBERS— UNITED STATES,

Pocket knives—Continued.
Standard—Continued.
Cocoa, iron-lined handle, 34-inch, 2 blades.
Cocoa, iron-lined handle, 3f-inch, 2 blades.
Cocoa, iron-lined handle, 4-inch, 1 blade.
Ebony, brass G. S. handle, 3|-inch, 2 blades.
Ivory, brass G. S. handle, 3-inch, pen, 2 blades.
Ivory, brass G. S. handle, 3£-inch, pen, 2 bladen.
Ivory, brass G. S. handle, 3£-inch, pen, 4 blades.
Ivory, brass G. S. handle, 3£-inch, pen, 3 blades.
Pearl, brass-lined handle, 3J-inch, pen, 3 blades.
Pearl, silver-lined handle, 3|-inch, pen, 3 blades.
Pearl, silver-lined handle, 3J-inch, pen, 4 blades.
Redwood, iron-lined handle, 4-inch pruner, 1 blade.
Redwood, iron-lined handle, 5-inch pruner, 1 blade.
Stag, brass G. S. handle, 4|-inch, 3 blades.
Stag, brass-lined handle, 3J-inch, pen, 3 blades.
Stag, brass-lined handle, 3J-inch, pen, 3 blades.
Stag, brass-lined handle, 3f-inch, 4 blades.
Stag, brass-lined handle, 4-inch, 4 blades.
Quicksilver.
Rope:
Manila.
Tarred, American.
Tarred, Russian.
Saws:
Circular, 52-inch, Disston’s.
Crosscut, 6-foot, Disston’s.
Hand, common, Disston’s.
Hand, standard, Disston’s.
Scythes.
Shovels, Ames No. 2, cast-steel D handle, square-point, back-strap.
Spelter, imported.
Wood screws, 1-inch, No. 10, flat head, iron.
Lumber and building materials.

Brick, common domestic building.
Carbonate of lead, in oil.
Cement, Rosendale.
Chestnut, lumber, in the log, not sawed.
Doors, pine, unmolded, 2 feet 4 inches by 6 feet 8 inches, 1J inches thick.
Hemlock, boards, first quality, 1-inch, not planed.
Hemlock, lumber, in the log, not sawed.
Lime, Rockland.
Maple, boards, first quality, 1-inch, rough.
Oak, boards, white, plain, first quality, 1-inch, rough.
Oxide of zinc, American, dry.
Pine, boards, white, clear, 1-inch, not planed.
Pine:
Boards, white, clear, extra, 1-inch, not planed (2 quotations).
Boards, white, common, 1-inch, not planed (2 quotations).
Boards, white, culls, 1-inch, not planed.
Flooring, white, extra, 1-inch, not planed.
Lumber, in the log, not sawed.



134

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

Plate glass, polished:
Unsilvered, area 1 to 3 square feet.
Unsilvered, area 3 to 5 square feet.
Unsilvered, area 5 to 10 square feet.
Unsilvered, area 10 to 40 square feet.
Unsilvered, area 40 to 80 square feet.
Unsilvered, area 80 to 100 square feet.
Putty.
Shingles, pine:
16 inches, XXX.
16 inches, extra X X X cut.
16 inches, extra X X X sawed.
Spruce boards, 1-inch.
Tar, Wilmington.
Turpentine.
Window glass:
American, 10 by 14.
French, 10 by 14, firsts, single.
French, 10 by 14, thirds, single.
Drugs and chemicals.

Alcohol.
Alum, lump, crystal.
Bichromate of potash.
Blue vitriol.
Brimstone, crude.
Calomel.
Copperas.
Flaxseed.
Glycerin, refined.
Linseed oil.
Mercury.
Muriatic acid.
Opium.
Quinine.
Soda ash.
Sugar of lead:
Brown.
White.
Sulphuric acid.
House-furnishing goods.

Furniture:
Chairs, bedroom, maple, cane seat.
Chairs, kitchen, common, spindle.
Tables, kitchen, pine, 3£-foot.
Glassware:
Bowls, 8-inch.
Goblets, common.
Pitchers, J-gallon.
Sets, finished.
Tumblers, J-pint.
Pails, wooden:
2-hoop (2 quotations).
3-hoop.
Tubs, wooden (4 quotations).



ISTDEX NUMBERS— UNITED STATES.

135

Miscellaneous.

Powder, rifle (2 quotations).
Rubber, Para.
Soap, castile, mottled, imported.
Starch:
Ontario.
Ordinary laundry.
Pearl.
Pure.
Refined.
Silver gloss.
SUBSTITUTIONS AND ADDITIONS.

In the compilation of the index numbers no substitution of one
grade or quality of an article for another grade or quality of the same
article or for a different article previously included was made. In
cases where quotations on a particular article could no longer be had,
or where the article had ceased to be representative, it was discon­
tinued and the index number was computed on the remaining articles.
Additions to the list of articles were made from time to time as
occasion demanded, such additions being carried into the index
number for the year.
INTERPOLATION.

Interpolation of prices was not resorted to in the preparation of
the index numbers. In cases where prices for particular periods were
lacking, the article in question was temporarily discontinued.
WEIGHTING.

The committee calculated three distinct index numbers. The first
was unweighted, while the second and third were weighted by assign­
ing to each article an importance in the result equal to its importance
in family consumption. The basis selected by the committee for
determining this consumption was the Seventh Annual Report of the
Commissioner of Labor, showing the values of various articles con­
sumed by a large number of families which were considered typical
of the expenditures of the mass of the people. This information is
summarized in the following table showing the distribution of ex­
penditure for 2,561 normal families.
DISTRIBUTION OF E XPEN DITU RE FOR 2,561 FAMILIES.




Group.

Bent...........................
Food...........................
Fuel...........................
Clothing....................
Lighting....................
All other purposes....

Per cent of
expendi­
ture for
each pur­
pose.

Propor­
tions of
10,000.

15.06
41.03
5.00
15.31
.90
22.70

1,506
4,103
500
1,531
90
2,270

100.00

10,000

186

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

This table shows the main groups of family expenditures only, and
these were not sufficiently specific for the purpose. In order to secure
accurate and specific data as to the composition of these groups them­
selves, therefore, 232 special budgets of family expenses were collected.
The table based on 2,561 families was then used to secure the propor­
tion of the groups entering into consumption, while a table based on
232 families was used to secure the distribution of expenditure within
the groups themselves. The figures thus secured, showing the relative
weight in consumption of each article contained in the family budgets,
were then applied to the various articles in the index numbers. Few
articles were found, however, with identical descriptions, hence a
method of grouping was resorted to. For example, two or more
articles contained in the index numbers were often grouped to repre­
sent one article of the family budget, thus, “ ham,” “ bacon,” and
“ pork” were considered equivalent to the “ hog products” of the
family budgets, and an arithmetical average was made of the index
numbers of these three articles, which was made the index number
of hog products and given its appropriate weight. Briefly described,
the method as applied to the food group was as follows: The figure
denoting the importance of each article in the group in a given year
wa3 multiplied by its simple index number (or the average where two
or more articles were grouped), and the sum of these products was
divided by the total of the figures denoting importance for the
weighted index number for the general group of food. This method
was used in securing similar weighted index numbers for the other
years of the period. It is seen that according to this method the
same weights were used for each year of the period, although they
were based on the consumption of families in 1891.
The following table shows the weighted relative prices or index
numbers in 1891 for each of the groups considered, together with the
total weighted relative price or index number:
R E LATIVE PRICES OF A LL ARTICLES IN 1891, MEASURED B Y CONSUMPTION.
(Base period, 1860=100.)

Impor­
tance.

Weight­
ed index
number.

Result.

Rent............................
Food.............................
Fuel.............................
Lighting.......................
Clothing.......................
All other purposes . —

1,506
4,103
500
90
1,531
2,270

100.0
103.7
98.1
48.1
75.1
95.3

1,506,000
4,254,811
490,-500
43,290
1,162,029
2,164,,096

Total....................

10,000

96.2

9,620,726

Group.

In the above table the Weighted index numbers for the various
groups were found in practically the same manner as has previously
been, described for food, except that rent and certain items entering



187

INDEX NUMBERS— UNITED STATES.

into the group “ All other purposes” were considered to have remained
unchanged. These were then multiplied by the figures denoting im­
portance, and the sum of the products divided by the total of the
figures denoting importance (10,000) to secure the weighted total of
96.2. The remaining years were treated in a similar manner.
The items of budget expenditure considered as remaining un­
changed (rent, taxes, insurance, etc.), constituted 31.40 per cent of
the total expenditure, leaving 68.60 per cent as affected by changes
in prices. Another set of index numbers for such articles was made
by the committee by assigning a total numerical weight of 6,860 to
such articles and working out the index numbers on that basis.
This resulted in a slightly different total index number.
TESTING.

The accuracy of the results secured was tested by comparison of the
index numbers with those of the London Economist and of Sauerbeck
for England.1
TABLES OF RESULTS.

The index numbers computed by the three methods, i. e., simple
average, average of all articles weighted according to consumption,
and average of fluctuating articles only, weighted according to con­
sumption, are shown in the following table. The prices are in currency.2
R E LA TIV E PRICES IN EACH Y EA R , 1840 TO 1891, FOR ALL ARTICLES GROUPED B Y
DIFFERENT METHODS.

Year.

1840
1841.
1842.
1843.
1844.
1845.
1846.
1847.
1848.
1849.
1850.
1851.
1852.
1853.
1854.
1855.
1856.
1857.
1858.
1859.
1860.
1861.
1862.
1863.
1864.
1865.

All articles All articles
averaged
averaged
according
according
to im­
to im­
portance,
All articles portance,
certain
comprising
simply
expendi­
68.60 per
averaged.
cent of
tures
being
total
expendi­
considered
uniform.
ture.
116.8
115.8
107.8
101.5
101.9
102.8
106.4
106.5
101.4
98.7
102.3
105.9
102.7
109.1
112.9
113.1
113.2
112.5
101.8
100.2
100.0
100.6
117.8
148.6
190.5
216.8

98.5
98.7
93.2
89.3
89.8
92.1
96.7
96.7
92.0
88.9
92.6
99.1
98.5
103.4
100.4
106.3
108.5
109.6
109.1
102.0
100.0
95.9
102.8
122.1
149.4
190.7

97.7
98.1
90.1
84.3
85.0
88.2
95.2
95.2
88.3
83.5
89.2
98.6
97.9
105.0
105.0
109.2
112.3
114.0
113.2
102.9
100.0
94.1
104.1
132.2
172.1
232.2

Year.

1866...............
1867...............
1868...............
1S69...............
1870...............
1871...............
1872...............
1873...............
1874...............
1875...............
1876...............
1877...............
1878...............
1879...............
1880...............
1881...............
1882...............
1883...............
1884...............
1885...............
1886...............
1887...............
1888...............
1889...............
1890...............
1891...............

All articles All articles
averaged
averaged
according according
to im­
to im­
All articles portance,
portance,
simplycertain
comprising
averaged.
expendi­
68.60 per
cent of
tures
being
total
considered expendi­
uniform.
ture.
191.0
172.2
160.5
153.5
142.3
136.0
138.8
137.5
133.0
127.6
118.2
110.9
101.3
96.6
106.9
105.7
108.5
106.0
99.4
93.0
91.9
92.6
94.2
94.2
92.3
92.2

160.2
145.2
150.7
135.9
130.4
124.8
122.2
119.9
120.5
119.8
115.5
109.4
103.1
96.6
103.4
105.8
106.3
104.5
101.8
95.4
95.5
96.2
97.4
99.0
95.7
96.2

187.7
165.8
173.9
152.3
144.4
13G.1
132.4
129.0
129.9
128.9
122.6

113.6
104.6
95.0
104.9
108.4
109.1
106.6
102.6
93.3
93.4
94.5
96.2
98.5
93.7
94.4

i Report from the Committee on Finance of the United States Senate on Wholesale Prices, Wages, and
Transportation, Pt. I, pp. 226,227, and 256.
* Idem, Pt. I, p. 9.




138

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

A convenient summary of the foregoing table, by periods of five
years, is found in the table following:1
R ELATIVE PRICES, B Y 5-YEAR PERIODS, 1840 TO 1891, FOR A L L ARTICLES GROUPED
B Y DIFFEREN T METHODS.
1

Period.

1840-1844..............
1845-1849..............
1850-1854..............
1855-1859..............
1860-1864..............
1865-1869..............
1870-1874..............
1875-1879..............
1880-1884..............
1885-1889..............
1890-1891..............

All articles All articles
averaged
averaged
according according
to
impor­
to
impor­
All articles tance, cer­
tance,
simply
comprising
tain
ex­
averaged. penditures 68.60 per
being con­
cent of
total ex­
sidered
uniform. penditure.
108.8
103.2
106.6
108.2
131.5
178.8
137.5
110.9
105.3
93.2
92.3

93.9
93.3
99.4
107.1
114.0
156.5
123.6
108.9
104.4
96.7
96.0

91.0
90.1
99.1
110.3
120.5
182.4
134.4
112.9
106.3
95.2
94.1

INDEX NUMBERS OF THE ANNALIST.
PUBLICATION.

The Annalist, a magazine of finance, commerce, and economics,
published weekly in New York City, has compiled an index number
based on the wholesale prices of 25 food commodities in the United
States. These articles are so selected as to represent a theoretical
family food budget.
HISTORY.

The publication of this index number began with the first issue
of the Annalist on January 20, 1913, and has been continued weekly
since that date in connection with the exhibit of various other items
of business activity appearing under the caption of “ Barometrics.”
SOURCE OF QUOTATIONS.

The prices used in the computation of the index number are those
prevailing in the New York and Chicago markets.
BASE PERIOD.

The 10 years 1890-1899 constitute the base period used in com-*
puting the index number.
PRICES: H O W SHO W N AND COMPUTED.

During the period from May 19 to September 1,1913, the Annal­
ist published in each week’s issue the mean price of each selected
commodity dtiring the preceding week, together with the relation
i Report from the Committee on Finance of the United States Senate on Wholesale Prices, Wages, and
Transportation, Pt. I , p. 10.




INDEX NUMBERS— UNITED STATES,

139

of such price to the price for the base period 1890-1899. The sum
of these relative prices, divided by 25 (the number of commodities),
constitutes the index number for the week. In all other issues of
the Annalist up to date no exhibit of wholesale prices is made in
connection with the presentation of the index number.
NUMBER AND CLASS OF COMMODITIES.

As previously stated, 25 articles of food are included in the index.
These are listed in the Annalist of May 19, 1913, and in subsequent
numbers to September 1 of the same year, as follows:
Steers.
Hogs.
Sheep.
Beef, fresh.
Mutton, dressed.
Beef, salt.
Pork, salt.
Bacon.
Codfish, salt.
Lard.
Potatoes.
Beans.
Flour, rye.

Flour, wheat, spring.
Flour, wheat, winter.
Corn meal.
Rice.
Oats.
Apples, evaporated.
Prunes.
Butter, creamery.
Butter, dairy.
Cheese.
Coffee.
Sugar, granulated.

DESCRIPTION AND GROUPING OF COMMODITIES.

The following description of the commodities included in the
index number has been supplied by the publishers of the Annalist:
New York markets.

Codfish (Georges), corn meal, rice, beans, evaporated apples, California prunes,
extra creamery butter, New York State dairy butter, cheese (New York State,
whole milk, held), No. 7 Rio coffee, fine granulated sugar, fresh beef, dressed mutton,
salt beef, salt pork, wheat flour (winter straights and spring patents), Middle West
lard, and rye flour.
Chicago markets.

Good to choice steers, hogs (250-300 pound packers and fair to select butcher’s),
sheep (good to choice wethers), bacon (short, clear sides), white potatoes, and cash
oats (2 white, 3 white, and standards).
SUBSTITUTIONS AND ADDITIONS.

The statement is made in the Annalist of October 13, 1913, that
“ a substitution has been made which affects the current numbers
seven-tenths of 1 per cent.” This was occasioned by the substitu­
tion of “ good to choice steers” for “ prime to fancy steers,” as quota­
tions on the latter grade had become nominal in the Chicago market.
The entire index number was recast so as to conform to the change
made in this respect. No additions to the list of commodities have
been made.




140

BULLETIN OF THE BUREAU OF LABOR 'STATISTICS,
INTERPOLATION.

No prices have been interpolated, as far as the published informa­
tion discloses.
WEIGHTING.

The index number is unweighted and is obtained by computing the
simple arithmetic mean of the relative prices of the different com­
modities.
TESTING.

No test of the index number by means of comparison with other
indexes or by other means is shown in any issue of the Annalist.
TABLE OF RESULTS.

The course of the index number by years from 1890 to 1914 and by
months during 1912, 1913, and 1914 is shown in the following state­
ment furnished by the publishers of the Annalist:
IN D E X NUMBERS, B Y YEARS, 1890 TO 1914, AND B Y MONTHS, 1912, 1913, AND 1914.
(Base period, 1890-1899=100.)

Yearly.
Index
number.

Year.

1890.............................
1891.............................
1892.............................
1893.............................
1894.............................
1895.............................
1896.............................
1897.............................
1898.............................

109.252
119.488
108.624
116.100
102.076
94.604
80.096
84.092
92.208

Index
number.

Year.

1 8 9 9 . : ......................
1900...........................
1901...........................
1902...........................
1903.........................
1904...........................
1905..........................
1906...........................
1907...........................

93.348
99.388
104.656
116.264
107.516
108.664
110.652
114.364
117.940

Index
number.

Year.
1908...........................
1909...........................
1910...........................
1911...........................
1912...........................
1913...........................
1914...........................

125.756
133.952
137.172
131.068
143.254
139.980
146.069

Monthly.
1913

1912

Month.

January...
February..
March.......
April.........
May..........
June..........
July..........
August___
September.
O ctober...
November.
December.




Index
number.
139.681
138.012
143.515
152.326
152.958
148.193
143.285
140.871
140.794
141.861
139.543
138.013

Month.

January...
February.
March.......
April........
May..........
June.........
July..........
August___
September
October...
November.
December.

1914

Index
number.
137.197
137.866
139.926
141.971
137.927
137.750
139.839
139.927
142.290
141.664
141.558
141.847

Month.

January...
February..
March.......
April........
May..........
June.........
July..........
August___
September
October...
November.
December.

Index
number.
142.452
141.278
142.079
141.120
139.231
141.433
144.879
152.106
160.776
150*245
150.008
146.779

INDEX NUMBERS— UNITED STATES.

141

INDEX NUMBERS OF BRADSTREET’S.
PUBLICATION.

This “ index” 1 represents the record of wholesale prices of staple
articles in the primary markets of the United States and is now pub­
lished every month. Formerly it was issued only every quarter.
HISTORY.

Bradstreet’s index had its beginning in the issue of September 21,
1895, of the periodical of that name, which presented a table of com­
parative prices of 110 staple articles for each quarter from October 1,
1890, to July 1, 1895, under the heading, “ Five years’ prices for 110
staple products.”
The compiler evidently had in mind a record of price movements
in the United States similar to that furnished by Sauerbeck’s index
of English prices, as he refers to it in the introductory paragraph as
follows:
“ In Sauerbeck’s latest record of prices of staple products in the
United Kingdom during the past 30 years it is shown that quo­
tations for 50 selected articles by groups averaged lower in 1894
than in any of the 16 next preceding years, as well as lower than
in the 11-year period from 1867 to 1877, which the eminent statis­
tician selected as representing the normal and called 100. ” 2
In explanation of the data presented in Bradstreet’s the compiler
says: “ In the accompanying exhibit of comparative prices of staple
articles at primary markets in the United States at quarterly intervals,
beginning with the autumn of 1890, prior to the Baring crash, and
ending with July 1, 1895, is furnished what should prove an oppor­
tunity for tracing the relative effects of panic and trade depression
on the prices in different lines of business.” 2
In the issue of October 26, 1895, the report was extended to
include prices for October 1 of that year and the statement made
that “ it will be recalled this work was first made public by Brad­
street’s late in the summer with a comparison of quotations for more
than 100 articles of merchandise and produce at quarterly intervals
during the past five years. ”
Again in the issue of January 11, 1896, the author has this to say
in discussing the compilation: “ Perhaps the most elaborate exhibit
which has been compiled of comparative prices of staple products,
breadstuffs, live stock, provisions, fresh and dried fruits, hides and
leather, raw and manufactured textiles, coal and coke, mineral and
vegetable oils, building materials, chemicals and drugs, and others
1Not an index in the true sense of the word, being rather a number representing the aggregate of per
pound prices of certain selected commodities.
2Bradstreet’s, Saturday, Sept. 21,1895, p. 594.




142

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

is presented in connection with this article. The quotations are
given for quarterly periods during the past five calendar years, and
probably few, if any, better outlines of the movement of quotations
have thus far been presented. ”
The comparative prices continued to be presented on the first of
each quarter until May 8, 1897, when in connection with “ A study
of prices” an index number was published for the first time.
The index as constructed was simply the sum obtained by adding
the per pound prices of the different articles included. At first
it was not expressed in dollars and cents, but as an abstract num­
ber. No attempt was made at weighting, nor was consumption
taken into account, so that the result was “ not an absolute indication
of the price movement based on the proportions in which each of the
products and articles are used, but a fair indication of the tendency. ”
The author stated that only 97 articles were included in the index,
but as actual prices were shown for 108 articles and only 10 articles
were stated to be excluded it would appear that the index comprised
98 articles.
In the issue of June 11, 1898, actual prices were shown for 107
articles, quotations for onions being dropped, and the index number
was revised to exclude the price of quicksilver. The only explana­
tion given for this was: “ It might be stated in passing that the low­
ering of the index number is accounted for by the deduction of the
price of quicksilver.” The index for the period October, 1890, to
June, 1898, was thus recomputed by deducting the price of quicksil­
ver; for instance, the old index for January, 1898, was 80,149 and the
new one was 75,084; that is, the price of quicksilver on January 1,
1898 ($0.5065 per pound), was deducted from 80,149, leaving 75,084
as the new index. There were still 10 articles, excluding quicksilver,
not included in the index, but for which comparative prices were
given.
Again on September 10, 1898, the index appeared with revised
figures. This revision was due to the quotation of a different grade
of hides. Previous to this time prices had been quoted for dry
Buenos Aires hides, but for some reason not stated the new quota­
tions were for No. 1 native steer hides. The difference between
the prices of these two grades of hides in August, 1898, was $0.0925,
and this deduction from the former index for August (77,481) leaves
76,556 as the new index. This amount was deducted from every
index figure already established as far back as October, 1890.
In the issue of October 12, 1901, the first group indexes were shown
and consisted of the sum of the per pound prices for all of the articles
included in the group. The sum of the 13 groups was the index
shown for all commodities. The general index was expressed in
dollars and cents and continued to be stated this way until April 9,




INDEX NUMBERS— UNITED STATES.

143

1904, when it was restated in dollars, cents, and fractions thereof.
This was not a revision of the index, but simply a change in the method
of pointing off. The index numbers for the groups had been
expressed in this way for some time before this date. The index
now began with January 1, 1892, instead of October 1, 1890, as
formerly, and was computed upon the basis of the revision of Septem­
ber, 1898, until December 16, 1905, when a general index “ revised
to exclude some staples showing wide fluctuations” in price was
published. It is not stated in connection with these figures what
articles were excluded or on how many commodities the revised
index number was based. The exhibit as published contained the
index number by quarters from January 1, 1892, to October 1, 1898,
and by months from January 1, 1899, to December 1, 1905, inclusive.
No further revision of the index number appears to have been made.
SOURCE OF QUOTATIONS.

The source of these quotations is not disclosed, but it is stated
that they are from primary markets.
BASE PERIOD.

No base period was selected in the compilation of the index num­
ber, the need of such being obviated by the method employed, which
consists simply in adding together the prices per pound of the various
selected articles at the date named.
PRICES: H O W SHOW N AND COMPUTED.

Prices are published each month for a selected list of representative
commodities. These prices are shown for the first day of the cur­
rent month and, for purpose of comparison, the first day of several
preceding months and the first day of the corresponding month in
the preceding year. No range of quotations is shown in any case,
and it is evident that a single price has been used, but whether either
extreme or the mean was taken it is impossible to determine with
the source of quotations unknown. No yearly average actual prices
are published.
In the issue of May 8, 1897, the price per pound of each article
was shown as quoted on the first of April, the articles being grouped
under the amount paid per pound. The list was prefaced by the fol­
lowing statement: “ Bradstreet’s exhibit of 98 staple, raw and manu­
factured articles, products, produce and live stock classified accord­
ing to the cost of 1 pound of each on April 1,1897.” This exhibit was
continued at intervals for about a year and then dropped. In many
cases the figures appear to have been approximations. The list
as published in Bradstreet’s of July 10,1897, follows. The prices are
for July 1.




144

BULKBTI.V OK THE BUREAU OF LABOR STATISTICS.

Cost per pound.

$0.0007 Connellsville coke, southern coke.
. 001 Bituminous coal, brick, iron ore.
. 002 Anthracite coal.
. 003 Salt, southern pig iron, crude petroleum, rosin, lime, phosphate rock.
. 004 Bessemer pig iron, pine lumber, cotton seed.
. 005 Corn, eastern pig iron, tar, spruce, hemlock.
. 006 Steel billets.
. 007 Oats, barley, rye, potatoes, hay, sulphuric acid.
. 009 Steel rails, steel beams, refined petroleum.
. 015 Wheat, milk, peas, nails, alum, bicarbonate of soda.
. 02 Flour, molasses, beans, paper, caustic soda.
. 03 Hogs, lemons, hemp, jute, tin plates, cottonseed oil, turpentine, glass, flax.
.04 Beeves, sheep, bread, barreled beef, pork, lard, codfish, rice, linseed oil,
raisins, lead, nitric acid.
. 05 Pigs, sugar, currants, borax, bacon.
. 06 Eggs.
. 07 Beef carcasses, mutton, coffee, olive oil, hops.
. 08 Horses, mackerel, cheese, cotton.
. 10 Hams.
. 11 Copper.
. 12 Castor oil.
. 14 Standard sheetings, cotton sheetings, tin, tobacco.
. 15 Butter.
. 175 Print cloths.
. 18 Tea, Buenos Aires hides, carbolic acid.
. 20 Hemlock hides, wool.
. 27 Union leather.
. 29 Oak leather.
. 31 Ginghams.
. 34 Alcohol.
.50 Australian wool.
. 52 Quicksilver.
. 84 Rubber.
NUMBER AND CLASS OF COMMODITIES.

In the beginning 110 articles were shown in the comparative table
of actual prices, but now only 106 are included, and of these only 96
are included in the index. Oranges, naphtha, onions, and aluminum
were the articles dropped from the table of comparative prices, but
the reason for their discontinuance is not given. Two of these, onions
and aluminum, were never included in the compilation of the index.
Two articles that at first were included in the index are no longer
included—namely, quicksilver and rubber—but these are still shown
in the table of actual prices. When these articles were dropped the
index was recomputed from that date to the beginning, necessitating
a new index figure for every previous date. The list of articles in­
cludes both raw and manufactured commodities that are of general
consumption in the United States.




INDEX NUMBERS---- UNITED STATES.

145

DESCRIPTION AND GROUPING OF COMMODITIES.

The articles on which the index is based are divided into 13 gen­
eral groups, as follows: Breadstuffs, live stock, provisions and gro­
ceries, fresh and dried fruits, hides and leather, raw and manufactured
textiles, metals, coal and coke, mineral and vegetable oils, naval stores,
building materials, chemicals and drugs, and miscellaneous. Since
October 12, 1901, an index has been computed usually for each of
the different groups separately. The sum of the indexes for the 13
groups is the index for the whole number of articles. Index num­
bers for years are computed by averaging the 12 monthly totals.
The following list is an enumeration of the articles, under the
various groups, for which actual prices are shown in the comparative
price table. As before stated, only 96 of these 106 articles are in­
cluded in the index as now compiled. This is the list and description
of articles as printed in Bradstreet’s of December 12, 1914:
Breadstuffs (6

Wheat, No. 2, red winter, in elevator.
Corn, No. 2, mixed, in elevator.
Oats, No. 2, mixed, in elevator.

Barley, No. 2 (Milwaukee).
Bye, western.
Flour, straight winter.

Live stock (4 articles).

Beeves, best, native steers (Chicago).
Sheep, prime (Chicago).

Hogs, prime (Chicago).
Horses, average, common to best (Chi-

Provisions and groceries (24 articles).

Beef, carcasses (Chicago).
Hogs, market pigs, carcasses (Chicago).
Mutton, carcasses (Chicago).
Milk (New York).
Eggs, State, fresh (New York).
Bread (New York).
Beef, family.
Pork, new mess.
Bacon, short ribs, smoked (Chicago).
Hams, smoked.
Lard, western steam.
Butter, creamery, State, best.

Cheese, choice, east factory.
Mackerel, No. 1, bays (Boston).
Codfish, large dried.
Coffee, Bio, No. 7.
Sugar, granulated.
Tea, Formosa Oolong, superior.
Molasses, New Orleans, prime.
Salt, fine domestic, sacks.
Bice, domestic, good.
Beans (New York), choice marrow.
Peas, choice (New York).
Potatoes, eastern.

Fresh and driedfruits (6 articles).

Apples (State).
Cranberries, Cape Cod, fancy.
Peanuts, best Virginia, in hull.

Lemons, choice.
Baisins, layer.
Currants, new, dried.

Hides and leather (4 articles).

Native steer hides, No. 1.
Hemlock, packer, middleweight^, No. 1.
94261°—Bull. 173—15------10




Union, middlebacks, tannery run.
Oak, scoured backs. No. 1,

146

BULLETIN OF THE BUREAU OF LABOR STATISTICS.
Raw and manufactured textiles (11 articles).

Cotton, middling uplands.
Wool, Ohio and Pennsylvania X, washed
(Boston).
Wool, Australian super combing, scoured.
Hemp, manila.
Jute, average of grades.

Silk, best No. 1, filature.
Flax, New Zealand, spot.
Print cloths, 64s (Boston).
Standard sheetings (Boston).
Ginghams, Amoskeag staple (Boston).
Cotton, sheetings, southern, 3 yards.

Metals (13 articles).

Iron ore, old range, Bessemer, hematite.
Pig iron, No. 1 foundry, eastern (New
York).
Pig iron, No. 2 foundry, southern (Bir­
mingham).
Pig iron, Bessemer (Pittsburgh).
Steel billets, Bessemer (Pittsburgh).
Steel rails, standard (Pittsburgh).

Tin plates, American (Pittsburgh).
Steel beams (Pittsburgh).
Silver, commercial bars (New York).
Copper, electrolytic (New York).
Lead, pig, western (New York).
Tin, pig, spot (New York).
Quicksilver (San Francisco).

Coal and coke (4 articles).

Anthracite, stove sizes (New York).
Bituminous (Pittsburgh), f. o. b. Chicago.

Connellsville coke, short ton, f. o. b.
Southern coke (Chattanooga).

Mineral and vegetable oils (6 articles).

Petroleum, crude, in barrels (New York).
Petroleum, refined, in cases.
Linseed.

Cottonseed, crude, prime (New York).
Castor, No. 1.
Olive, Italian, in barrels.

Naval stores (3 articles).

Rosin, good, strained (Savannah).
I Tar, regular (Wilmington, N. C.).
Turpentine, machine, regular (Savannah). |
Building materials (7 articles).

Brick, Hudson River, hard.
Lime, eastern common.
Nails, wire, from store, base prices.
Glass, window, 10 by 15.

Pine, yellow, 12-inch and under.
Timber, eastern spruce, wide random.
Timber, hemlock, Pennsylvania, random.

Chemicals and drugs (11 articles).

Alum.
Bicarbonate of soda, American.
Borax, crystals.
Carbolic acid, in bulk.
Caustic soda, 60 per cent.
Nitric acid, 36 degrees.

Sulphuric acid, 66 degrees.
Phosphate rock, South Carolina, ground.
Alcohol, 94 per cent.
Opium.
Quinine, domestic, in bulk.
Miscellaneous (7 articles).

Hops, New York State, choice.
Rubber, upriver, Para, fine new.
Tobacco, medium leaf, Burley (Louis­
ville).




Paper, news, roll.
Ground bone, fine, steamed.
Hay prime (New York).
Cotton seed (Houston).

147

INDEX NUMBERS---- UNITED STATES.
SUBSTITUTIONS AND ADDITIONS.

Numerous changes in description of the articles have occurred from
time to time, but only once, apparently, has a substitution been con­
sidered of enough importance to justify any change in the index.
This was in the case of dry Buenos Aires hides, for which were sub­
stituted No. 1 native steer hides, when the index was recomputed
back to the beginning.
INTERPOLATION.

No method of supplying missing data is disclosed, if such has been
found necessary.
W EIGHTING.

Apart from the basic plan of expressing in terms of dollars and cents
the value of 1 pound avoirdupois of each commodity, there is no
attempt at assigning varying degrees of importance to the different
articles included in the index.
TESTING.

No test has been made of the index so far as known, other than a
comparison in parallel columns of the numbers with those published
by the London Economist, the Statist (Sauerbeck’s), the Canadian
Department of Labor, and La Reforme ficonomique for the same
dates.
TABLES OF RESULTS.

The following table, appearing in Bradstreet’s issue of December
12, 1914, illustrates the manner in which the group index numbers
for different dates are shown:
B R A DSTREET’ S IN D E X NUMBERS FOR SPECIFIED DATES.

Commodity.

Dec. 1,
1913.

Aug. 15,
1914.

Nov. 1,
1914.

Dec. 1,
1914.

Breadstuffs........................................................................
Live stock..........................................................................
Provisions..........................................................................
Fruits.................................................................................
Hides and leather.............................................................
Textiles..............................................................................
Metals...............................................................................
Coal and coke....................................................................
Oils.....................................................................................
Naval stores.......................................................................
Building materials............................................................
Chemicals and drugs.........................................................
Miscellaneous...................................................................

$0.0947
.4480
2.4513
.1950
1.3500
2.5625
.6720
.0070
.3539
.0771
.0831
.5867
.3477

$0.1001
.4860
2.5006
.2305
1.4300
2.3704
.8707
.0067
.3755
.0784
.0822
1.0096
.3088

$0.1116
.4415
2.3753
.1736
1.4175
2.1854
.5279
.0067
.3434
.0794
.0816
.8529
.2652

$0.1139
.4220
2.3689
.1648
1.4250
2.1892
.5830
.0066
.3503
.0770
.0821
.9979
.2547

Total........................................................................

9.2290

9.8495

8.8620

9.0354




148

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

For some years past a yearly index has been computed by averag­
ing the 12 monthly indexes. The manner of presenting this in­
formation is shown by the following table, which is reproduced
from Bradstreet’s of December 12; 1914.
1914..................... .................... $8.9035
1902.....................
1913..................... .................... 9.2076 1901.....................
1912..................... .................... 9.1867 1900.....................
1911..................... .................... 8.7132 1899.....................
1910..................... .................... 8.9881 1898.....................
1909..................... .................... 8.5153 1897.....................
1908..................... .................... 8.0094 1896.....................
1907..................... .................... 8.9045 1895.....................
1906..................... .................... 8.4176 1894.....................
1905..................... .................... 8.0987 1893.....................
1904..................... .................... 7.9187 1892.....................
1903..................... ..................... 7.9364
Ten-year average, 1902 to 1911, inclusive, $8.3377.
Ten-year average, 1892 to 1901, inclusive, $6.9696.

.................... $7.8759
.................... 7.5746
.................... 7.8839
.................... 7.2100
.................... 6.5713
.................... 6.1159
.................... 5.9124
.................... 6.4346
.................... 6.6846
.................... 7.5324
.................... 7.7769

The index numbers computed from the wholesale prices of 96
articles on the first day of each month from January, 1903, to Decem­
ber, 1914, inclusive, are shown in the subjoined table, also reproduced
from Bradstreet’s issue of December 12, 1914.
BRADSTREET'S IN D E X NUMBERS, JANUARY, 1903, TO DECEMBER, 1914, INCLUSIVE.

Index number: First of each month.
Year
Jan.
1903.
1904.
1905.
1906.
1907.
1908.
1909.
1910.
1911.
1912.
1913.
1914.

Feb.

Mar.

Apr.

May.

June.

July.

Aug.

Sept.

Oct.

$8.0789 $8.0824 $8.1300 $8.1247 $7.9567 $7.8751 $7.8706 $7.7473 $7.7583 $7.9083
7.9885 8.0973 8.0882 7.9690 7.9352 7.9877 7.6318 7.7623 7.7845 7.9213
8.0827 8.0805 8.0979 7.9996 7.9700 7.9073 7.9160 8.1111 8.2795 8.2298
8.3289 8.2415 8.2321 8.2987 8.3054 8.3203 8.2835 8.3376 8.4528 8.5580
8.9172 8.9953 8.1293 '8.9640 8.9356 8.9901 9.0409 8.9304 8.8297 8.85Co
8.2949 8.1289 7.9862 8.0650 7.9629 7.7227 7.8224 7.9328 7.9051 8.0139
8.2631 8.3022 8.2167 8.3157 8.3016 8.3960 8.4573 8.5039 8.5906 8.7478
9.2310 9.0730 9.1113 9.1996 9.0385 8.9105 8.9246 8.8222 8.9519 8.9267
8.8361 8.7662 8.6929 8.5223 8.4586 8.5294 8.5935 8.6568 8.8191 8.8065
8.9493 8.9578 8.9019 9.0978 9.2696 9.1017 9.1119 9.1595 9.2157 9.4515
9.4935 9.4592 9.4052 9.2976 9.1394 9.0721 8.9521 9.0115 9.1006 9.1526
8.8857 8.8619 8.8320 8.7562 8.6224 8.6220 8.6566 8.7087 9.7572 9.2416

Nov.

Dec.

$7.8671 $7.8383
8.0015 8.0579
8.2097 8.3014
8.7509 8.9023
8.7468 8.5246
8.0674 8.2133
8.9635 9.1262
8.8841 8.7844
8.8922 8.9824
9.4781 9.5462
9.2252 9.2290
8.8620 9.0354

INDEX NUMBERS OF DUN.
PUBLICATION.

An “ index” 1 number based on the wholesale prices of a large num­
ber of representative commodities in general use in the United States
is published by the mercantile agency of R. G. Dun & Co., of New
York City. The information appears monthly in Dun’s Review, the
weekly journal of finance and trade issued by the above-named
company.
i Not an index in the true sense of the word, but a statement in dollars and cents of the per capita cost
of a year's supply of certain commodities at each date named.




INDEX NUMBEBS---- UNITED STATES.

149

HISTORY.

The publication of this index number was begun in 1901 and
covered a period of time extending back to 1860. From 1901 to
1907 periodical presentation of the index in Dun’s Review appears
to have been made. With the issue of May 11, 1907, however, its
publication ,was discontinued and apparently was not resumed until
May 9, 1914. The issue of the latter date contained data for the first
five months of the years 1912, 1913, and 1914, respectively, but no
attempt was made in this number to supply figures for all of the period
intervening since 1907. Data for other months of 1912, 1913, and
1914 are shown in subsequent issues; and in Dun’s Review of January
9, 1915, a presentation is made of the index number on the first
of each month for the entire period from 1907 to 1914, inclusive,
thus furnishing a continuous series since the inception of the under­
taking.
SOURCE OF QUOTATIONS.

The price quotations on which the index number is based are
those gathered by Dun & Co. in the principal markets of the country,
New York and Chicago prices predominating.
BASE PERIOD.

Under the method followed in the computation of the index num­
ber no base period is employed, the index in the case of each article
and group being the actual amount in dollars and cents required to
purchase a year’s supply for a single individual at the date named.
PRICES: H O W SHO W N AND COMPUTED.

With regard to the method of calculation, the following statement
is reproduced from Dun’s Review of May 9, 1914:
Quotations of all the necessaries of life are taken and in each case
the price is multiplied by the annual per capita consumption, which
precludes any one commodity having more than its proper weight in
the aggregate. Thus, wide fluctuations in the price of an article little
used do not materially affect the “ index,” but changes in the great
staples have a large influence in advancing or depressing the total.
* * *
per capita consumption used to multiply each of many
hundreds of commodities does not change. There appears to be
much confusion on this point, but it should be seen at a glance that
there would be no accurate record of the course of prices if the ratio
of consumption changed. It was possible, however, to obtain figures
sufficiently accurate to give each commodity its proper importance
in the compilation. This was done by taking averages for a period
of years when business conditions were normal and every available
trade record was utilized, in addition to official statistics of agricul­
ture, foreign commerce, and census returns of manufactures.




150

BULLETIN OF THE BUREAU OF LABOR STATISTICS.
NUMBER AND CLASS OF COMMODITIES.

The following excerpt from the same source shows what com­
modities are included:
For convenience of comparison and economy of space the prices
are grouped into seven classes: Breadstuffs include quotations of
wheat, corn, oats, rye, barley, beans, and peas; meats include live
hogs, beef, sheep, and many provisions, lara, tallow, etc.; dairy and
garden products embrace eggs, vegetables, fruits, milk, butter, cheese,
etc.; other foods include nsh, liquors, condiments, sugar, rice, also
tobacco, etc.; clothing covers the raw material of each industry, as
well as quotations for woolen, cotton, silk, and rubber goods, also
hides, leather and boots and shoes; metals include various quota­
tions for pig iron and partially manufactured and finished products,
as well as the minor metals, tin, lead, copper, etc., and coal and petro­
leum; miscellaneous includes many grades of hard and soft lumber,
lath, brick, lime, glass, turpentine, hemp, linseed oil, paints, fertili­
zers, and drugs.
The precise number of articles included in the index is not stated;
but in Dun’s Review of January 9, 1915, it is said that “ about 200
products are taken.”
DESCRIPTION AND GROUPING OF COMMODITIES.

As previously stated, the commodities are divided into seven
groups: viz, breadstuffs, meats, dairy and garden products, other
foods, clothing, metals, and miscellaneous articles. No further
description of the articles entering into the index is given.
SUBSTITUTIONS AND ADDITIONS.

Additions to the list of commodities for which index numbers
have been computed, or substitutions of a particular grade or quality
of an article for another grade or quality of the same article, if any,
are not shown in connection with any of the published data.
INTERPOLATION.

So far as can be determined from the information at hand concern­
ing the long period covered, no interpolation of prices has been made.
WEIGHTING.

As stated in a preceding paragraph, weighting is accomplished by
multiplying the price of each commodity, at the date named, by its
annual per capita consumption “ for a period of years when business
conditions were normal,” 1 as nearly as could be ascertained by
reference to reliable statistical records. It is stated in Dun’s Review
of January 9, 1915, that “ while it is obvious that the consumption of
some commodities has increased during recent years, it would defeat
the purpose of the index to change the multiplier in any instance,




Dun’s Review, May 9,1914, p. 23.

151

INDEX NUMBERS---- UNITED STATES.

because there would no longer be a comparative record of the cost of
the same quantities of the same articles back to I860, as is now the
case.” The issue of September 7, 1901, states that “ while the figures
can not be considered exact, the approximation is sufficiently close
to attain the desired result, and the ratio being constant the com­
parison with different dates shows to a cent the rise or fall in the cost
of living.” 1
TESTING.

No comparison of the index number with those compiled by
others has been made, nor have other means of testing been employed
so far as can be determined.
TABLE OF RESULTS.

The following statistics, showing the trend of wholesale prices
from January 1, 1860, to December 1, 1914, have been compiled from
Dun’s Review of May 11, 1907, and January 9, 1915:
W HOLESALE PRICES OF SPECIFIED COMMODITIES, JAN. 1, 1860, TO DEC. 1, 1914.

Date.

Breadstuffs.

Meats.

Dairy
and
garden
products.

Other
foods.

Clothing.

Metals.

Miscella­
neous.

1860, Jan. 1.................
1864, Sept.. 1.................
1870, Jan. 1.................
1875, Jan. 1.................
1880, Jan. 1.................
1885, Jan. 1.................
1888, Jan. 1.................
1889, Jan. 1.................
1890, Jan. 1.................
1891, Jan. 1.................
1892, Jan. 1.................
1893, Jan. 1.................
1894, Jan. 1.................
1895, Jan. 1.................
1896, Jan. 1.................
1897, Jan. 1.................
1897, July 1 (low)___
1898, Jan. 1.................
1899, Jan. 1.................
1900, Jan. 1.................
1901, Jan. 1.................
1902, Jan. 1.................
1903, Jan. 1.................
1904, Jan. 1.................
1905, Jan. 1.................
1906, Jan. 1.................
1907, Jan. 1.................
Feb. 1.................
Mar. 1.................
Apr. 1.................
•May 1.................
June 1.................
July 1.................
Aug. 1.................
Sept. 1.................
Oct. 1.................
Nov. 1.................
Dec. 1.................
1908, Jan. 1.................
Feb. 1.................
Mar. 1.................
Apr. 1.................
May 1.................
June 1.................

$23.652
46.138
29.076
26.048
22.955
16.342
18.565
la 195
13.765
19.725
17.700
15.750
13.530
14.311
11.380
11. 729
10.587
13.511
13.816
13.254
14.486
20.002
17.104
17.102
18.278
16.554
16.079
16.389
17.478
16.982
18.165
20.089
20.306
19.872
22.483
22.940
21.987
21.290
22.254
21.120
21.480
22.032
22.882
23.163

$10,084
17.789
15.255
11.932
9.206
9.432
8.920
8.705
7.620
7.810
7.895
9.315
8.655
8.359
7.540
7.327
7.529
7.336
7.520
7.258
8.407
9.670
9.522
8.138
7.950
8.426
9.350
9.693
9.673
9.629
9.641
9.982
10.196
10.090
10.150
9.667
9.229
8.929
8.146
8.246
8.546
9.221
9.777
9.620

$14.169
29.426
21.178
17.832
14.007
14.304
15.030
14.670
12.675
16.270
13.180
15.290
13.945
12.196
10.969
10.456
8.714
12.371
11.458
13.702
15.556
15.248
14.613
15.287
13.948
14.399
14.965
14.411
15.727
14.792
14.461
15.417
14.767
15.458
15.019
15.646
15.840
17.169
17.380
15.643
15.904
14.369
14.303
13.114

$8,978
29.562
16.240
14.546
11.873
8.996
10.340
10.480
9.935
10.215
9.185
9.595
8.945
8.607
8.898
8.170
7.887
8.312
9.096
9.200
9.504
8.952
9.418
9.653
10.699
9.822
9.760
9.804
9.767
9.817
9.824
10.100
10.013
10.041
10.180
10.446
9.629
10.152
10.236
10.384
10.354
10.501
10.397
10.314

$22,094
91.667
32.986
25.718
22.673
18.081
15.140
15.170
14.845
14.135
13.430
13.900
12.880
11.886
12.787
12.407
13.808
14.654
14.150
17.484
16.024
15.547
15.938
17.316
16.319
19.313
19.637
19.798
20.000
19.997
20.098
20.252
20.355
20.281
20.529
20.169
19.933
19.389
18.849
18.313
17.731
17.200
16.804
16.919

$26,082
61.964
27.682
22.833
25.002
15.065
17.330
17.360
16.240
15.875
14.665
15.985
14.565
12.026
12.803
13.014
11.642
11.572
11.843
18.085
15.810
15.375
17.185
15.887
16.188
17.141
18.087
18.162
18.135
17.372
17.524
17.689
17.688
17.667
17.626
17.296
17.179
16.937
17.232
16.944
17.122
17.176
16.872
16.659

$16,572
36.191
23.056
18.669
16.963
14.245
14.577
14.496
15.Ill
14.217
23.767
14.320
13.512
13.607
13.403
12.399
12.288
12.184
12.540
16.312
15.881
16.793
16.576
16. 759
16.936
18.809
19.386
19.109
19.133
19.305
19.242
20.125
20.335
20.319
20.086
19.976
19.836
19.406
19.185
19.264
19.252
18.229
19.150
18.198

Total.

$121.631
312.737
165.473
137.578
122.679
96.465
99.902
99.076
90.191
98.247
89.822
94.155
86.032
80.992
77.780
75.502
72.455
79.940
80.423
95.295
95.668
101.587
100.356
100.142
100.318
104.464
107.264
107.366
109.913
107.895
108.955
113.654
113.660
113.728
116.073
116.140
113.633
113.272
113.282
109.910
110.389
108.728
110.185
107.987

1 The issue of May 9,1914, contains the statement that “ Dun’s index number does not propose to show
the cost of living, because wholesale prices are taken and all luxuries omitted. Its economic value lies in
showing the percentage of advance or decline from month to month.”




152

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

WHOLESALE PRICES OF SPECIFIED COMMODITIES, JAN. 1, 1860, TO DEC. 1, 1914—
Concluded.

Date.




Breadstuffs.

Meats.

$22.826
24.161
24.176
23.990
23.579
21.879
21.480
22.900
23.967
24.129
25.696
26. 781
25.854
23. 705
22.002
21.530
21.638
22.315
23.830
23.509
23.423
22.172
20.992
20.590
21.690
21.863
20.263
19.120
18.830
18.567
18.010
18.175
17.762
18.176
19.973
20.508
21.283
21.695
22.145
23.828
24.864
23.125
23.523
24.278
24.718
25.590
27.637
27.391
25.964
25.760
24.088
21. 765
22.371
20.665
19.883
19.565
19.596
19.966
20.673
21.277
21.192
21.632
22.975
22.586
22.610
23.006
21.961
20.962
22.146
21.402
21.544
23.162
21.086
22.567
26.253
24.441
25.300
24.426

$10.197
9.992
9.488
9.534
9.175
9.135
9.142
10.277
8.860
9.247
9.022
9.498
9.955
9.617
9.540
9.450
9.351
9.546
9.642
9.683
10.786
12.359
11.542
11.692
11.406
11.080
11.029
10.370
9.897
9.788
9.483
9.963
10.146
9.742
9.363
9.638
9.414
9.900
10.080
9.612
9.218
8.924
8.920
9.173
9.514
10.59D
11.283
11.016
10. 715
10.848
11.186
10.923
10.457
10.629
10.912
11.522
13.047
13.478
13.183
12.963
13.090
13.080
12.786
13.053
12.211
12.059
12.150
12.625
13.168
12.868
12.813
13.068
12.979
13.427
12.839
12.093
11.907
11.324

Dairy
and
garden
products.
$12,552
13.357
13.924
14.620
15.016
17.019
18.104
15.645
15.212
16.142
15.705
16.053
15.268
15.767
16.014
16.265
17.508
19.164
18.906
17.564
16.927
15.237
14.321
14.325
14.663
15.457
15.738
16.234
16.810
18.013
18.073
16.468
14.588
13.634
14.759
14.701
17.473
19.248
18.001
16.501
19.190
22.177
21.286
21.898
19.364
21.774
20.776
18.087
15.501
16. 752
16.491
18.627
19.416
19.223
17.925
16.651
16.142
15.319
15.112
16.525
13.039
14.916
16.604
17.934
19.978
20.454
20.087
18.056
16.009
15.872
16.437
16.114
17.244
16.201
17.432
17.326
18.586
19.825

Other
foods.

$10.465
10.349
10.090
10.090
10.314
10.428
10.396
10.506
10.417
10.680
10.620
10.650
10.628
10.810
10.740
10.975
11.073
11.052
10.803
10.810
10.906
10.778
10.515
10.549
10.556
10.830
11.037
11.038
10.866
10.509
11.196
11.258
11.018
11.078
11.283
10.981
11.384
11.604
12.055
12.339
12.597
12.610
12.261
12.237
12.222
12.323
11.753
11.976
11.828
11.705
11.590
11.757
11.103
11.112
11.073
10.877
10.732
10.165
10.120
10.250
10.213
10.267
10.571
10.700
11.068
11.010
10.950
11.002
11.361
10.684
10.467
10.610
10.449
10.284
11.729
11.423
10.880
10.548

Clothing.

$17,233
17.348
17.325
17.226
17.308
17.828
18.024
18.277
18.893
18.633
19.078
19.587
20.062
20.924
21.061
21.528
22.145
22.130
20.635
21.671
21.785
22.061
22.194
21.281
21.173
20.508
20.556
19.932
19.896
20.042
19.644
19.596
19.789
19.355
20.021
18.845
19.324
18.778
18.509
18.638
18.191
18.191
18.630
19.048
19.493
19.868
19.979
20.003
20.449
20.588
20.703
20.705
20.789
21.066
21.015
20.835
21.143
20.938
20.807
20.705
20.534
20.250
20.507
20.947
21.074
20.815
20.664
20.241
20.434
20.641
19.969
20.686
20.834
20.975
20.398
20.259
19.970
19.883

Metals.

$16,542
16.537
16.720
16.821
16.788
16.920
16.919
16.935
16.652
16.388
16.353
16.453
16.426
16.615
16.948
17.200
17.304
17.437
17.496
17.419
17.265
17.132
16.937
16.894
16.744
16.587
16.652
16.574
16.144
16.092
16.519
16.591
16.742
16.718
16.694
16.617
16.583
16.526
16.502
16.307
16.294
16.361
16.371
16.356
15.961
15.550
15.918
16.104
16.349
16.664
17.022
17.633
18.029
18.046
17.942
17.850
17.379
16.924
16.753
16.760
16.512
16.528
16.742
16.760
16.758
16.596
16.170
16.185
15.881
15.784
15.559
15.695
15.691
15.764
16.126
15.974
15.849
16.134

Miscella­
neous.

$18,359
17.751
17.608
17.710
17.734
17.781
17.783
18.914
21.419
21.635
21.789
22.003
20.828
20.582
20.656
21.362
21.751
21.770
22.122
21.743
21.748
21.816
21.806
21.910
22.936
22.171
22.156
22.181
22.180
21.653
22.177
22.201
22.243
22.225
22.166
22.083
22.669
22.024
22.040
22.067
2L 616
21.534
22.437
22.435
22.255
22.354
21.640
21.411
21.471
21.575
21.465
21.696
21.360
21.313
22.082
22.428
22.422
22.427
21.676
21.570
21.739
21.842
21.868
21.922
21.804
21.794
22.546
22.570
22.772
22.540
21.441
21.761
21.425
21.522
22.198
22.015
21.848
22.043

Total.

$108.174
109.495
109.331
109.991
109.914
111. 008
111. 848
113.454
115.420
116.864
118.263
121.025
119.021
118.020
116.961
118.301
120.770
123.414
123.434
122.399
122.840
121.555
118.307
117.241
119.168
118.524
117.431
115.449
114.623
114.664
115.102
114.252
112.288
110.928
114.259
113.373
118.130
119. 775
119.332
119.292
121.970
122.922
123.438
125.425
123.527
128.049
128.986
125.988
122.277
123.892
122.545
123.106
123.525
122.054
120.832
119.728
120.461
119.217
118.324
120.050
116.3J9
118.515
122.053
123.902
125.503
125.734
124.528
121.641
121.771
119.791
118.230
121.096
119.708
120.740
126.975
123.531
124.340
124.183

INDEX NUMBERS---- UNITED STATES.

153

INDEX NUMBERS OF GIBSON.
PUBLICATION.

This index of wholesale prices in the United States is published
by Thomas Gibson, New York, every Saturday, in his weekly market
letter.
HISTORY.

In March, 1910, Prof. J. Pease Norton published a “ report on a
new method of compiling index numbers on the Sauerbeck selection
of commodities modified with the Dun system of weighting,” which
was prepared for use in the weekly market report of Thomas Gibson.1
The work was undertaken as a continuation of the Dun index,
which had been suspended in May, 1907.
In this compilation 50 articles, divided into four general groups,
were used instead of the much larger number included in Dun's index.
The general food group was in turn divided into vegetable foods
and animal foods. The descriptions of the 50 articles whose prices
formed the index were the same as those used for these 50 articles
in Bulletin of the United States Bureau of Labor, No. 75. The
actual and relative prices for 1907 of these 50 articles appear to
have been taken from the latter source. The plan followed in the
compilation of this index was intended to be that used by Sauerbeck.
It is claimed that no manufactured or derivative products are
included, but that only primary commodities have been used.
Since November, 1912, only 22 articles, all of which belong to the
food group alone, have been included in the index number.
SOURCE OF QUOTATIONS.

As previously stated, the quotations used to join this index number
with the one compiled by Dun were those published for January,
1907, in Bulletin of the United States Bureau of Labor, No. 75.
The source of later quotations is not given.2
BASE PERIOD.

The years 1890 to 1899 are used as the base period in the compu­
tation of the index number.
PRICES: HOW SHOWN AND COMPUTED.

The actual prices of the articles are not shown for any period, the
only data published in Gibson's weekly market report being the
index for all commodities.
1 See also article by Prof. Norton in Quarterly Journal of Economics, August, 1910, pp. 750-759. Pub­
lished by Harvard University, Cambridge, Mass.
2 It is stated, however, in the Quarterly Journal of Economics, August, 1910, p. 758 (footnote) that “ Sta­
tistics collected from trade journals were used from January, 1909, to compute relative prices.”




154

BULLETIN OF THE BUREAU OF LABOR STATISTICS.
NUMBER AND CLASS OF COMMODITIES.

As has been stated, when this index was first published it covered
50 articles from the farm, mines, and other sources, and included
such as had been subjected only to an initial manufacturing process.
Since November, 1912, it has been calculated on the food group
alone, including 22 articles. It is stated that the articles covered
are those essentially primary in their nature.
DESCRIPTION AND GROUPING OF COMMODITIES.

The present list of articles is divided into two groups, as follows:
Wheat, contract price.
Wheat flour, spring patents.
Wheat flour, winter patents.
Barley, by sample.
Oats, cash.
Corn, No. 2, cash.
Corn meal, fine yellow.
Potatoes, white.
Rye, No. 2.
Sugar, 89°, fair refining.
Sugar, 96°, centrifugal.
Coffee, Rio, No. 7.
Tea, Formosa, fine.

Vegetable foods {13 articles).

Animal foods (.9 articles).

Beef, steers (average of quotations for two grades).
Beef, fresh native sides.
Beef, salt.
Mutton, sheep (average of quotations for two grades).
Mutton, dressed.
Pork, hogs (average of quotations for two grades).
Bacon, short rib sides.
Hams.
Butter (average of quotations for three grades).
SUBSTITUTIONS AND ADDITIONS.

Since the adoption of the present list there have been no additions
of new articles nor substitutions in the place of those carried, so far
as can be ascertained from the material published.
INTERPOLATION.

Methods of supplying lacking statistical data, if resorted to, are
not disclosed.
WEIGHTING.

The weights assigned to the four groups formerly included in the
index number were 50 for foods, 18 for textiles, 16 for minerals, and
16 for other commodities.
The weighting was accomplished by using a combination of
figures from Dun’s report and the material published by the



INDEX NUMBERS---- UNITED STATES.

155

United States Bureau of Labor. The first operation was to secure
an average of Dun’s general index numbers for the years 1890 to
1899, which was found to be 0.843. The sum of the relative prices
for all the articles in a group as published for January, 1907, in
Bulletin No. 75 of the Bureau of Labor was then found. These
relative prices were based upon the average for 1890 to 1899 as
100. The sum of these relatives was then divided by the average
of the Dun number, 0.843. The quotient thus obtained was termed
a multiplier. The total of the relatives of a group was then multi­
plied by this multiplier. The result divided by 100 was the index
for that group, and the sum of the indexes for the four groups was the
general index number.
The following statement explains the process of calculating the
index by the above method:
ILLU STRATIVE EXAM PLE OF CALCULATING GIBSON’S IN D E X NUMBERS.

Commodity.

Total
relatives
of the
groups.

Multi­
pliers.

Weighted
product.

Foods........................
Textiles....................
Minerals....................
Other........................

2422.1
1264.0
1324.8
1408.7

1.9159
1.6860
1.4987
1.3488

46.40"0
21.3010
19.8548
19.0005

Index number

106.5613

Since the reduction of the number of articles on which the index
number is calculated from 50 articles of all classes to 22 food com­
modities, no explanation has been given concerning the method of
weighting employed, so far as can be ascertained. It is stated, how­
ever, in Gibson’s weekly market letter of January 11, 1913, and in
subsequent issues that the index number is weighted according to
Dun’s method.
TESTING.

The compiler of Gibson’s index compares the result obtained under
his method, 106.5613, with 107.2640, Dun’s number for the same
period. As a further test to show that figures compiled by this
method would take the course of Dun’s index number, the following
figures are shown:




Date.

January, 1907..........................
February, 1907.......................
March, 1907.............................
April, 1907..............................
Mav, 1907................................
1896..........................................

Gibson
numbers.

Dun
numbers.

106. f 6
108.01
109.38
110.56
113.41
72.22

107.26
107.37
109.91
(107.90)
(109.00)
74.32

156

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

The compiler expresses the opinion that in April and May of 1907
the comparison of the two index numbers can not justly be made,
inasmuch as it appears probable the Dun calculator changed his
system of weighting for those months.
TABLES OF RESULTS.

The average yearly index numbers for the cost of foodstuffs, the
only part of the original series now published, as computed by this
process from 1890 down to the present time, are shown in the follow­
ing table appearing in Gibson’s weekly market letter of December
31, 1914:
AVERAGE Y E A R L Y IN D E X NUMBERS, 1890 TO 1914.

Average
yearly
index
number.

Year.

1

Average
yearly j
index
number.

Year.

j

1890.............................
1891.............................
1892.............................
1893..........................
1894
. .

43.4
50.8
45.3
4P>. 0
43.4

1900...........................
1901...........................
1902...........................
1903...........................
1904...........................

44.2
44.5
53.5
49.0
48.3

1895.............................
1898.............................
1897..........................
1898..........................
1899..........................

42.0
34.0
34. 7
38. 7
41.6

1906...........................
1907...........................
1908...........................
1909.........................

47.3
49.8
50.9
54.2
59.2

Year.

1910...........................
1911...........................
1912...........................
1913...........................
1914...........................

Average
yearly
index
number.
59.3
56.9
62.6
58.1
60. 8

|

Monthly averages for 1913 and 1914, also shown in the publication
referred to above, are as follows:
MONTHLY AVERAGES, 1913 AND 1914.

1913

ji

1914

'l

i:
Month.

Monthly j
average. |

January............
Februarv.........
March.............
\pril
May. . .
June.................




55.5
57.0
57.8
59.0
57.8
57.3

Month.
July..............
August..........
September. . .
October........
November__
December___

Monthly
average.
58. 6
59.3
60.0
58.4
58.4
58.2

Month.

January........ !
February___
M arch..!.......
April.............
Mav..............
June..............

Monthly
average.
58.2
58. 2
57. 8
57. 7
57.9
59.4

Month.

July..............
August..........
September. . .
October........
November...
December___

Monthly
average.
58.9
64.9
68.6
62.9
63.1
62.3

INDEX NUMBERS---- AUSTRALIA.

157

AUSTRALIA.
INDEX NUMBERS OF THE COMMONWEALTH BUREAU OF CENSUS AND
STATISTICS.
PUBLICATION.

In December, 1912, a report entitled “ Prices, Price Indexes, and
Cost of Living in Australia/ ’ compiled by G. H. Knibbs, Common­
wealth statistician, was published by the recently organized Labor
and Industrial Branch of the Commonwealth Bureau of Census and
Statistics at Melbourne as its Report No. 1. Both wholesale and
retail prices, together with import and export index numbers for
Australia, are among the subjects considered in the report.
This publication is the first of a proposed series designed to include
topics covering general industrial conditions as well as prices. Its
main object, as stated in the preface, is “ to furnish information
as to prices in past years in such a form as to be fully comparable
with that which it is proposed to publish periodically in the future.”
The continuation of these index numbers is to be found in Report
No. 2, entitled “ Trade Unionism, Unemployment, Wages, Prices,
and Cost of Living in Australia, 1891 to 1912,” under date of April,
1913.
HISTORY.

The author of these reports, having studied the various systems
of index numbers published in other countries, became convinced
that the methods ordinarily followed were so defective as to be mis­
leading. He believed that an accurate system of inquiry should be
determined upon to secure reliable and satisfactory data on which
to base index numbers, and further that a uniform method for the
international study of prices as a basis for world index numbers
should likewise be worked out by those economists interested in the
subject. These conditions he undertook to meet.
SOURCE OF QUOTATIONS.

At first an attempt was made to secure from market reports whole­
sale prices of a representative list of commodities for the capital
towns of each State. This plan was abandoned when it was found
to be impracticable owing to the lack of complete records and to the
difficulty in obtaining comparable returns. Moreover, the compila­
tion of figures for Melbourne alone involved so much labor that no
attempt was made to complete any other city.
The figures for Melbourne were obtained mainly from market prices
published in the ordinary press and in special trade reviews. Where
there was any question as to the reliability of the quotations the
figures were verified by “ reference to well-known and important
business firms dealing in the articles in question.”




158

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

The prices quoted are taken from the following sources: Journal
of Commerce, 1861, 1866, 1871, 1872, and 1883 to 1912; Australasian
Trade Review, 1871 to 1882 and occasionally to 1892; Melbourne
papers, 1873 to 1912; for meats, Gippsland Mercury, 1890 to 1892;
for coal, Federated Steamship Owners, 1901 to 1912.
BASE PERIOD.

The basic period for the computations of the wholesale price indexes
was the year 1911, the aggregate expenditure on all articles and on
each group of articles in this year being taken as 1,000. To quote:
“ The index numbers show the amount which would have had to be
expended in each of the years specified in order to purchase what
would have cost £1,000 in 1911, distributed in purchasing the rela­
tive quantities (indicated by the mass units) of the several commodi­
ties included in each group and in all groups respectively.”
PRICES: HOW SHOWN AND COMPUTED.

A table in the appendix to the first report shows the average annual
wholesale prices in Melbourne of all commodities included in the
investigation except meats from 1871 to 1912 (first 9 months only),
inclusive. Complete 1912 data for all of the 80 commodities, except
silk, and for 13 additional commodities, are contained, in the appendix
to the second report.
Prices for meat were not obtained for the years prior to 1884 and
1885 nor for the years 1886 to 1889, inclusive. The unit of meas­
urement is given for each article and the price stated in shillings
and pence. The articles for which prices are quoted are divided
into eight groups. In most cases monthly prices were obtained, and
from these the yearly averages were computed. For tea, coal, cotton,
wool, and silk, however, monthly prices were not available, so that
yearly averages based in each case on expert opinion 1 were secured.
The monthly quotations, from which the yearly average is. com­
puted, are not shown.
The group and general index figures for 1861 and 1866 are shown
elsewhere in the report, but no actual prices are given for these years.
NUMBER AND CLASS OF COMMODITIES.

In the computation of index numbers for the years prior to 1911
(the base year) the aggregate expenditure on 80 commodities was
used, while for 1912 the number of commodities included was
increased to 92. The author states that the commodities included
are generally in the nature of raw materials—that is, materials in
which the labor cost is relatively low.




i Source of this expert opinion not stated.

INDEX NUMBERS---- AUSTRALIA.

159

There are no articles of clothing, boots or shoes, or furniture
included. The reason assigned for their omission was the imprac­
ticability of obtaining periodic prices for predominant grades and
qualities and of satisfactorily determining the relative importance in
consumption of the various items, the author contending that the
character of clothing and of furniture includes the element of change
due to the influence of fashion, and that where incomes are limited
economy strikes first at these articles.
DESCRIPTION AND GROUPING OF COMMODITIES.

The 80 commodities used in computing the index numbers for
years prior to 1911 are divided into 8 groups as follows:
I.
II.
III.
IV.
Y.
VI.
VII.
VIII.

Metals and coal, 12 commodities.
Jute, leather, etc., 9 commodities.
Agricultural produce, 13 commodities.
Dairy produce, 7 commodities.
Groceries, 21 commodities.
Meat, 5 commodities.
Building materials, 9 commodities.
Chemicals, 4 commodities.

The list of articles, with their description or brand, the source of
information, the unit upon which quoted, the quantities consumed,
and the “ mass unit” —i. e., the extent to which a commodity is
used—are shown in detail on pages 162 to 164 of this bulletin.
In both reports index numbers are given for all groups taken as a
whole and for each group. Under each group in the first report is
shown the index number for a few individual articles of importance
computed on the price in 1911 as the base, but no table is given
showing an index for each of the 80 articles separately.
SUBSTITUTIONS AND ADDITIONS.

Cases of substitution of a particular grade or quality of an article
for another grade or quality of the same article, if any, are not
apparent in the tables, owing to the manner in which the information
is presented. The author states, however, that “ every care was
taken to insure that the prices quoted for each article refer to a uni­
form quality” and that “ special precautions were taken to insure
substantial continuity of quality or grade.”
In the computation of the index numbers for 1912 in the second
report, as previously stated, the author added 13 articles and dropped
raw silk, so that the index for 1912 covers 92 articles instead of 80,
as formerly. The aggregate expenditure on these 92 articles in 1911
formed the base for the 1912 index. In group 3 the mass unit for
hay was changed from 270 to 135, and oaten chaff, a new article in this




160

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

group, was assigned a mass unit of 135, thus making the sum of the
mass units used for hay and oaten chaff in 1912 equal the mass unit
for hay in the earlier years. The 13 articles added are as follows:
GROUPS OF ARTICLES ADDED IN THE COMPUTATION OF THE IN D E X NUMBERS
FOR 1912.

Unit,

Commodity.

Group I:
Tinned plates, I. C. coke............
Quicksilver..................................
Grom> II:
Twine, reaper and binder...........
Tallow, prime mutton................
Group III:
Chaff, good oaten........................
Malt, Victorian............................
Onions..........................................
Group IV:
Beeswax......................................
Condensed milk, Bacchus Marsh
Group VII:
Slates, Welsh, 20 by 10................
Group VIII:
Caustic soda.................................
Alum, lump.................................
Cyanide potassium......................

Mass
unit.

Hundredweight
Pound..............

60

Pound..............
Ton...................

150

Ton.................. .
Bushel..............
Ton.................. .

135
140
3

Pound...............
Dozen pounds...

40
160

12

1,000....................

Hundredweight.
Hundredweight.
Pound...............

N ote.—Groups V and VI have no additions.

INTERPOLATION.

As already stated, prices for meats were not secured prior to 1890
except for 1884 and 1885. For the full period since 1871 the index
numbers were accordingly worked out for the seven groups, excluding
meats, and also for the period since 1890 for the eight groups, including
meats. The figures for the general index for 1871 to 1889 (except
1884 and 1885) were then adjusted, on the basis of the results of
succeeding years, so as to include meats. The exact procedure has
not been disclosed.
WEIGHTING.

The system of weighting used differs materially from the system
generally employed by compilers of index numbers. The author
bases his index numbers on what he terms the aggregate expendi­
ture method. By this method the cost of an unvarying bill of
goods is calculated at the varying prices prevailing during differ­
ent years. The extent to which a commodity is used is expressed
by a number termed the “ mass unit.” The mass unit is devel­
oped from the figures which denote the quantity used or consumed,
which latter amount has in general been obtained by adding
to the production of each commodity in Australia the amount
of imports and from this sum subtracting the amount of exports.
The figures have, in general, been based on the average production
and the average export and import returns for the five years 1906 to
1910, inclusive, No further explanation as to the source of his con­




161

INDEX NUMBERS---- AUSTRALIA.

sumption figures has been given by the author. Reference to the
table on pages 162-164 of this Bulletin will indicate the manner in which
the mass units were obtained. In this table the quantities consumed
of the various commodities are expressed in thousands only. The
mass unit is obtained in each case by dividing the figure denoting
the quantity consumed by 10 and approximating the quotient. For
instance, the average annual consumption of pig iron is stated to be
64 thousand tons, which is restated as a mass unit of 6J.
The mass unit having been established and the average yearly
price for the year determined, the process then was to multiply the
mass unit by the price. Thus, the price of pig iron was 81s. 2d. ($19.48)
for a ton in 1911. This price, 81s. 2d. ($19.48), multiplied by the mass
unit (6J) gives the aggregate expenditure on pig iron in 1911. This
process was applied to each article of the metals group in 1911 and
each of the years during the entire period (1890-1912), the sum of
such products producing the yearly aggregate group expenditure.
The same mass unit was used for all the years of the period. The
index for a single group and for all groups was obtained by dividing
the total expenditures of a given year by the total expenditures of
the basic year, i. e., 1911, and the result multiplied by 1,000.
The following table illustrates the system of weighted index num­
bers used:
COMPUTATION OF IN DEX NUMBERS—ILLUSTRATIVE EXAM PLE OF AGGREGATE
EXPENDITURES METHOD.

Articles.

Butter.......................................
Bread........................................
Mutton.....................................
Milk..........................................

Unit.

Pounds..........
2-pound lo a f..
Pounds..........
Quarts............

Total..............................

Quantities
consumed
(in mil­
lions).

90
470
330
300

Total expenditures
(in millions).

Prices.
1901

1911

a.

d.

15
3
3
4

1901
d.

18
4
5
5

1,350
1,410
990
1,200
4,950

1911
d.

1,620
1,880
1,650
1,500
6,650

Thus 6,650 millions is the total expenditure for this group in 1911,
which is the base or 1,000. To secure the index figure for 1901, the
total aggregate number 4,950 millions for 1901 is divided by 6,650
millions, the base, which quotient multiplied by 1,000 equals 744 as
the index for 1901. A similar process was used for each of the
groups represented in the report. The author lays particular
emphasis on the fact that his index numbers are reversible, by
which he means that they may be recomputed with any other year
than 1911 as the base and the results be as accurate as if that year
had been taken as the base originally.
94261°—Bull. 173—15----- 11




162

BULLETIN OF TH E BUREAU OF LABOR STATISTICS.
TESTING.

The author tests his system of weighting by mass units instead of
actual quantities consumed by a comparison of results obtained under
the two methods by taking a list of prices from 1871 to 1911 for 73
articles, the year 1911 being used as a base. In the first instance the
actual figures were used and in the second the mass units. The
result in the first case was 1,194, and in the second 1,193, the slight
difference thus shown appearing not to warrant the extra arithmetical
labor required by the use of the actual figures instead of the rounded
numbers or mass units.
He further tests his index numbers by a comparison with index
numbers obtained by other methods. The following table illustrates
this comparison:
PRICE IN DEXES: EXAM INATION AS TO R E L IA B IL IT Y OF VARIOUS METHODS.
Aggregate
expendi­
ture
method.
Index number for 1871, with 1911 (equaling 1,000) as
base year.........................................................................

1,194

Weighted Weighted
Un­
price-ratio price-ratio
weighted
method
method
(geometric (arithmetic price-ratio
method.
mean).
mean).

1,195

1,289

1,310

He considers the first two as valid but the last two as invalid for
his use.
TABLES OF RESULTS.

The following table shows the list of 80 commodities divided into
8 groups, with the brand, source of information, unit, quantities con­
sumed, and mass unit in each case:
MELBOURNE WHOLESALE PRICES, COMMODITIES INCLUDED, SOURCES OF INFOR­
MATION, QUANTITIES CONSUMED, AND MASS UNITS.

Group I .— Metals and coal (12 commodities).

Commodity.

Iron:
Pig.........................
Rod and bar..........
Angle and T..........
Plate......................
H oop......................
Galvanized............
Fencing wire................
Zinc, sheet...................
Lead, sheet...................
Lead p ipes..................
Copper, sheet...............
Coal..............................

Total...................

Brand.

Sources of infor­
mation.

Unit.

Trade journals1. . . Ton....................
____d o___________!
rio
.......do.................... .......do.................
.......do.................... ___do
.......do.................... ....... do
.......
26-gauge
___ .......d o.................... .......do
No. 8
......... .......do.................... ___do
___do
.......do
...........
do
do
Pound
N e w c a s t le , on Journal of Com­ Ton....................
merce, 1871 and
wharf.
1872, quoted in
Bay, Melbourne
papers, 1873 to
1990, Federated
Steamship Own­
ers, 1901 to 1912.

Mixed numbers. . .
Stall'ord.................
.......do....................
.......do
.......

Quantities
consumed
(in thou­
sands).
64
34
34
31
6
52
60
8
7
6
20,000
6,004

Mass
unit.

11

?
54
6
1

2,000^
600

2,630*

1When ‘ ‘ trade journals” is mentioned it signifies the Journal of Commerce, 1861,1866, 1871, 1872, and
1883 to 1912, and the Australasian Trade Review, 1871 to 1882, and occasionally to 1892.




163

INDEX NUMBERS---- AUSTRALIA.

MELBOURNE W HOLESALE PRICES, COMMODITIES INCLUDED, SOURCES OF IN FOR­
MATION, QUANTITIES CONSUMED, AND “ MASS UNITS” —Continued.

Group I I . —Jute, leather, wool, etc. (9 commodities).

Commodity.

Brand.

Sources of infor­
mation.

Bran, bags...................
Trade journals1. . .
.......d o ...................
Com, sacks...................
....... d o....................
Woolpacks...................
....... d o....................
Leather, kip (small
beasts, as calf, sheep).
....... d o....................
Leather, calf
__
....... d o....................
Leather, basils (leather
tanned in bark as
distinguished from
“ roan,” tanned in
sumach).
Cotton.......................... Raw...................... London prices.......
Silk
........................ .......do.................... ....... d o....................
W ool............................. ....... do.................... L o n d o n prices
(A u s t r alasian
wool).

Unit.

Quantities
consumed
(in thou­
sands).

Mass
unit.

Dozen.................
.......do................
Each..................
Pound...............

1,090
2,500
2,000
10,710

110
250
200
1,070

.......d o ................
Dozen..
___

6,946
257

700
25

Pound...............
.......do.................
....... do................

243,200
2,635
122,000

24,000
250
12,200

38,805

Total...................

Group II I . —Grains, etc. (IS commodities).
Wheat..........................
Flour
...................
Bran
........................
Pollard.........................
Oats
.................
Oatmeal
.........

Barley..........................
D o..........................
Maize ..........................
Hay..............................
Straw
....................
Peas............. .*..............
Potatoes.......................

Trade journals1 Bushel...............
and Melbourne
papers.
Ton....................
....... d o.................... Bushel...............
____d o ___________ .......do.................
Feed____ _______ I____ do __________ .......do................
Colonial................. Trade journals1 Ton....................
and Melbourne
papers, 1891 to
date.
Malting................. .......do.................... Bushel...............
Feed...................... .......do.................... .......d o................
.......d o.................... .......do................
Best manger......... Melbourne papers. Ton....................
Victorian.............. .......d o .................. .......do
.......
Trade j ourn als! . . . Bushel...............
Melbourne papers. Ton....................

4,853

500

477
14.350
14.350
12,365
16

48
1.400
1.400
1,200
1h

1,500
1,000
9,624
2,695
265
554
402

150
100
1,000
270
25
55
40

Total...................

6,189§

Group I V .— Dairy produce (7 commodities).
Ham.............................
Bacon...........................
Cheese..........................
Butter.......................... Best fresh..............
Lard ........................... In bladders...........
Eggs............................. Ordinary...............
Honey..........................

Melbourne papers.
.......d o....................
.......d o....................
.......d o....................
.......do................
.......do....................
.......do....................

Pound...............
.......d o................
.......do.................
.......d o ...............
.......do
.......
Dozen................
Pound...............

8,000
32,500
15.000
95.000
2,000
18.000
5,847

Total...................

800
3,200
1.500
9.500
200
1,800
600
17,600

Group V.— Groceries (21 commodities).
Currants.......................
Trade journals1. . . Pound................
Raisins......................... Sultanas................ ....... d o.................... .......do
.......
Herrings....................... 1-pound tins......... .......d o.................... Dozen 1-pound
tins.
Salmon......................... .......do.................... ....... d o.................... .......d o ...............
Sardines....................... Halves...................
Dozen halves___

14.000
14.000
500

1.400
1.400
50

500
1,000

50
100

1 When “ trade journals ” is mentioned it signifies the Journal of Commerce, 1861,1866, 1871, 1872, and
1883 to 1912, and the Australasian Trade Review, 1871 to 1882, and occasionally to 1892.




164

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

MELBOURNE WHOLESALE PRICES, COMMODITIES INCLUDED, SOURCES OF INFOR­
MATION, QUANTITIES CONSUMED, AND u MASS U N ITS” — Concluded.

Group V .— Groceries (21 commodities)—Concluded.

Commodity.

Sources of infor- 1
rnation.
j

Brand.

Quantities
consumed
(in thou­
sands).

IT ..
Umt-

Coffee........................... Plantation............ Trade journals 1.. _ Pound ..............
Cocoa............................ Taylor’s ................ ....... do.................... .......do.................
Sugar............................ No. A1 or its .......do.................... Ton....................
equivalent in
former years.
i.......d o.................... Pound ............
Macaroni......................
.......d o.................... .......do
Sago..............................
___
Rice.............................. Patna.................... .......do.................... Ton....................
.......d o....................
Salt............................... Liverpool, fine
D o.......................... Rock..................... .......do....................
Mustard........................ 1871-1883 D. S. F.; .......do.................... Dozen 1-pound
in 1884-1911, 1tins.
p o u n d tins,
Coleman’s.
.......do.................... Pound...............
Starch........................... Coleman’s white
Blue
........................ Keen’s ................... .......do.................... .......d o................
Matches ...................... W ooden safety___ .......d o.................... Gross
............
.............. Pound ............
Candles
.............. Gouda................... .......do
Tobacco........................ Two Seas, in pock­ .......do.................... .......d o.................
et pieces.
Tea...............................
London prices....... .......d o................
Trade journals1. . . Gallon................
Kerosene......................

Mass
unit.

2,100
1,000
220

200
100
22

2,000
7,750
22
70
10
64

200
800
2
7
1
6

1,000
500
860
16,000
13.000

100
50
90
1,600
1,300

30.000
17,500

3,000
1,700

Total...................

12,178

Group V I.— Meat (5 commodities).
Beef.............................. Average quality.. . Metropolitan meat
market reports.2
Mutton......
.......... .......do.................. ......... d o....................
Veal.............................. .......do.................. ......... d o....................
Lamb
...................... j .......d o ................ .1.......do ................
Pork
......................i .......do.................. ......... do....................

100 pounds.........

3,875

390

Pound................
....... do.................
Each
Pound ............

332,000
20,000
2,047
37,000

33,000
2.000
200
3,700
39,290

T o t a l . . . ...............................1

!

!

Group V II.— Building materials (9 commodities).
Timber......................... Flooring:
6 by 1§............
6 i)y |..............
6 by |..............
6 b y £ .........
Weatherboards
Oregon..................

Trade journals1. ..
.......do....................
.......do....................
.......do
...........
.......do....................
.......do....................

100 feet, linear..
.......do.................
.......d o.................
.......do
___
.......d o.................
1,000 feet, super­
ficial.
Shelving................ .......do.................... .......d o.................
Cement......................... Portland............... i.......d o.................... Cask...................
.......do.................... Ton....................
White lead...................
Total...................

300
300
300
300
2,000
200

30
30
30
30
200
20

100
312
8

10
30
i

380f
i

Group V III.— Chemicals (4 commodities).
Cream of tartar............ In kegs..................
Carbonate of soda
................................
SaltmAtAr
_________ Refined_________
Sulphur ......................

i

Trade journals1. . .
.......do
..............
.......do ................
.......d o....................

4,030
3
1
2

Pound................
Ton....................
.......do.................
....... d o.................

400
4

t
400|

Total...................
1

1 When “ trade journals” is mentioned it signifies the Journal of Commerce, 1861, 1866, 1871, 1872, and
1883 to 1912, and the Australasian Trade Review, 1871 to 1882, and occasionally to 1892.
8 Gippsland Mercury, 1890 to 1892; Melbourne papers, 1893 to 1912.




165

INDEX NUMBERS— AUSTRALIA.

The next table contains the index numbers for each group and for
all the groups as a whole from 1871 to 1912, inclusive. It will be
noticed that no index numbers for meats are given before 1884 or
for the years 1886 to 1889.
MELBOURNE

WHOLE SALE-PRICE IN DEX NUMBERS, 1871 TO 1912, COMPUTED TO
Y EA R 1911 AS BASE.

III.
Agricul­
IV.
V.
tural
Dairy
produce, produce. Groceries.
etc.

I.
Metals
and coal.

II.
Jute,
leather,
etc.

1871 .........
1872 .........
1873
___
1874 .......
1875
___
1876
1877
1878
1879
1880 .......
1881 .......
1882
1883 .......
1884...........
1885...........
1886 .......
1887
1888
___
1889
1890...........
1891...........
1892...........
1893...........
1894...........
1895...........
1896...........
1897...........
1898...........
1899...........
1900...........
1901...........
1902...........
1903...........
1904...........
1905...........
1906...........
1907...........
1908...........
1909...........
1910...........

1096
1456
1816
1635
1487
1406
1400
1329
1266
1347
1178
1297
1231
1208
1216
1164
1053
1216
1061
1402
895
889
856
752
720
808
813
842
933
1042
1061
1007
923
821
772
882
1037
1033
1014
1004

1257
1394
1362
1240
1230
1146
1149
1094
1060
1101
1115
1032
1021
997
921
835
883
870
886
911
847
800
783
721
684
749
706
683
717
861
774
756
834
885
850
978
1017
901
907
1052

1236
1246
1422
1456
1361
1446
1347
1269
1298
954
1012
1444
1237
1124
1156
1222
1184
1123
1505
1022
1024
971
834
644
734
1116
1063
920
670
703
928
1192
1209
754
894
916
973
1312
1000
969

1000

1000

1000

1912...........

1021

991

1370

Year.

1911............

864
1019
1032
1160
1345
1415
1303
1112
1146
900
935
1347
1114
1156
1316
1286
1091
1210
1082
1099
995
1066
842
708
712
875
937
1034
814
838
1029
1215
1059
876
980
972
1020
1198
1119
1100

1000

1206

1586
1608
1581
1476
1435
1462
1502
1378
1371
1412
1421
1414
1408
1326
1158
1139
1128
1122
1152
1074
1032
997
1033
1057
1016
1021
1009
1000
1003
1039
1048
945
936
916
942
923
948
968
978
999
1000

1052

VI.
Meat.

1151
1042

1007
888
901
816
695
682
808
1072
1091
960
1168
1345
1447
1443
1427
1209
1110
1294
1335
1088
1008

VII.
VIII.
Building Chemi­
materials.
cals.

1044
1097
1446
1138
1009
1054
1047
886
852
943
1091
1005
910
876
880
<30
790
937
940
880
780
704
739
731
789
780
766
838
805
911
841
837
875
845
801
896
968
935
911
996

1409
1537
1661
1668
1554
1532
1569
1411
1444
1626
1587
1493
1484
1471
1432
1398
1401
1378
1228
1257
1194
1149
1018
934
1003
1065
971
933
892
908
917
881
921
875
859
864
961
891
815
898

1000

1000

1000

1357

1057

978

All com­
modities
together.

1229
1335
1451
1387
1337
1350
1311
1216
1210
1109
1121
1289
1183
1132
1105
1089
1055
1074
1171
1053
945
918
850
749
760
922
925
895
809
894
974
1051
1049
890
910
948
1021
1115
993
1003
1000

1170

The author presents the following table of index numbers by quin­
quennial periods to show the average level of prices over periods
of several years. The average for each 5-year period is 1000 and
that for 1911 and 1912 is compared with this base. For instance,
taking the average price level of 1871-1875 as 1000, that for 19111912 has fallen to 806, and taking that of 1876-1880 as 1000 that
for 1911-1912 in comparison is 877. The other figures are to be
read in the same manner.




166

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

IN D E X NUMBERS FOR 1911-1912, W ITH AVERAGE EXPEN DITU RE IN EACH SUCCESSIVE
QUINQUENNIAL PERIOD AS BASE.

Base period
(prices=
1000).

1871-1875..
1876-1880..
1881-1885..
1886-1890..
1891-1895..
1896-1900..
1901-1905..
1906-1910..

I.
Metals
and
coal.
672
746
821
854
1,225
1,134
1,098
1,013

II.
Jute,
leather,
etc.
792
926
1,012
1,172
1,340
1,383
1,224
1,059

III.
IV.
Agri­
V.
Dairy
cultural
Groceries.
produce.
produce.
841
895
946
934
1,345
1,265
1,137
1,094

1,037
957
957
974
1,299
1,249
1,089
1,039

678
731
775
928
1,015
1,028
1,089
1,082

VI.
Meat.

1,438
1,122
833
981

VII.
V III.
Building Chemi­
cals.
materials.

889
1,067
1,071
1,193
1,362
1,244
1,214
1,084

621
641
651
730
917
1,019
1,091
1,097

All
groups
together.

806
877
932
999
1,288
1,222
1,115
1,070

A table contained in the first of the two reports compares index
numbers of wholesale prices in Australia with those of the United
Kingdom, Belgium, Germany, Italy, France, Canada, the United
States, and New Zealand. Complete data for all countries are
shown for the years 1890 to 1911, inclusive. In the cases of the
United Kingdom and the United States the comparison is extended
back to 1840. This comparative table does not show, of course,
the relative prices as published in the different countries, but as
recomputed for each country on the base 1911 equals 1,000. The
last, column of the table contains figures computed by weighting
the index number for each country by its relative population, thus
supplying what the author says may be termed the world's index
number.
AUSTRIA-HUNGARY.
INDEX NUMBERS OF DR. BELA YON JANKOYICH.1
PUBLICATION AND HISTORY.

This index for the years 1867-1897 appeared under the title
“ Die Fluktuation der Waarenpreise im Grosshandel und die
Schwankungen der Wechselkurse der oesterreich-ungarischen Papiervaluta in den Jahren 1867-1897" (The fluctuation of wholesale
prices and the variation in the rate of exchange of the AustroHungarian paper values). It was printed in the Hungarian eco­
nomic review entitled “ Kozgazdasagi Szemle 1899."
It was continued to 1909 and reprinted in the Bulletin of the
International Institute of Statistics, volume 19, Part III, page 136
et seq., in 1912 (B u lle tin d e U l n s t i t u t In te r n a tio n a l d e S ta tistiq u e ),
under the title “ Index Nummer von 45 Waaren in der oesterreichungarischen Monarohie, 1867-1909; System Sauerbeck, zum Teil vom
Verfasser korrigiert” (Index number of forty-five articles in the
1 Dr. von Jankovich was in 1911 professor of the theory of finance and credit in the University of Buda­
pest (according to the Minerva Yearbook of the Learned World) and vice president of the Hungarian
Chamber of Deputies.




167

INDEX NUMBERS---- AUSTRIA-HUNGARY.

Austro-Hungarian monarchy, 1867-1909, according to the system
of Sauerbeck, with some corrections by the author).
Sauerbeck’s method of computation was followed as closely as
possible in order that the Austrian index might be comparable
with Sauerbeck’s index, since England throughout the period had
a gold standard. Articles were also selected to correspond as closely
as practicable with those entering into Sauerbeck’s index.
SOURCE OF QUOTATIONS.

The sources of the actual prices vary, being mostly official and
semiofficial publications of Austro-Hungarian cities and Provinces.
BASE PERIOD.

The years 1867-1877 were used as a base for all articles except
petroleum, for which 1873-1877 constitutes the base, and flax, for
which 1874-1877 constitutes the base.
NUMBER AND CLASS OF COMMODITIES.

The index includes 45 articles. Actual as well as relative prices
are shown for all articles except tea, but in a few cases the actual
prices are not complete. Relatives are also shown for each of six
groups into which the 45 articles are separated, and for the entire 45
taken together. The six groups are: Grains (1-8), animal products
(9-15), colonial goods (16-19), minerals (20-26), textiles, raw matetials (27-34), and miscellaneous (35-45).
TABLE OF RESULTS.

The following table summarizes the results of Dr. von Jankovich’s
compilations:
SUMMARY TABLE SHOWING IN DEX NUMBERS FOR THE PRICES OF 45 ARTICLES IN
THE WHOLESALE MARKETS OF AUSTRIA-HUNGARY.1
[Source: Bulletin de L’Institut International de Statistique, vol. 19, Pt. HI, p. 156.]

Year.

1867...........
1868...........
1869...........
1870...........
1871...........
1872...........
1873...........
1874...........
1875...........
1876...........
1877...........
1878...........
1879...........
1880...........

Grains
(1-8).

Animal Colonial
products
goods
(9-15).
(16-19).

110
95
88
95
101
101
108
114
91
96
103
91
89
103

90
96
98
101
102
108
111
103
97
97
97
94
90
96

101
99
104
103
104
102
99
98
95
94
103
86
82
82

All
foods
d-19).

101
96
95
99
102
104
107
106
94
96
101
91
88
96

Index
Miscella­ All raw number,
Mineral Textiles
(raw
neous
raw
Austriaproducts materials) materials materials Hungary
(20-26). (27-34).
(35-45). (20-45).
for 45
articles.
101
93
99
104
116
106
107
98
86
97
93
89
81
86

117
104
109
106
105
110
95
87
87
86
94
85
82
87

104
99
103
106
106
103
100
95
92
96
98
89
82
82

107
99
103
106
109
106
101
94
89
93
96
88
82
84

104
98
100
102
106
105
104
99
91
95
98
89
89
85

* Numbers appearing in the boxes refer to the column numbers of the articles making up the six groups
as they appear m the detailed table.




168

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

SUMMARY TABLE SHOWING IN D E X NUMBERS F O R THE PRICES OF 45 ARTICLES IN
THE W HOLESALE M ARKETS OF A U STR IA -H U N G A RY 1—Concluded.

Year.

1881...........
1882...........
1883...........
1884...........
1885...........
1886...........
1887...........
1888...........
1889...........
1890...........
1891...........
1892...........
1893...........
1894...........
1895...........
1896...........
1897...........
1898...........
1899...........
1900...........
1901...........
1902...........
1903...........
1904...........
1905...........
1906...........
1907...........
1908...........
1909...........

Grains
(1-8).

99
99
93
93
84
82
79
79
78
83
89
80
75
74
76
71
81
89
79
76
81
83
83
91
96
89
95
109
115

Animal Colonial
products
goods
(9-15).
(16-19).

97
101
103
101
97
93
92
94
90
S2
96
90
91
91
97
94
94
99
97
98
97
103
112
109
121
129
131
125
130

79
77
72
64
62
01
66
67
70
64
64
64
68
62
58
54
47
45
46
47
41
37
39
45
46
44
46
46
48

All
foods
(1-19).

Index
Miscella­ All raw number,
Mineral Textiles
(raw
neous
raw
Austriaproducts materials) materials materials Hungary,
(20-26). (27-34). (35-45). (20-45).
for 45
articles.

94
95
92
90
84
82
81
82
81
82
86
80
79
78
80
76
78
83
78
78
78
81
84
88
95
91
98
100
107

86
89
82
77
74
74
75
80
78
81
82
79
78
73
72
74
76
81
96
109
106
92
93
90
96
112
117
105
103

82
81
78
78
74
73
74
71
72
70
67
65
69
65
64
63
62
59
67
76
71
71
76
78
76
85
88
79
79

79
78
83
85
80
74
73
71
73
70
70
67
71
66
66
65
66
68
68
74
73
69
6768
69
74
79
76
79

82
80
82
81
77
74
74
74
74
73
72
70
72
68
67
66
68
68
75
84
81
76
77
77
79
88
92
84
86

87
86
86
85
80
77
77
77
77
77
78
74
75
72
72
71
72
75
76
82
80
78
80
82
85
91
95
91
94

1 Numbers appearing in the boxes refer to the column numbers of the articles making up the six groups
as they appear in the detailed table.

INDEX NUMBERS OF MARIO ALBERTI.
PUBLICATION.

A volume entitled “ II costo della vita i salari e le paghe a Trieste
nelT ultimo quarto di secolo” (The Cost of Living, Salaries, and
Wages in Trieste During the Last Quarter of the Century), by Mario
Alberti, was published in Trieste in April, 1911, under the direction
of the Museo Commerciale, a nonofficial organization of that city.
A second volume, II movimento dei prezzi e dei salari a Trieste, 1911
(The Movement of Prices and Wages in Trieste, 1911), was published
by the same author in 1912.
HISTORY.

Mario Alberti in his two works has made a study of prices and has
computed index numbers for certain commodities purchased by
contract in Trieste.
In his cost of living study the author traces from the time of ancient
Rome to the present the interest manifested in the question of the
cost of living and the means adopted to study the movement of
prices in various epochs and countries. Chapter IV of this work
deals exclusively with conditions in Trieste. The volume on the
movement of prices and wages also relates exclusively to Trieste.




INDEX NUMBERS---- AUSTRIA-HUNGARY.

169

SOURCE OF QUOTATIONS.

The prices used in the tables of index numbers are contract or semi­
wholesale prices quoted by the Istituto dei Poveri (Institution for the
Poor) and by the Austrian Lloyd Steamship Co. Those for the
Istituto dei Poveri extend from 1885 to 1911, inclusive, and for the
Austrian Lloyd Steamship Co. from 1892 to 1911, inclusive.
BASE PERIOD.

For the index numbers relating to the Istituto dei Poveri the
year 1885, taken as 100, is used as a base. For those of the Austrian
Lloyd Steamship Co. the years 1892-1896, taken as 100, are used
as a base.
PRICES: H O W SHOW N AND COMPUTED.

Tables are presented showing by years the average annual price of
each commodity purchased by the Istituto dei Poveri and by the
Austrian Lloyd Steamship Co. In addition, the simple index, the
coefficient or weight attributed to the commodity, and the weighted
index for the commodity are given. The totals for each year show
the figures on which the simple index and the weighted index for all
commodities taken as a whole are computed.
NUMBER AND CLASS OF COMMODITIES.

The commodities included are:
(a ) For the Istituto dei Poveri:
Bread.
Flour.
Paste (macaroni, etc.).
’ Rice.
Kidney beans.
Potatoes.
Oil.
( b)

Cheese.
Meat.
Sugar.
Wine.
Vinegar.
Coal.
Soap.

For the Austrian Lloyd Steamship Co.:
Bread. (x)
Flour N. O. (national).
Paste (macaroni, etc.), national,
fine.
Rice, fine, in transit.
Kidney beans.
Potatoes.
Oil, fine, in transit.
Cheese, Parma, in transit.
Beef (cow).

Sugar.
Wine, Dalmatian, in transit.
Vinegar.1
Soap.1
Fresh butter.
Coffee, Santos, in transit.
Milk.
Lard.
Eggs.
Meal, yellow.

i Prices are average prices paid by the Istituto dei Poveri.




170

BULLETIN OF THE BUREAU OF LABOR STATISTICS.
SUBSTITUTIONS AND ADDITIONS.

For 1910 two sets of prices were used in the computation of the
indexes. The report on the cost of living contained prices based
on estimates, while the volume devoted to the movement of prices
and wages contained prices based on actual expenditures. On page
21 of the latter the author states that for this reason “ The index
number for 1910 was newly calculated on the base of prices furnished
by the Istituto dei Poveri.”
WEIGHTING.

The prices of the different articles are weighted by the US3 of
coefficients which represent the relative amounts of the commodities
consumed. These coefficients are shown for the tables relating to the
Istituto dei Poveri in the following list:
Bread..................................• 26
Flour, wheat.......................
1
Paste (macaroni, etc.).........
4
Bice.................................... 2. 5
Kidney beans.....................
2
Potatoes..............................
10
Oil.......................................
2

Cheese.
Meat..
Sugar..
Wine...
Vinegar...............................
Coal....................................
Soap....................................

1-5

20
4

20
3-5
50
1

The coefficients used in the tables for the Austrian Lloyd Steam­
ship Co. are as follows:
26
Wine...................................
Bread..............................
20
Flour, wlieat....................
Paste (macaroni, etc.)---Bice................................
Kidnev beans.................
Potatoes..........................
Oil...................................
Cheese.............................
Beef................................
Sugar...............................

1
4
2.5
2
10
2
1-5
20
4

Vinegar........................ ......
Soap....................................
Fresh butter........................
Coffee..................................
Milk....................................
Lard....................................
Eggs....................................
Meal, yellow.......................

3-5

1
2
1
30

2
50

1

TESTING.

No comparison with other index numbers or other means of
•testing the accuracy of the results obtained has been attempted so
far as the published information discloses.
TABLES OF RESULTS.

The first of the three following tables shows the average price of
each of 14 articles for the base year 1885; the coefficients used in
computing the weighted index; the average price, the simple index
and the weighted index for each of the 14 articles for the years 1910
and 1911, and the totals of the simple and the weighted indexes for
those two years. The prices in the table are based on reports from
the Istituto dei Poveri.




171

INDEX NUMBERS---- AUSTRIA-HUNGARY.

The second table shows the indexes, both simple and weighted,
for the 14 articles taken as a whole, in yearly periods from 1885 to
1911, inclusive.
The third table shows the average prices for 1892-1896 of 19
articles (18 food articles and soap); the coefficients; the average
price, the simple index and the weighted index for each article in the
years 1910 and 1911; and the totals of the indexes for 1910 and 1911,
respectively. The prices are based on the period 1892-1896, taken as
100, and were obtained from the Austrian Lloyd Steamship Co.
IN D E X NUMBERS BASED ON PRICES OF ISTITUTO DEI P OVERI.1

Name of article.

Bread.........................
Flour, wheat.............
Paste (macaroni, etc.)
R ice...........................
Kidney beans............
Potatoes....................
O il.............................
Cheese........................
Meat...........................
Sugar.........................
W ine.........................
Vinegar......................
Coal............................
Soap...........................

1910

Price
in base
period,
1885.

Coeffi­
cient.

32.00
25.52
36.00
30.50
15.52
8.00
120.00
137.60
92.00
43.36
66.00
17.00
2.00
52.00

26
1
4
2.5
2
10
2
.2
20
4
20
.6
50
1

Price.

38.17
35.69
48.47
31.82
29.69
8.06
102.43
213.57
105.00
79.14
46.50
6.00
2.14
56.90

Simple
index.

1911
Weighted
index.

119.28
139.85
134.64
104.33
191.26
100.75
85.36
155.22
114.13
182.52
70.45
35.29
107.00
109.42

3,101.28
139.85
538.56
260.82
382.52
1,007.50
170.72
31.04
2,282.60
730.08
1,409.00
21.18
5,350.00
109.42

1,650

15,535

Price.

37.54
34.19
49.97
33.91
35.05
11.04
106.70
219.98
138.00
82.13
78.00
6.03
2.04
59.33

Simple Weighted
index.
index.
117.31
133.97
138.81
111.18
225.84
138.00
88.92
159.87
150.00
189.42
118.18
35.47
102.00
114.10

3,050.06
133.97
555.24
277.95
451.68
1,380.00
177.84
31.97
3,000.00
757.68
2,363.60
21.28
5,100.00
114.10

1,823

17,415

IN D E X NUMBERS BASED ON PRICES OF ISTITUTO DEI POVERI, FOR EACH YEA R ,
1885 to 1911, INCLUSIVE.1

Year.
1885.............................
1886.............................
1887.............................
1888............................
1889............................
1890............................
1891............................
1892............................
1893............................
1894.............................
1895.............................
1896............................
1897............................
1898............................
1899............................
1900............................
1901............................
1902............................
1903............................
1904............................
1905............................
1906............................
1907............................
1908............................
1909............................
1910............................
1911.............................

Simple
index.
100.00
97.21
95.57
94.14
90.21
92.43
93.80
97.86
96.86
95.43
90.28
91.86
96.93
99.57
99.87
101.21
101.78
103.00
102.43
100.86
108.00
105.00
111.43
116.79
117.14
117.86
130.21

Weighted
index.
100.00
95.74
93.52
90.54
88.96
94.11
97.55
92.77
92.46
89.69
86.32
90.79
97.40
99.99
102.43
114.95
116.11
108.09
103.31
100.36
104.97
104.36
111. 72
117.55
112.33
108.41
121.53

1 Mario Alberti, II Movimento dei prezzi e dei salari nelF anno 1911 a Trieste, pp. 24 and 25.




172

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

IN D E X NUMBERS BASED ON PRICES OBTAINED FROM THE AUSTRIAN LLO YD
STEAMSHIP CO.i

Name of article.

Price
in base
period, Coeffi­
1892- cient.
1896.

Bread2.......................................... 27.916
Flour, N. 0 ., national................... 25.952
Paste(macaroni, etc.),national, fine 35.920
36.880
Kidney beans................................ 22.740
6.392
Potatoes........................................
Oil, fine, in transit........................ 97.908
Cheese, Parma, in transit.............. 181.752
Beef (co w )2................................... 94.300
Sugar............................................. 64.996
Wine, Dalmatian, in transit........ 23.322
Vinegar2........................................ 12.896
Soap2............................................. 50.092
Fresh butter.................................. 224.992
Coffee, Santos, in transit.............. 193.300
Milk............................................... 20.000
Lard............................................... 102.612
Eggs............................................... 6.232
Meal, yellow.................................. 17.364

26.00
1.00
4.00
2.50
2.00
10.00
2.00
.20
20.00
4.00
20.00
.60
1.00
2.00
1.00
30.00
2.00
50.00
1.00

1910

Price.

1911

Simple Weighted Price.
index. index.

38.17 136.73
40.61 156.48
45.46 126.56
43.95 119.17
30.55 134.35
7.57 118.43
135.40 138.29
252.08 138.70
105.00 111.34
80.99 124.61
96.61
22.53
46.52
6.00
56.90 113.59
231.68 102.98
108.87
56.37
22.00 110.00
187.67 182.90
6.60 105.90
23.68 •136.37

3,554.98
156.48
506.24
297.94
268.69
1,184.30
276.58
27.74
2,226.80
498.44
1,932.20
27.91
113.59
205.96
56.37
3,300.00
365.80
5,295.00
136.37

2,256

20,431

37.54
35.22
47.86
45.11
31.72
9.86
145.60
240.52
138.00
83.66
44.06
6.03
59.33
259.96
154.38
23.80
160.65
7.18
22.68

Simple Weighted
index. index.
134.48
135.71
133.24
122.32
139.49
154.25
148.71
132.34
146.35
128.72
188.92
46.76
118.44
115.55
79.87
119.00
156.56
115.05
130.62

3,496.48
135.71
532.96
305.79
278.98
1,542.50
297.42
26.47
2,927.00
514.8g
3,778.40
28.06
118.44
231.10
79.87
3,570.00
313.12
5,752.50
130.62

2,446

24,060

1 Mario Alberti, II movimento dei prezzi e dei salari nelP anno 1911 a Trieste.
2 Index numbers based on prices paid by the Istituto dei Poveri.

BELGIUM.
INDEX NUMBERS OF HECTOR DENIS.
HISTORY.

Hector Denis, professor at the University of Brussels, is believed to
be the author of the only series of index numbers of prices in Belgium
that has been presented with any measure of continuity and com­
pleteness. This series has not, however, been published in uniform
manner, but has appeared as a gradual development in various pub­
lications credited to its author. Even at the present time it does not
seem to have reached its final form, since every succeeding presenta­
tion either alters or omits former tables or includes additional ones.
The only exception seems to be a table of index numbers computed
for 28 articles of export, which is given in 1911 in practically the same
form in which it first appeared in 1895.




INDEX NUMBERS---- BELGIUM.

173

PUBLICATION.

Among the publications of Prof. Denis in which his various
tables of indexes appear are the Economic and Social Depression and
the History of Prices (L a d e p r e s s io n econorrdque et s o cia le et VTiistoire
d es p r i x ) , 1895,1 and the Index Numbers of Moral Phenomena (L e s
in d e x n u m b e rs — n om b res in d ic e s — d es p h en o m e n es m o r a u x ), 1911.2
The most nearly complete examples of his indexes, however,
are published in the Bulletin of the International Institute of Statis­
tics, volume 19, Part III, pages 157-195, and are computed to
include the years 1909, 1910, or 1911, as the case may be. Only
two of the earlier tables appear in the bulletin. Of these the
index numbers based on the 28 exports are continued to include
the year 1910, as is likewise a comparative table that had appeared in
his Index Numbers of Moral Phenomena. This comparative table is
used by Prof. Denis to test his own general index for the 28 articles of
export of Belgium. He reduced the general index numbers of France
(Palgrave), Germany (Soetbeer), England (Sauerbeck), and the
United States (Falkner-Hardy) to the common base 1867-1877. and
presented them in parallel columns with his own for each year and for
five-year periods from 1860 to 1893.3 In the same publication the
table appears a second time, but with data for each year from 1850 to
1910,4 and without the reduction of the indexes to a common base. A
second table of index numbers of exports, similar to but not identical
with the one published in 1895 on the base period 1865, also appears
in the bulletin. Separate tables of index numbers are also given for
vegetable products, meats and butter, cereals, and other groups of
articles.
SOURCE OF QUOTATIONS.

Prof. Denis bases his computations for the index numbers appear­
ing in his study of the economic and social depression on the quota­
tions given in the tables of foreign commerce of Belgium (ta blea u x d u
co m m erce e x te r ie u r d e la B e lg iq u e ), i. e., on the prices used for fixing
1 In this appear four tables of indexes. The first is an index for 23 exports yearly from 1850 to 1890, com­
puted on the base period 1867-1877. The second index covers the same period of years and almost the same
articles, except that some are grouped and one or two are added, the base period being the single year 1865.
The third index is similar to the first, but is based on 22 imports. The fourth table presents the general
index number for imports and also for exports for every year from 1865 to 1890, the period 1867-1877 again
being used as the base.
2 Published by the Royal Academy of Belgium in its Memoires, second series, 4, Brussels, 1908-1911.
Separate tables of price index numbers are given for cereals, beef, wheat, coal, and metals, all computed on
the base period 1867-1877. A comparative table shows index numbers fcr the United States, Belgium,
Germany, and England.
3 This table is given on p. 175 of this bulletin.
* The general index computed for Belgium by Prof. Denis, which appears in Hi is table, is printed on p. 175
of this bulletin for the years 1894-1910.




174

BULLETIN OF TH E BUREAU OF LABOR STATISTICS.

the customs values. Of these Mr. Armand Julia1 remarks that he
finds that the averages adopted by the commission for the official
values correspond but remotely with the real fluctuations in prices.
Therefore the accuracy of Prof. Denis’s index numbers may appear
rather doubtful. However, this criticism can apply only to the single
original table of 28 exports that is continued to include 1910, since
in the additional tables published in the bulletin of the International
Institute of Statistics referred to above he states in a footnote that
prices up to 1852 were taken from the secular almanac of the observa­
tory (V A lm a n a c k s ec u la ir e d e V O b serva toire) and those for later years
from the statistical yearbooks (A n n u a ir e s s ta tis tiq u es ).
BASE PERIOD

For the base period Prof. Denis selected the years 1867-1877, as
did Sauerbeck for England. His reason for such a choice appears to be
the fact that during those years there were periods of both rise and
depression in prices, while a later period would not include the eco­
nomic depression that followed 1873 and an earlier one would cover a
time of rising prices only.
DESCRIPTION OF COMMODITIES

The articles for which index numbers have been computed are not
described specifically in the publications mentioned above, but it is
stated in the one dealing with the economic and social depression
that the 28 exports (whose indexes are continued to 1910) and the
22 imports (not computed after 1890) were selected with a view to
including those most prominent, and that the total of 50 articles so
selected comprises two-thirds of the total exports and imports of
Belgium.
WEIGHTING.

The indexes are not weighted, and in his Economic and Social De­
pression and the History of Prices Prof. Denis justifies himself by
stating that in spite of the use of weighted averages the variation in the
curve of prices as shown in the diagrams accompanying that publica­
tion remain essentially the same and that therefore he has not
abandoned the simpler method, but presents his indexes with the
caution that the results are to be taken only as an approximation of
the truth.
TABLES OF RESULTS.

Below follows the comparative table of general index numbers for
several countries as it appears in his treatise on the Index Numbers
of Moral Phenomena mentioned above:
1 “ The economic progress of Belgium from 1880 to 1908,” In the Journal of the Royal Statistical Society,
1911, p. 268.




175

INDEX NUMBERS---- BELGIUM.

COMPARISON OF GENERAL IN D E X NUMBERS FOR FRANCE (PALG R AV E ), GERMANY
(SOETBEER), ENGLAND (SAUERBECK), UNITED STATES (FA LK N E R-H A R D Y), AND
BELGIUM (DENIS), REDUCED TO A COMMON BASE.
(Base period, 1867-1877=100.)

Year.

1860.
1861.
1862.
1863.
1864.
1865.
1866.
1867.
1870.
1871.
1872.
1873.
1874.
1875.
1876.
1877.
1878.
1879.
1880.
1881.
1884.
1885.
1886.,
1887.
1888.
1889.
1890.
1891..
1892.
1893.

Germany
(Soetbeer).

Trance
(Palgrave).

United
States
(FalknerHardy).

England
(Sauerbeck).

Belgium
(Denis).

103.6

101
103
105
97.6'
97.6
97.6
97.6
94.6
101.51
108.3
107.3
98.5
99.5,
97.6
99.5

101
1021
100
99
98
96

►96.7

117.6

100:
109
111•103.6
102
96
95
94
87
83

86.8

90.7
83.9
85.8.
82.9
81.9 79.7
80
74.1

- 96.7

9.2

•76.7

88

100.1

85:
84
82 79.8
76
72
69;

•92.1

104.2
116.6
106.9
106.9
93.1
95.8 95.8
91.6
91.8
98.8
109
108.6 107.6
108.3
113.4
108.31
103
98.4 101.1
95.8

- 73.7

68

70 70.2
72
72
72'

68
68

97.8
85
85.7 16.1
82.8
79.1
77.8:
76.2
75.6 74.2
71.
70
69.9
66.5
64.4

A continuation of the general index numbers for Belgium as com­
puted by Prof. Denis in the above table appears on page 158 of volume
19, Part III, of the Bulletin of the International Institute of Statis­
tics, as follows:




Year.

Index
numbers.

1894.... ... .
1895
1896
1897
1898
1899
1900
1901
1902 .......

59.0
61.5
61.0
56.0
58.5
61.8
63.4
63.1
64.0

Year.

Index
numbers.

1903
1904
1905
1906
1907........ ...
1908....... ...
1909
1910
.

60.6
61.8
60.9
64.8
69.2
67.7
65.9
64.7

176

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

CANADA.
INDEX NUMBERS OF THE DEPARTMENT OF LABOR.
PUBLICATION.

This compilation of wholesale prices for the Dominion of Canada is
published yearly by the Department of Labor of Canada at Ottawa.
Index numbers for each group of commodities and for its main sub­
divisions are also published monthly in the Labor Gazette, the official
organ of the department of labor.
HISTORY.

The first report on wholesale prices made by the Canadian Depart­
ment of Labor was published in 1910 and covered the years 1890 to
1909, inclusive. The object in undertaking this work was to deter­
mine as accurately as possible the nature and extent of the general
rise in wholesale prices which had occurred in Canada during recent
years. Prior to the beginning of this work the Labor Gazette, the
official organ of the department, had for some time published each
month certain data regarding prices in connection with its review of
industrial and labor conditions. The importance of the subject and
the unsatisfactoriness of general statements in a matter of this kind
led the department in 1910 to adopt a more comprehensive and sys­
tematic method of treating the subject of prices in the monthly sum­
mary and also to extend the inquiry into the wholesale prices of a
selected list of representative staple commodities back over the pre­
ceding 20 years.1 In subsequent annual and monthly reports the
price data have been brought down to the present time.
SOURCE OF QUOTATIONS.

It is stated that the practice followed throughout the investigation
was “ to collect and collate the best available published information
and to submit the result for verification to long-established firms at
the wholesale center in question.” The daily press and weekly trade
journals of Canada and the printed reports of exchanges, boards of
trade, etc., are mentioned as the principal sources of data. When
reliable printed matter failed, information was obtained from books
of manufacturers and wholesalers.
A source used for verification purposes in the case of a few impor­
tant raw materials imported by manufacturers direct from the primary
markets of the world, and in which there is no wholesale trade in
Canada, was the declared import values, which were divided by total
quantities to show the average prices. Toronto and Montreal markets
furnish the great mass of the quotations published in the reports.2
i Wholesale prices in Canada, 1890-1909, p. 2.




2 idem,

p. 8.

INDEX NUMBEKS— CANADA.

177

BASE PERIOD.

The base period selected for the computation of index numbers for
practically all commodities is the decade 1890-1899. Two reasons
are given for this selection: (1) The period was considered as repre­
sentative of normal conditions as any available, containing a time of
falling and a time of rising prices, and (2) direct comparison with the
similar study of the United States Department of Labor was con­
sidered very desirable, and this was made possible by choosing the
same base period.1 In a few instances, owing to special reasons, a
period other than the decade 1890-1899 has been chosen as the base.
PRICES: H O W COMPUTED AND SHO W N .

Tn the first report the prices quoted are stated to be “ for the most
part those prevailing on the opening day of each month, though if,
in particular cases, these were found to be abnormal, an average of
the week was taken.” 2 In the report for 1912 it is stated that the
manner of quoting prices is the same as in the earlier reports except
that for certain articles subject to rapid fluctuations (grains, live
animals, certain meats, butter, eggs, potatoes, and fresh fruits— 40
in all) weekly instead of monthly quotations were obtained.3 This
plan was continued in the preparation of the wholesale-price data for
1913.
Difficulty was encountered in obtaining quotations of a uniform qual­
ity of certain articles, particularly of manufactured articles, through a
series of years. It is stated in the reports that wherever such articles
are quoted, care has been taken to see that changes in quality are
accounted for in the prices given.4 In a few cases—as, for example, in
the case of cotton goods—the prices published are not simple quota­
tions on a single variety, but averages of a large number of varieties.
In the annual reports the actual prices are published for each com­
modity by months, or, in some cases, by weeks, and the average of
these quotations is given as the price for the year. Index numbers
are published in the annual reports for each commodity by years and
in the Labor Gazette for each group and subgroup (56 items in all)
by months currently. Index numbers do not seem to be published
for single commodities by months. Many of the actual prices are
stated in the form of a range of price, and apparently the mean is
used for computations based on these figures.
Some commodities whose price is largely governed by seasonal
conditions are quoted for only those months of the year when they are
in season—as, for example, blue grapes, for which quotations are given
only for September and October.
1 Wholesale prices in Canada, 1890-1909, p. 440.
2 Idem, p. 439.

94261°—Bull. 173—15----- 12



8 Idem, 1912, p. 2.
* Idem, 1890-1909, p. 439.

178

BULLETIN OF THE BUREAU OF LABOR STATISTICS.
NUMBER AND CLASS OF COMMODITIES.

The index number for the 20-year period 1890-1909 is based on 230
commodities so-called, some of these quotations being, however, as in
the case of cotton fabrics, the average of a large number of varieties
of the articles. In the first annual report (covering the year 1910)
one quotation was dropped and six new quotations were added, thus
making the number of commodities 235. In the second annual report
(covering the year 1911) one quotation was dropped, but the entire
number covered by the index was increased to 261. In the latter
report the statement is made that it is hoped ultimately to include
about 280 commodities. The total was increased to 272 in 1912 by
the addition of nine new articles and additional series of quotations
in the case of two articles previously included. No change in the list
was made in 1913. The new articles have been included in the index
numbers since 1910, so as to assist immediate comparisons, but no
recalculation of the entire series of index numbers back to 1890 is to
be made on the enlarged basis until the number of commodities is
completed.1
In recent reports, prices for a number of articles which it is hoped
ultimately to include in the index number are published in connec­
tion with prices for the 272 commodities included at present. Both
raw materials and manufactured articles are included in the com­
modities used in computing the index number. Difficulties attend­
ing the employment of manufactured articles were recognized, but
their inclusion on a conservative basis seemed imperative in selecting
a sufficiently large number of representative commodities. With
respect to the original number, 230, it was said that “ the effect of
tendencies incidental to the manufacturing process are present in
about 40 per cent of the quotations.” 2
DESCRIPTION AND GROUPING OF COMMODITIES.

The commodities for which index numbers have been computed
are shown in the following list, which is taken from the report for 1913
(pp. 218-240).
I. Grains and fodder.

Barley, Western.
Barley, No. 2, Ontario.
Bran.
Corn, No. 3, yellow.
Flaxseed, No. 1, Northwestern.
Hay, Montreal.
Hay, Toronto.
Oats, No. 2, white, Western.
i Wholesale Prices in Canada, 1912, p. 2.




Oats, No. 2, white, Ontario.
Peas, No. 2, Ontario.
Rye, No. 2, Ontario.
Shorts.
Straw.
Wheat, No. 1, Northern.
Wheat, No. 2, white, Ontario.
2 Idem,

1890-1909, p. 438.

INDEX NUMBERS— CANADA.
II. Animals and meats.

Bacon, English boneless breakfast.
Beef, plate.
Beef, dressed, hind quarters.
Beef, dressed, forequarters.
Cattle, Western prime.
Cattle, choice steers, Toronto.
Fowls.
Hams, city cured, medium.
Hogs, selects.

Hogs, dressed.
Lard, pure.
Mutton, dressed.
Lamb.
Pork, Canada, heavy short-cut mess.
Sheep, export ewes.
Turkeys.
Veal, dressed.

m . Dairy products.

Eggs,
Milk,
Milk,
Milk,

Butter, creamery, Montreal.
Butter, creamery solids, Toronto.
Butter, dairy, prints, Toronto.
Cheese, western colored.
Eggs, fresh, Montreal.

storage, Toronto.
at Montreal.
at Toronto.
at Victoria, B.C.

IV. Fish.

Lobster, canned.
Mackerel, salted.
Salmon, B. C., canned.
Salmon trout, fresh.
Whitefish, fresh.

Codfish, dry, f. o. b.
Haddock, dry, f. o. b.
Halibut.
Herring, salted.
Lobster, fresh.

V. Other foods.

(a) Fruits and vegetables.

2. Dried fruits.
1. Fresh fruits.
Native—
Apples, evaporated.
Apples, good seasonable.
Currants, Patras.
Cherries.
Prunes, Bosnia.
Grapes, Blue.
Raisins, Sultanas.
Peaches, Leno covers, No. 1 3. Fresh vegetables.
fruit.
Beans, hand-picked.
Pears, early, Bartletts and
Onions, Canadian Red.
winter.
Potatoes, Montreal.
Plums, early, Lombards, greenPotatoes, Toronto.
Turnips.
Raspberries, red.
Tomatoes.
Strawberries.
4. Canned vegetables.
Foreign—
Com, standards, 2’s.
Bananas, yellow.
Peas, standards, 2’s.
Lemons, Messinas and VerdelTomatoes, 3’s.
lis.
Oranges, navels and Valencias.
(b) Miscellaneous groceries and provisions.

1. Breadstuffs.
Flour, straight rollers.
Flour, strong bakers.
Flour, winter wheat patents.
Flour, Manitoba 1st patents.
Bread, Toronto.




1. Breadstuffs—Concluded.
Bread, Victoria, B. C.
Biscuits, soda.
Oatmeal, standard.
Rice, Patna.
Tapioca, medium pearl.

179

180

BULLETIN OF TH E BUREAU OF LABOR STATISTICS.
V. Other foods— Concluded.

(6) Miscellaneous groceries and provisions—Concluded.
2. Tea, coffee, and chocolate.
Chocolate, Diamond.
Coffee, Rio, No. 7.
Coffee, Santos.
Tea, good common Japan.
3. Sugar, etc.
Glucose.
Honey, strained.
Maple sugar.
Molasses, New Orleans.

3. Sugar, etc.—Concluded.
Sugar, Montreal granulated.
Sugar, Montreal yellow.
4. Condiments, etc.
Pepper, black, pure.
Cream of tartar.
Salt, fine, dairy, cheese, and table.
Soda, bicarbonate of.
Vinegar, white wine, proof strength.
VI. Textiles.

(а) Woolens:
Wool (Ontario), washed.
Wool (Ontario), unwashed.
Yarn.
Woolen underwear.
Beaver cloth.
(б) Cotton:
Cotton, upland middling.
Gray cottons.
Woven colored fabrics.
Prints.
(c) Silk:
Silk, raw, Italian, classical.
Belding’s, 50 yards spool silk (A).
Belding’s prize medal (16-ounce,
machine).

Flax products:
Flax sewing twine.
Linen rope, white.
Flax fiber.
Tow, fine.
(e) Jute:
Jute, first marks.
Hessian, lOJ-ounce, 40-inch.
(/) Oilcloths:
Floor oil cloth, No. 3 quality.
Table oilcloth, assorted patterns
(5-4 wide).
(d)

VII. Hides and tallow, leathers, and boots and shoes.

Hides and tallow:
No. 1 inspected steers and cows.
No. 1 green calfskins.
Horsehides.
Tallow rendered, No. 1 stock, in
barrel.
Leather:
No. 1 Spanish sole, for jobbing.
No. 1 slaughter sole, heavy.

Leather—Concluded.
Harness, No. 1, U. O.
Heavy upper.
Boots and shoes:
Men’s split blucher bal., pegged.
Men’s box calf blucher bal., G. W.
Women’s Dongola blucher bal.,
F. S.

V in . Metals and implements.

(a)

Metals:
Antimony.
Brass.
Copper.
Iron, pig, Summerlee.
Iron, pig, No. 1 foundry, N. S.
Iron, common bar.
Iron, black sheets.
Iron, galvanized sheets.




(a)

Metals—Continued.
Iron, tin-plate, charcoal.
Iron, tin-plate, Bessemer.
Iron, boiler-plate.
Wrought iron, No. 1.
Lead, imported.
Lead, domestic, Trail.
Nickel.
Quicksilver.

INDEX NUMBERS---- CANADA.
V m . Metals and implements—Concluded.

Metals—Concluded.
Silver.
Solder.
Spelter.
Steel billets.
Steel, cast.
Tin, ingots.
Zinc, sheets.
(б) Implements:
Anvils, Wrights’ , 80 pounds and
over.
(а)

(6) Implements—Concluded.
Axes, standard.
Chains, coil.
Crowbars.
Grindstones, 40-200 pounds.
Horseshoes.
Mallets, carpenters’ hickory.
Screws, bench wood.
Soldering irons.
Vises, Wrights’ .

IX. Fuel and lighting.

Coal, N. S., run of mines.
Coal, Crow’s Nest Pass.
Coal, Pennsylvania anthracite.
Coke, Connellsville furnace.
Coke, Crow’s Nest Pass.

Gasoline.
Coal oil, Canadian standard.
Coal oil, United States standard.
Calcium carbide.
Matches.
X . Building materials.

(a) Lumber.

Pine, all grades, Ottawa.
Pine, good sidings, Ottawa.
Pine, ISTo. 1 cuts, Toronto.
Laths, pine, Ottawa.
Pine, shipping, culls, Ottawa.
Pine, box boards, Ottawa.
Hemlock, Ottawa.

Spruce, New Brunswick.
Shingles, New Brunswick.
Birch, Toronto.
Maple, Toronto.
Oak, Toronto.
British Columbia fir.
British Columbia shingles.

( b) Miscellaneous building materials.
Brick, fire.
Brick, common building.
Cement, Canadian Portland.
Hinges.
Iron pipe.
Lead pipe.
Lime.
Nails, cut.
Nails, wire.
Pitch.

Plaster of Paris.
Red lead, dry.
Sash cord.
Sash weights.
Soil pipe, medium.
Tar.
Wire, copper.
Wire, iron.
Wire cloth.
Wire fencing.
(c) Paints, oil, and glass.

Benzine.
Glue.
Linseed oil (raw).
Linseed oil (boiled).
Paris green.
Prepared paints.
Putty.




Rosin, white.
Shellac.
Turpentine.
Varnish.
Venetian red (dry color).
White lead.
Window glass.

181

182

BULLETIN OF THE BUREAU OF LABOR STATISTICS.
XI. House furnishings.

(а) Furniture:
Kitchen chairs (common spindle).
Kitchen tables (4-foot, with
drawer).
Dining tables (hardwood, exten­
sion, 8-foot).
Sideboards (hardwood, with mirror
16 by 28).
Bedroom sets (dresser and stand,
hardwood).
Iron beds, continuous pillars, 4-foot.
(б) Crockery and glassware:
Tumblers, tank glass, J-pint.
Earthenware: White cups and
saucers.

(6) Crockery and glassware—Concluded.
Earthenware: 10-piece printed toi­
let sets.
Earthenware: 97-piece printed din­
ner sets.
(c) Table cutlery:
Knives, celluloid handle, medium
size.
Silver-plated knives and forks, 6
pennyweight, medium.
(d) Kitchen furnishings:
Pails.
Tubs.
Brooms.
Sadirons, Mrs. Potts.

XII. Drugs and chemicals.

Copperas.
Glycerine.
Muriatic acid.
Opium.
Quinine.
Soda ash.
Sulphuric acid.

Alcohol, 65 0. P.
Alcohol, wood.
Alum.
Bleaching powder.
Borax.
Brimstone.
Carbolic acid.
Caustic soda.

XIII. Miscellaneous.

(a) Furs.

Mink, dark.
Muskrat, best fall and winter.

Raccoon.
Skunk, black Canadian.
(6) Liquors and tobacco.

Hops.
Malt.
Whisky (Can. Club 20-8 up).

Ale and porter (draft).
Tobacco, smoking.
Tobacco, raw leaf.
(c) Sundries.

Binder twine.
Gunpowder.
Paper, news print.
Rope.

Rubber, Para Island.
Soap.
Starch.
SUBSTITUTIONS AND ADDITIONS.

In cases where new articles have been included in the index number
an effort was made to secure a series of quotations back to 1890. In
some cases this was not found possible, owing to “ poverty of records,
changes in industrial methods or consumption standards, etc.” It is
stated that no satisfactory solution of this problem has been found,
but that “ the method followed was to assign to the new commodity
the index number of the commodity displaced or most nearly repre­
sented in the year in question. Thus the index number of the lowest



INDEX NUMBERS---- CANADA.

183

grade of pine lumber was assigned to hemlock in the year in which
that article first makes its appearance in the quotations. In this way
the new commodity creates a minimum of disturbance in the index
number of the year in which it first occurs, whilst subsequent varia­
tions make themselves duly felt.” 1 In the case of calcium carbide, an
entirely new commodity introduced into the index in 1894, and the
case of Crow’s Nest Pass coal and Crow’s Nest Pass coke, both intro­
duced in 1899, the price first quoted was taken as 100, or the base,
while in the case of cotton prints, introduced in 1893, the average
price for the years 1893-1899 was taken as the base.
INTERPOLATION.

So far as can be determined, no price quotations have been inter­
polated. In the case of a few commodities, however, the statement
is made that, owing to incomplete records, associated data have been
drawn upon in calculating the base prices. Thus the price of flax­
seed for the base period 1890-1899 was estimated from the price at
Chicago from 1890-1910, as published in the reports of the United
States Bureau of Labor Statistics, and the price at Winnipeg from
1906-1910. The base price of plate beef was in like manner “ cal­
culated from the percentages of cattle and beef prices from 1906 to
1911,” while the base price of dressed veal is “ based on the average
prices of other meat products, 1890-1899.” 2
WEIGHTING.

The general index number is the simple average or arithmetic
mean of the index numbers of the several commodities; i. e., the sum
of the relative prices of the different commodities, divided by the
number of commodities. Certain commodities are represented by
more than one quotation and, as would happen in any extensive
list including both raw materials and manufactured products, some
commodities are represented indirectly more than once, as, for exam­
ple, lumber, which is also represented by furniture. In the opinion
of the compiler “ an extended list of articles tends to weight itself.”
TESTING.

With the object of testing the results obtained by the use of the
simple arithmetical average of the index numbers representing the
several commodities, a weighted index has been computed. This is
based on the table prepared by the British Association for the
Advancement of Science, a committee of which dealt exhaustively
with the whole subject of index numbers in 1887-1890. The table
follows.3
1 Wholesale Prices in Canada, 1890-1909, p. 447.
pp. 129,130.

2 Idem, 1913,




a idem, 1890-1909, p. 442.

184

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

BRITISH ASSOCIATION FOR THE ADVANCEMENT OF SCIENCE TABLE OF WEIGHTS
FOR THE CONSTRUCTION OF AN IN D E X NUMBER.
[The following explanation of the table is given in the Canadian report:
The second column gives in round numbers (000,000’s being omitted) the average national expenditure
(in pounds) on each class of article at present and for the last few years, and presumably also for the imme­
diate future the proportion at least, if not the absolute amounts, of expenditure. In the estimated amount
of consumption allowance is made for the addition to the value made before the articles are in the form in
which they are finally consumed.
In the third column these amounts (or proportions) are reduced to percentages (of the total amount
expended on such articles).
In the last column the relative importance proposed to be assigned to each article in the index number
is stated, mainly on the basis of the percentages in the third column, but with modifications so as to substi­
tute even figures for the convenience of handling.]

Articles consumed or used up.

Breadstuffs:
Wheat..............................................................................................
Barley..............................................................................................
Oats..................................................................................................
Potatoes, rice, etc............................................................................

Relative
importance
Estimated
proposed
expendi­ Percentage
for each
of each
ture per
annum on amount to article in
index
each
total.
number
article.
reduced to
percentage.

60
30
50
50

6.5
*3.3
5.4
5.4

5
5
5
5

100
20

110.9
2.2

10
2J

60

6.5

20
Meat.................................................................................................
Fish..................................................................................................
Cheese...............................................................................................
Butter .........................................................................................
Millc..................................................................................................

7i
20

Mass luxuries:
Sugar................................................................................................
Beer..................................................................................................
Spirits.............................................................................................
Win ft................................................................................................
Tobacco...........................................................................................

30
20
100
40
10
10

3.3
2.2
110.9
4.3
U .l
il.l

24
%

9

?

2i
20

Clothing:
Cotton..............................................................................................
W ool................................................................................................
Sills...................................................................................................
Leather............................................................................................

20
30
20
10

2.2
3.3
2.2
1.1

2§
2l
2|
2|
10

Metals and minerals:
Coal..................................................................................................
Iron..................................................................................................
Copper..............................................................................................
Lead, tin, zinc, etc..........................................................................

100
50
25
25

110.9
5.4
2.7
2.7

10
5
2£
2$
20

Miscellaneous:
Timber.............................................................................................
Petroleum........................................................................................
Indigo..............................................................................................
Flax and linseed..............................................................................
Palm oil...........................................................................................
Caoutchouc......................................................................................

30
5
5
10
5
5

3.3
1.5
!,5
1.1
1.5
1.5

920

100.0

3
1
1
3

1
1
10

Total..........................................................................................

100

1 This percentage does not agree with that found in the Candian report, but is correct according to the
expenditure given in the preceding column.




185

INDEX NUMBERS---- CANADA.

Slight modifications were made in the above table to meet the
groupings adopted in the Canadian investigation and the different
standards of consumption in that country. In the absence of statistics
directly bearing on consumption standards in Canada, apart from
statistics of import and export trade and of production, use was
made of the special studies of family expenditures of the British,
United States, and Massachusetts labor departments.
The following figures show the weighting used in the Canadian
study:1
Group.

Weight.

Grains and fodder.......................................
Animals and meats.....................................
Fish.............................................................
Dairy produce.............................................
Other foods..................................................
Textiles.......................................................
Hides, leather, boots and shoes................
Metals and implements:
(a) Metals.............................................
(6) Implements....................................
Fuel and lighting........................................
Building materials:
(a) Lumber..........................................
(6) Miscellaneous building materials..
(c) Paints, etc......................................
House furnishings..............................................
Drugs and chemicals..................................
Miscellaneous:
(a) Furs............. ..................................
(6) Liquors and tobacco......................
(c) Sundry...........................................

18
10
2*
7*
15
8
2

10

Total..................................................

100

8
1

6

2
1
2
2
1
2
2

The results of the testing may be seen in the following table, which
gives both the weighted and unweighted index numbers for each year
from 1890 to 1913, inclusive.2




Year.

1890...................................
1891...................................
1892...................................
1893...................................
1894...................................
1895...................................
1896...................................
1897...................................
1898...................................
1899...................................
1900...................................
1901...................................
1902...................................
1903...................................
1904...................................
1905...................................
1906...................................
1907...................................
1908...................................
1909...................................
1910...................................
1911...................................
1912...................................
1913...................................

Weighted
number.
112.0
111.3
104.9
103.9
97.2
95.6
90.6
89.9
95.5
99.0
105.8
106.0
109.6
109.7
110.6
113.8
120.1
129.2
125.1
126.3
128.0
131.1
143.9
139.6

Unweighted
number.
110.3
108.5
102.8
102.5
97.2
95.6
92.5
92.2
96.1
100.1
108.2
107.0
109.0
110.5
111.4
113.8
120.0
126.2
120.8
121.2
124.2
127.4
134.4
135.5

1 Wholesale Prices in Canada, 1890-1909, p. 12.
2 Idem, 1913, p. 11.

186

BULLETIN OF THE BUREAU OF LABOR STATISTICS.
TABLE OF RESULTS.

The following table, reproduced from the 1913 report (p. 3), shows
by groups of commodities the index numbers for the 24 years 18901913, inclusive:
IN D E X NUMBERS OF COMMODITIES, B Y GROUPS, 1890-1913.
(Base period, 1890-1899=100.0.)

Group.
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.

11.
12.
13.

Grains and fodder.........
Animals and meats.......
Dairy products..............
Fish................................
Other foods....................
Textiles.........................
Hides, leather, boots....
Metals and implements:
Metals........................
Implements...............
Fuel and lighting..........
Building materials:
Lumber......................
Miscellaneous............
Paints, oils, glass......
House furnishings.........
Drugs and chemicals. . .
Miscellaneous:
Furs...........................
Liquors and tobacco.
Sundries....................

1890

1891

1892

1893

116.7
111.2
103.0
103.3
120.3
111.4
100.6

123.9
104.7
106.2
97.3
121.3
104.2
102.6

106.7
108.5
105.8
90.6
104.7
102.2
99.8

99.1 94.3 98.8 85.2 80.6 98.8 96.7 99.9 107.3
117.7 98.7 92.2 82.4 90.4 97.9 95.1 103.4 111.3
110.4 104.6 94.8 90.1 90.1 92.9 101.4 109.0 102.5
99.7 96.4 101.4 102.6 98.6 99.6 110.0 106.4 113.2
102.1 95.0 95.2 87.1 86.0 94.3 93.6 96.4 98.6
101.2 97.3 93.6 96.9 98.0 95.2 99.8 100.0 103.6
101.8 89.9 98.6 92.9 100.1 105.0 109.4 113.8 112.8

1.
2.
3.
4.
5.
6.
7.
8.
9.
10.

11.
12.
13.

Grains and fodder.........
Animals and meats.......
Dairy products..............
Fish...............................
Other foods....................
Textiles.........................
Hides, leather, boots__
Metals and implements:
Metals........................
Implements...............
Fuel and lighting..........
Building materials:
Lumber......................
Miscellaneous............
Paints, oils, glass___
House furnishings.........
Drugs and chemicals.. .
Miscellaneous:
Furs...........................
Liquors and tobacco..
Sundries....................

1895

125.4 114.4 107.6 102.1 91.1 87.0
103.8 103.2 102.9 102.6 102.2 101.0
107.4 106.7 106.6 102.9 97.5 97.0
103.5
117.6
109.5
100.2
110.5

102.7
110.4
103.8
100.5
110.3

104.4
106.8
98.2
100.9
104.4

103.7 104.6 102.8
103.7 98.7 95.2
98.6 95.5 96.1
101.1 101.3 97.9
104.4 103.1 100.3

86.5 99.7 103.7 123.6 113.5
94.9 99.0 99.7 99.4 98.7
112.0 106.7 98.8 100.3 93.7

Total....................... 110.3 108.5 102.8 102.5
Group.

1894

97.2

1896

1897

1898

1899

1900

87.5
98.5
98.9

85.7
93.1
96.4

97.1
93.9
96.2
97.5
99.8

93.9 90.8 95.8 114.0
87.7 87.4 97.2 111.8
95.5 100.0 107.6 125.9
99.8 99.0 100.2 110.2
96.5 96.8 93.3 101.5

1901

87.6 111.9 121.2 110.4
94.3 98.0 100.1 102.2
93.5 96.9 100.8 98.1
114.6
106.0
121.9
107.9
99.8

80.5
99.4
91.3

80.7 88.0 111.1 111.8 147.3 140.9
98.0 103.9 103.9 102.3 103.3 103.3
92.6 91.2 103.3 109.5 113.0 110.9

95.6

92.5

92.2

96.1 100.1 108.2 107.0

1902

1903

1904

1905

1906

1907

1908

1909

1910

1911

1912

1913

116.1
122.2
106.9
110.2
98.4
101.0
118.2

106.5
117.9
108.9
116.2
98.1
105.9
115.7

115.5
111.3
107.2
119.5
101.8
110.4
113.6

116.4
120.7
115.1
115.7
100.7
114.6
119.6

118.5
130.1
120.2
121.8
103.1
123.4
128.1

140.2
133.8
131.5
129.5
112.5
126.1
125.5

148.3
129.6
136.3
120.5
110.3
111.0
120.0

149.9
148.6
133.6
134.0
107.6
108.3
135.4

140.7
163.6
135.7
145.1
111.3
114.6
135.4

148.4
146.6
136.2
143.6
118.7
119.2
139.6

167.3
160.8
159.0
155.7
126.0
120.7
152.4

136.8
180.8
154.7
158.0
117.4
130.8
163.9

102.8 105.5 99.7 108.4 128.6 134.8 106.3 101.9 97.6 108.3 117.4 119.1
104.7 105.7 106.2 106.1 106.0 107.1 104.2 102.4 104.5 104.5 104.7 105.6
104.9 111.0 103.0 104.1 106.4 108.8 102.2 103.8 103.0 100.5 113.3 118.2
122.0
104.6
128.1
109.2
102.2

128.8
107.7
126.3
109.6
105.5

131.3
107.2
122.4
112.7
109.6

134.1
106.8
125.3
107.3
106.4

152.7
104.7
135.3
113.0
106.3

165.2
108.7
141.2
112.7
108.5

162.6
107.5
136.8
112.8
107.1

154.6
105.7
135.2
110.4
103.9

158.5
109.2
145.5
110.6
109.5

165.4
102.6
154.5
110.4
112.1

166.5
105.4
148.6
114.5
115.5

181.3
112.7
144.8
126.2
113.3

145.2 168.1 171.3 217.4 229.2 239.4 231.8 227.2 234.5 252.9 297.3 307.9
103.7 107.0 107.8 108.1 108.1 125.5 118.0 117.5 132.9 151.2 155.2 134.7
116.8 115.9 119.1 121.1 120.9 123.0 117.6 121.6 118.0 100.3 104.3 113.1

Total....................... 109.0 110.5 111.4 113.8 120.0 126.2 120.8 121.2 124.2 127.4 134.4 135.5

DENMARK.
INDEX NUMBERS OF THE STATE STATISTICAL BUREAU.
HISTORY AND PUBLICATION.

This series of index numbers is based on the values of Danish
imports and exports. It was first compiled in 1907 by Michael
Koefoed, chief statistician of the State Statistical Bureau of Denmark,




INDEX NUMBERS-— DENMARK.

187

and since then has been published annually in the trade statistics
of that country,1 with a brief advance statement of it appearing in
the journal of the statistical office.2 It covers a period extending
from 1876 down to the present time.
SOURCE OF QUOTATIONS.

The index numbers are computed from average annual prices
reported by various corporations, public authorities, and a consid­
erable number of private business houses, upon the basis of which
the customs officials determine the value of imported and exported
commodities. For the four varieties of grains included in the index,
the figures taken are the official Government prices.
BASE PERIOD

The decade 1891-1900 constitutes the base period used in the com­
putation of the annual index numbers.
NUMBER AND GROUPING OF COMMODITIES.

From a list of about 100 commodities entering into the import and
export trade of Denmark, there were selected for inclusion in the
index 38 of the more important ones classified into three groups,
as follows:
Group I. —Fats, oleomargarine, wheat flour, flaxseed, copra and palm kernel, rice,
coffee (green), cocoa bean, tobacco (raw), cotton, cotton yam (undyed), tallow,
copper (including brass, copper plates, and tin), and petroleum—in all, 14 commodities.
Group I I . —Horses, eggs, salt herring, wheat, potatoes, wool, wool yam (undyed),
hides (raw), boots and shoes (not including those with silk tops), paper (writing and
print), firewood bricks—in all, 12 commodities.
Group I I I .—Salt pork, meat (other than pork), butter, rye, barley, oats, maize, oilmeal cakes, sugar (3 articles or more—granulated, whiter than Dutch standard No. 18;
rock sugar, etc.; also granulated, whiter than No. 9), lumber (rough, for ships, etc.),
coal, bar and hoop iron—in all 12 commodities.

It is not possible to ascertain with any certainty the number and
variety of articles included. The classification used in the adminis­
tration of the customs laws determines the nature of the commodi­
ties which enter into this index number. A somewhat arbitrary
method of combining articles has, therefore, been adopted. Thus,
the articles coming under the single designation of "boots and shoes”
apparently include all boots and shoes except those with silk tops;
“ sugar” includes two separate items in the tariff schedule and forms
in reality three or more articles, while “ paper” includes two kinds
(writing and print) made up of various qualities combined for the
purposes of collecting the customs duty.
1 Danmarks vareindf^rsel og -udfjirseli aaret 1906-1912. Udgivet afdet Statistiske departementet. Copen­
hagen, 1907-1913. (Danmarks Statistisk Tabelvaerk. 5. rsekke, Litra D.)
2 Statistiske efterretninger, udgivet af det Statistiske departementet. Copenhagen, 1909-1914.




188

BULLETIN OF THE BUREAU OF LABOR STATISTICS.
WEIGHTING.

The system of weighting is unique. The commodities have been
placed in three distinct groups, as already noted, and these three
groups in their numerical order have been given the relative impor­
tance in the total index of 1, 2, and 3, respectively. No state­
ment is made as to the reason why certain commodities were thrown
into any particular group. An examination of the grouping of the
commodities, however, leads to the inference that they were thrown
into one group or the other on the basis of their relative importance
in consumption.
TABLE OF RESULTS.

No group index numbers are given, only a general index for all 38
commodities being published. This table follows:
IN D E X NUMBER OF W HOLESALE PRICES OF 38 IM PORTANT ARTICLES ENTERING
INTO THE IMPORT AND E X P O R T TRAD E OF DENM ARK, 1876 TO 1913.
[Source: Danmarks vareindfjzfrsel og-udf0rsel i aaret 1912. Udgivet af det Statistiske departementet.
Copenhagen, 1913: Pt. 2, p. 7*.]
(Base period, 1891-1900=100.)

Year.

Index
number.

Year.

Index
number.

Year.

Index
number.

1876.................
1877.................
1878.................
1879.................
1880.................

145
135
122
120
128

1886...............
1887...............
1888...............
1889...............
1890
.........

101
99
105
109
109

1896...............
1897...............
1898...............
1899...............
1900...............

93
95
99
105
110

1881.................
1882.................
1883.................
1884.................
1885.................

129
127
126
120
109

1891...............
1892...............
1893...............
1894...............
1895...............

112
101
100
94
92

1901...............
1902...............
1903...............
1904...............
1905...............

106
108
105
107
110

Index
number.

Year.

1906...............
1907...............
1908...............
1909...............

114
118
113
115

1910...............
1911...............
1912...............
1913...............

120
123
130
1129

i Statistiske efterretninger, 6te aargang, 1914. Utgivet af det Statistiske departementet.
1914, p. 53.

Copenhagen,

FRANCE.
INDEX NUMBERS OF ANNUAIRE STATISTIQUE DE LA FRANCE.
PUBLICATION.

The statistical annual (a n n v m r e sta tis tig u e d e la F r a n c e ) pub­
lished by the general statistical office of France {s ta tis tig u e g en era te
d e la F r a n c e ) in Paris, under the direction of the Ministry of Labor
and Social Welfare, contains index numbers for a group composed
of certain food commodities and for a second group comprising other
commodities, such as mineral products, textiles, hides, oils, etc.,
for the years from 1857 to the present time.
In addition to these there are now shown in the report, for pur­
pose of comparison, Sauerbeck’s index number for the United Kingdom, as published in the Journal of the Royal Statistical Society,
the index number for Hamburg, Germany, based on import values,
the index numbers for the United States, published by the Senate
Committee on Finance and by the Bureau of Labor Statistics, and



INDEX NUMBERS---- FRANCE.

189

the one compiled by Jules Domergue for France and published in
La Reforme ficonomique.
HISTORY.

Index numbers were first published in the Annuaire Statistique of
1904 (page 151*). Previous to this date average wholesale prices
for certain food commodities, for fodder, and for fuels had been shown,
but no totals were made for these average prices, each of which rep­
resented data for one year. The prefatory note to the table ap­
pearing in the 1904 report states that the index numbers for France
are based on the import values of 43 articles, the period 1867 to 1877
being taken as the base. In this table all commodities are divided
into two classes—foodstuffs and miscellaneous materials. An index
number is also given for all articles combined.
In the preparation of the 1907 report the period 1891-1900 was
adopted as the base instead of the years 1867-1877 and, in accord­
ance with this change, new index numbers were computed for all
preceding years.1 In 1912 a further change was made by the sub­
stitution for the years since 1905 of index numbers based on the aver­
age annual prices of 45 articles in interior markets of the country
instead of the import values of 43 articles, as in*former reports.2
SOURCE OF QUOTATIONS.

The index numbers for years prior to 1906 are based on data
published by the customs administration showing the values of
imports. These values were fixed by the board of appraisers {c o m ­
m is s io n d es v a le u r s e n d o u a n e ).
Beginning with 1906, as has been
stated, the index numbers are computed from the average yearly
prices of the different articles in interior markets of France. These
average prices are compiled mainly from records of transactions on
the Paris Bourse and from periodicals.3
BASE PERIOD

Prior to the 1907 report the base period used was that of 1867-1877.
In the 1907 report, as already stated, the base period was changed to
1891-1900 and recomputations of the index number for previous years
were made.
PRICES: H O W SHOW N AND COMPUTED.

The prices shown in the reports are in all cases averages of those
prevailing during the year. For years prior to 1906 these averages
are based on values assigned to imported articles by officials of the
customs service. The figures for years beginning with 1906 repre­
sent in each case the average of 12 monthly quotations in selected
interior markets.
1 Annuaire Statistique de la France, 1907, p. vii.




2Idem, 1912, p. 223*.

3 Idem, pp. 88*, 89*.

190

BULLETIN OE THE BUREAU OF LABOR STATISTICS.
NUMBER AND CLASS OF COMMODITIES.

The index numbers for the years from 1857 to 1905, inclusive, are
computed on the import values of 43 articles, while those for years
since 1905 are based on the market prices of 45 articles. Both raw
and manufactured commodities are included, the former predomi­
nating.
DESCRIPTION AND GROUPING OF COMMODITIES.

The commodities for which average yearly prices are published in
the Annuaire Statistique are shown in the following list appearing in
the report for 1912 (pages 88* and 89*):
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.

Wheat.
Wheat flour
Bye.
Barley.
Oats.
Maize (com).
Potatoes.
Bice.
Beef (Villette).
Veal (Villette).
Mutton (Villette).
Pork (Villette).
Beef (Hailes Centrales).
Veal (Hailes Centrales).
Mutton (Hailes Centrales).
Pork (Hailes Centrales).
Salt meats.
Butter.
Cheese (soft).
Cheese (dry).
Sugar (white, No. 3).
Sugar (refined, good quality).
Coffee.
Cocoa.
Bar iron (No. 2).
Cast iron (pipes).

27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.

Cast iron (columns).
Cast iron (plates).
Copper (bars).
Tin.
Lead.
Zinc.
Coal.
Cotton.
Flax (raw).
Hemp.
Jute.
Wool.
Silk (raw).
Hides (cattle).
Hides (horses).
Tallow.
Bapeseed oil.
Linseed oil.
Alcohol.
Petroleum (refined)
Soda (carbonate).
Soda (nitrate).
Indigo.
Timber (Bussian fir).
Timber (Austrian oak).
Bubber (Para, fine).

It is stated on page 223* of the 1912 report that since 1905 the index
numbers are computed on 45 of the above-named articles. It is not
shown which articles are not included.1 As previously stated, the
commodities are arranged in three groups: Foodstuffs, miscellaneous
articles other than food, and all commodities combined. No descrip­
tion of the articles appears in direct connection with the index num­
bers as published in the Annuaire Statistique.
SUBSTITUTIONS AND ADDITIONS.

Except for the changes made in the preparation of the 1912 report,
no additions to the list of articles or substitutions of one grade or
1 See, however, page 191 of this bulletin for list of articles published in the Bulletin dela Statistique
G6n6rale de la France, October, 1912.




191

INDEX NUMBERS---- FRANCE.

quality of an article for another have been made, so far as the printed
information discloses.
INTERPOLATION.

No prices have been interpolated, as far as can be ascertained from
the reports.
WEIGHTING.

The index numbers are unweighted.
TESTING.

Other than the arrangement by which the index numbers are ex­
hibited in comparison with other index numbers, by years, no testing
as to accuracy of results is apparent from the information at hand.
TABLES OF RESULTS.

The following table, showing the variation in the index number by
years, from 1857 to 1913, inclusive, is reproduced from the Annuaire
Statistique for 1912 (XXXII® volume, page 224*):
FLUCTUATION IN W HOLESALE PRICES, 1857 TO 1913, B Y YEARS.

Year.

1857....
1858....
1859....
I860....
1861....
1862....
1863....
1864....
1865....
1866....
1867....
1868....
1869....
1870....
1871....
1872....
1873....
1874....
1875;...

Food.

Miscel­
lane­
ous
com­
modi­
ties.

136
117
121
130
141
131
125
122
118
127
131
137
133
137
148
138
141
136
130

195
180
176
183
170
179
185
184
170
167
157
155
153
157
157
176
173
155
153

Total.

169
152
152
160
157
158
159
157
147
149
146
147
144
148
153
159
159
147
143

Year.

1876....
1877....
1878....
1879....
1880....
1881....
1882....
1883....
1884....
1885....
1886....
1887....
1888....
1889....
1890....
1891....
1892....
1893....
1894....

Food.

Miscel­
lane­
ous
com­
modi­
ties.

135
144
135
137
136
133
130
122
109
110
105
100
104
107
105
109
109
110
104

151
146
132
125
130
128
125
122
114
110
106
104
109
113
115
110
103
99
90

Total.

144
145
133
130
133
130
127
122
112
110
106
102
107
111
111
109
106
104
96

Year.

1895....
1896....
1897....
1898....
1899....
1900....
1901....
1902....
1903....
1904....
1905....
1906....
1907....
1908....
1909....
1910....
1911....
1912....
1913....

Food.

97
90
94
98
95
95
97
95
95
95
102
100
107
106
106
111
122
127
119

Miscel­
lane­
ous
com­ Total.
modi­
ties.
93
91
90
93
110
121
111
109
110
110
115
128
132
115
117
127
128
134
136

94
91
92
95
103
110
105
103
104
103
109
115
121
112
112
120
126
131
128

An index number apparently based on the same 45 articles that are
included in the Annuaire Statistique index is published in the quar­
terly bulletin which is also issued by the General Statistical Office of
France (Bulletin de la statistique generale de la France). The list
of articles as published in the bulletin for October, 1912, pages 22 to
25, is as follows: Wheat, wheat flour, rye, barley, oats, maize, potatoes,
rice, beef (2 kinds), mutton (2 kinds), pork, salt meats, butter, cheese,
raw sugar, refined sugar, coffee, cocoa, cast iron, bar iron, copper, tin,
lead, zinc, coal, cotton, flax, hemp, jute, wool, silk, salted hides, skins,
rapeseed oil, linseed oil, alcohol (90°), petroleum, soda carbonate,
soda nitrate, indigo, lumber, and rubber.



192

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

The publication of this index number began with the April, 1912,
issue of the bulletin and has been continued in each subsequent issue.
For purpose of comparison the index numbers of the R6forme ficonomique, those of Sauerbeck, and those published in the Economist
are shown in connection with it, all four being reduced to the common
base period 1901-1910. The following table showing the manner in
which the information is presented has been reproduced from the
bulletin of January, 1914:
TABLE SHOWING FOUR IN D E X NUMBERS FOR PURPOSES OF COMPARISON.

France.
Year and month or quarter.

Statis­
tique
G&ierale.

England.

RSforme
Economique.

Sauerbeck.

The
Economist.

1913
December...........................................................................
November..........................................................................
October..............................................................................

113.9
114.7
114.4

114.0
115.0
116.0

114.2
113.4
115.1

111.8
113.5
114.5

Fourth quarter..................................................................
Third quarter....................................................................
Second quarter..................................................................
First quarter.....................................................................

114.3
115.1
116.7
116.2

115.0
114.5
115.7
117.2

114.2
115.7
116.2
117.6

113.3
115.0
115.0
116.1

Entire year..............................................................

115.6

115.6

115.9

114.8

1912
Fourth quarter..................................................................
Third quarter....................................................................
Second quarter..................................................................
First quarter....................................................................

116.7
118.3
119.7
116.6

116.6
117.2
119.5
116.1

116.8
117.6
116.1
113.0

116.7
116.9
114.9
114.5

Entire year..............................................................

117.8

117.4

115.9

115.8

INDEX NUMBERS OF THE STATISTIQUE GENERALE DE LA FRANCE.
PUBLICATION AND HISTORY.

In 1911 the general statistical office (statistique generate) of the
French Ministry of Labor and Social Welfare published a volume
devoted exclusively to the subject of wages and the cost of living in
France at different epochs. Under the cost of living topic is included
a study of wholesale prices, contract prices, and retail prices, with
index numbers for each class. Index numbers are also given for
wages and for family budgets.
SOURCE OF QUOTATIONS.

The prices published in connection with the data on the cost of
living relate to the city of Paris. They were obtained from various
sources and may be classified as follows:
1.
Wholesale prices (current prices published by the Bourse de
Commerce; import prices, as fixed by the board of appraisers of cus­
toms; prices of various articles in the municipal markets of Hailes
Centrales; and meat prices in the market of Villette).




INDEX NUMBERS---- FRANCE.

193

2. Contract prices (prices paid for large quantities of supplies fur­
nished on the basis of awards on competitive bids to the Department
of Public Aid and to the Lycee Louis-le-Grand).
3. Retail prices (prices charged by the company stores of two rail­
road companies, by cooperative stores, and by a few large groceries).
BASE PERIOD.

In the computation of the index numbers the average of the prices
for the years 1891-1900 was taken as the base, or 100.
PRICES: H O W SHOW N AND COMPUTED.

Only annual prices are shown in the various tables included in the
report.
NUMBER AND CLASS OF COMMODITIES.

Table 2, on pages 44 and 45 of the report, contains 10 commodities
for which index numbers are given for the years 1880 to 1909, in­
clusive. The articles are as follows:
Bread.
Butter.
Cheese.
Potatoes.
Bice.

Oil (edible).
Wine (ordinary).
Sugar, refined.
Coal.
Oil (illuminating).

For each of these articles separate index numbers are given for
wholesale or import prices, for the prices paid by the Department
of Public Aid, and for retail prices.
A continuation of this table on page 46 covers the period from 1867
to 1910, inclusive, and contains index numbers for the following
articles:
Beef (steer).1
Veal.1
Mutton.1
Pork.1

Beef (cow).2
Fresh meats of all qualities.5
Bacon.4
Lard.4

On page 47 of the report is found a table of index numbers for the
years 1875 to 1910, inclusive, in which the following articles are
included:
Milk.
Eggs.
Coffee.
Salt.

Vinegar.
Candles.
Petroleum (refined).

1 Based on prices in the markets of Villette and Hailes Centrales, respectively.
2 Based on prices in market of Villette.
8 Based on import prices and those paid by the Department of Public Aid, and by the Lyc6e Louis-leGrand, respectively.
* Based on prices paid by the Department of Public Aid and on retail prices, respectively.

94261°— Bull. 173— 15-------13




194

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

For all of these articles except eggs index numbers, based on import
prices, are given in addition to those based on retail prices or prices
paid by the Department of Public Aid, or both. In the case of eggs,
the index numbers are computed from retail prices and from those
paid by the Department of Public Aid and by the Lyc6e Louis-leGrand, respectively. Prices paid by the last-named institution also
furnish the basis for index numbers shown for vinegar and salt.
TABLES OP RESULTS.

In the following table, which has been compiled from data pub­
lished on pages 44 and 45 of the report, are shown index numbers
covering the period from 1880 to 1909, inclusive, computed from the
wholesale or import prices of 10 commodities in common use. The
figures for refined sugar and illuminating oil are based on wholesale
prices of the Bourse de Commerce of Paris. Those for wine are com­
puted from prices published annually by the minister of finance in
the Bulletin de Statistique et de Legislation. The index numbers
for the remaining articles are based on import prices. As stated in
a preceding paragraph, the average of the prices for 1891-1900, taken
as 100, constitutes the base.
R E LA TIV E PRICES OF COMMODITIES, 1880 TO 1909.

Year.

1880...........................
1881...........................
1882...........................
1883...........................
1884...........................
1885...........................
1886...........................
1887...........................
1888...........................
1889...........................
1890...........................
1891...........................
1892...........................
1893...........................
1894...........................
1895...........................
1896...........................
1897...........................
1898...........................
1899...........................
1900...........................
1901...........................
1902...........................
1903...........................
1904...........................
1905...........................
1906...........................
1907...........................
1908...........................
1909...........................

Bread. Butter. Cheese. Pota­
toes.
137
135
134
114
105
101
107
99
112
109
100
106
101
84
78
78
100
120
121
105
105
105
106
107
110
120
120
129
129
130

100
100
98
98
95
91
89
82
82
82
82
82
100
107
98
96
107
96
102
104
107
111
107
100
102
104
111
107
107
105

117
117
121
121
121
110
102
99
102
110
113
110
100
104
97
98
98
98
99
99
104
110
113
104
106
110
88
120
120
128

150
142
150
145
133
150
100
92
125
92
118
117
83
100
100
67
83
117
100
117
117
167
150
167
150
167
183
167
199
199

Rice.

132
129
118
122
122
125
125
125
107
125
125
111
125
129
118
100
82
93
90
79
82
82
82
75
80
84
86
89
104
100

Oil,
edible.

169
169
150
166
184
183
183
175
161
149
146
134
138
124
115
105
88
78
73
73
73
101
121
128
128
131
142
149
181
192

Wine, Sugar,
ordi­
nary. refined.
148
156
157
144
155
155
156
140
118
122
138
116
111
90
89
122
98
92
112
99
70
56
78
109
65
62
71
66
62
70

133
114
106
101
100
100
92
94
102
111
102
103
101
109
101
95
96
92
99
101
100
96
91
78
50
63
56
55
58
60

Coal.

98
100
100
82
82
77
71
71
82
120
122
93
93
88
90
88
90
95
105
113
147
131
111
101
93
93
112
128
117
128

Oil,
illumi­
nating.
132
133
130
156
121
109
94
97
107
120
122
123
97
98
88
91
103
100
93
89
117
110
104
90
83
88
115
141
109
106

The following table, reproduced from page 45 of the report, contains
three series of index numbers representing in each case the average
of index numbers computed for the 10 articles included in the pre­




195

INDEX NUMBERS---- FRANCE.

ceding table. In addition to index numbers based on wholesale or
import prices, similar data for contract prices paid by the Depart­
ment of Public Aid and for retail prices are given in this table.
THREE SERIES OF IN D E X NUMBERS COMPUTED FOR 10 ARTICLES, 1880 TO 1909.

Year.

Whole­
sale or
import
prices.

Contract
prices
paid
by the
Depart­
ment of
Public
Aid.

132
129
126
125
122
120
112
108
no
114
117
110
105
102
98

126
121
125
120
116
109
110
104
106
104
105
108
106
101
100

1880
...............
1881
.............
1882.........................
1883
............
1884 ........................
1885
............
1886
1887
___
1888
..................
1889
.................
1890
...............
1891
............
1892
............
1893
..............
1894
...............

Retail
prices.

121
111
117
105
106
101
94
104
105
103
104
108
103
105

Year.

1895...........................
1896...........................
1897...........................
1898...........................
1899...........................
1900...........................
1901........................
1902...........................
1903...........................
1904...........................
1905...........................
1906...........................
1907...........................
1908...........................
1909...........................

Whole­
sale or
import
prices.

Contract
prices
paid
by the
Depart­
ment of
Public
Aid.

94
94
98
99
98
102
107
107
106
99
102
109
115
122
122

96
96
99
98
99
99
99
95
96
90
85
89
95
96
93

Retail
prices.

96
93
93
99
100
97
99
95
93
94
94
95
95
99
101

INDEX NUMBERS OF LA REFORME ECONOMIQUE.
PUBLICATION.

This series of index numbers is based on wholesale prices in France
and is published weekly in La Reforme Economique, a journal of
social and political economy, of which Jules Domergue is the editor.
HISTORY.

For a number of years prior to 1900 comparative prices for a
varying number of commodities were published in this journal. In
the earlier numbers the increase or decrease in prices of seven com­
modities as compared with the primary period (1890) was shown by
means of a chart. In 1894 prices for corresponding dates in 1892,
1893, and 1894 were given, and for some articles an average monthly
price was computed. In 1896 a weekly table of prices was presented
for the first time, and on May 9, 1897, a series of tables was begun
showing the average monthly prices of all the commodities used in
the computation. Beginning with January, 1899, an annual average
price was computed and published for the years 1890 to 1898, inclu­
sive, and in addition current prices were compared with these by
means of annual average prices. No comparison was made by means
of index numbers, however, until 1900, when the method of presenting
the variation in prices for stated periods was changed by showing in
addition to the average price of each commodity the simple percent­
ages of increase or decrease in the various commodities.




196

BULLETIN OP THE BUREAU OF LABOR STATISTICS.
SOURCE OF QUOTATIONS.

Actual commercial transactions are sought for in the collection of
price data. The prices quoted are said to be those obtained from the
records of licensed brokers (courtiers assermentes) and private
brokers (courtiers libres) in different parts of Paris and in the depart­
ments; official quotations of the Department of Agriculture; prices
obtained in the municipal markets of La Villette and Hailes Cen­
trales; quotations on importations as stated in the records of the
Government warehouses in Paris and elsewhere; and official quota­
tions of the price of bread furnished by the prefecture of the Seine.
BASE PERIOD.

The year 1890 is used as the base period. Nothing is stated in
connection with the figures, so far as can be ascertained, as to why
this year was chosen in preference to any other year or period of
years.
PRICES: H O W SHOW N AND COMPUTED.

As has been stated, no comparison of prices was made by means of
percentages until 1900. In that year, after noting the principal
objections to an index number representative of the combined prices
of all commodities for which prices are quoted, it was decided to
construct an index for each article and an index for each group of
articles, in addition to the index for the entire list. The prices used
in the calculations were the average annual prices already published
in La Reforme ficonomique for the period 1890-1895, the quarterly
average prices for the years 1896-1898, and the average monthly
quotations for the year 1899.
A special presentation of the price of cotton is made in the issue of
September 23, 1900, showing for five grades the range of prices of this
commodity from 1875 to September, 1900, with an index based on
the price in 1875 (equal to 100). In the supplement to the issue of
November 24, 1901, the average prices of wheat, sugar, wine, and
alcohol for the periods 1884-1891 and 1893-1900 have been computed
and the divergence in price between the two periods is shown. The
prices for the year 1892 are not a factor in either period.
NUMBER AND CLASS OF COMMODITIES.

In tracing this index number through the period elapsing since its
inception, there is great difficulty in determining the number and
class of commodities which have been included in the compilation.
Between 1890 and 1899 prices and index numbers were shown for
from 40 to 56 or even more articles, according to the system of count­
ing and classification adopted. Thus the compiler evidently com­
bined four kinds of meat (beef, pork, mutton, and veal) into one
commodity, while from two to four kinds of wool, silk, or cotton were




INDEX NUMBERS---- FRANCE.

197

each sometimes given a separate index number. From 1900 to date
it would appear that approximately 48 articles make up the series of
index percentages, although average annual wholesale prices are
given for several additional articles.1 After 1904 a statement printed
on the cover page and called “ La Thermometre des Affaires en
France” shows a separate index number for only 21 leading commodi­
ties, although the general percentage index number includes addi­
tional articles, as may be verified by actual arithmetical test.
DESCRIPTION AND GROUPING OF COMMODITIES.

The classification of the various commodities has differed from time
to time. The first summary table, presented in the issue of January
14, 1900, page 67, shows index percentages for 56 commodities (43 if
certain grades of textiles and bar and structural iron are not considered
separate commodities), classified in five main groups as follows:
F o o d p r o d u c ts .—Wheat, flour, rye, barley, meats, wine, sugar,
alcohol, coffee, butter, sirup, edible starches (fecula),2 oleomargarine,
tallow, lard, cocoa, rice, and bread.
T e x t ile s — Silk (2 grades), wool (raw and yarn), linen (raw and
thread), cotton (raw and spun), hemp, jute.
A g r ic u lt u r a l p r o d u c ts .—Hides (raw), leather, oats, maize, fodder,
fatty acids (3), rapeseed oil, linseed oil.
M in e r a ls a n d m e ta ls .—Coal, petroleum, copper, tin, zinc, lead, steel,
iron (2 grades), sheet iron.
M is c e lla n e o u s .—Rubber, sulphuric acid, hydrochloric acid, chloride
of lime, carbonate of soda, sal soda, sulphate of ammonia, super­
phosphates.
In the issue of April 14, 1901, average annual wholesale prices are
presented for all of these 56 commodities, while relative prices are
shown for only 43 of them, divided into five groups a3 above, except
that the fifth group is termed “ Chemicals and fertilizers” but contains
the same commodities as the group which is designated above as
“ Miscellaneous” commodities. Certain interchanges were also made
as between “ Food products” and “ Agricultural products.” This
classification was continued until January, 1902, so as to include
indexes for the year 1901.
In 1904 the list of commodities for which separate relative prices
were presented was reduced to 21 commodities as follows: Wheat,
meat, wine, sugar, alcohol, coffee, coal, petroleum, copper, tin, zinc,
lead, steel, iron, silk, wool, flax, cotton, nitrate of soda, superphos­
phates, and sulphuric acid.
Beginning with the issue of December 10, 1905, a change was made
in the form of classification, the articles being grouped as (1) agricul­
1 Annuaire Statistique de la France, 1912, p. 223*.
2 The French term “ fecule” includes such articles as potato flour, tapioca, sago, arrowroot flour, etc.




198

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

tural products and (2) industrial products. Under this classification
average annual wholesale prices are presented for 21 articles in the
first group, if forage is counted as one commodity and not as two, and
wines as one instead of three, meats as one instead of four (beef, pork,
veal, and mutton), and fatty acids as a single commodity instead of
three; whereas if all these subdivisions are counted as separate com­
modities, the number would appear as 29. In the second group, that
of industrial products, there are 28 or 40 commodities, according to
the system of counting adopted.
For the above commodities general index percentages are given
only for both groups combined, and not for each group separately or
for each commodity. This plan of presentation has continued since
1905, although the commodities differ a trifle from year to year. As
already stated a separate index number for 21 commodities is shown
in “ Le Thermometre,” which forms part of the cover page.
SUBSTITUTIONS AND ADDITIONS.

In the issue of La R6forme Uconomique for January 15, 1899, the
average prices of raw and spun silk of several grades was included for
the period of 1890-1898, and on February 12 of the same year average
prices of iron (two grades), steel rails, sheet iron, bar copper, and
tin (Banka) were added for the years 1890 to that date.
Two further groups were added on April 2,1899. The first of these
included wine, coffee (Santos, good average), hops (Burgundy),
nitrate of soda, sulphate of ammonia, and superphosphates; the sec­
ond, sulphuric acid 66°, hydrochloric acid 20-21°, chloride of lime
105-110°, carbonate of soda 90-92°, and caustic soda 80°, for manu­
facturing and laundry purposes. Quotations for hides, raw (three
quotations), and tanned (four quotations) were added on April 23,
1899. The index for tanned leather does not appear after 1900, nor
do those for cocoa, rice, rubber, oleomargarine, bread, lard, and
fatty acids.
During the year 1902 percentage relatives for two articles were
dropped from the list, and in the first issue of 1903 two more indexes
were dropped. After 1905 no quotations appear for the following
classes of wine: Alicante, Huelva, Aragon, Valence, and Haw.
Apparently no adjustment of previous percentages have been made.
No further change in the list of articles or method of presenting the
variation of prices appears to have been made since 1905.
Oleomargarine did not figure in the index series until 1896, and it
was dropped in 1900. Up to 1900 separate relative prices were cal­
culated for two kinds of sugar, raw and refined; after that date only




INDEX NUMBERS— FRANCE.

199

one index is shown, although wholesale prices are quoted for both
kinds. Apparently, however, the index percentage is based on the
average of the prices shown for each kind. Similar changes have
taken place in the coal index. Thus, four index percentages were
presented up to 1900, but after that date only one is shown, which is
based on the average wholesale prices of all four kinds.
In general, when any changes were made in the number or classes
of commodities the index percentages were recomputed back to the
base year, 1890.
INTERPOLATION.

No method of price interpolation has been resorted to, so far as
can be determined from the information published in La Reforme
fCconomique.
WEIGHTING.

No scientific method of weighting has been used, the arithmetic
average alone being employed in the construction of the index num­
bers. The method of calculating the yearly general index for groups
of commodities and for all commodities seems to be as follows: For
the years already covered by the reports on wholesale prices, the sum
of the average prices for the year of the different articles was divided
by that of the basic year (1890). Thereafter, average monthly prices
were obtained by getting the average of the weekly quotations made
during the month as published in La R6forme Economlque, and from
these average monthly prices the yearly average price was computed,
the yearly general index then being computed as before. The rela­
tives for each article (when given), the index for each group, and the
general index are in all cases simple percentages based on the prices
of corresponding items for the corresponding period in 1890.
TESTING.

In the earlier years Sauerbeck’s table of index numbers for Eng­
land was occasionally given for comparison, and in the later issues it
has appeared quite regularly. No other comparisons are made.
TABLE OF RESULTS.

The following table has been compiled from the numbers of La
Reforme ficonomique published during the years 1892-1913. The
items extending throughout the entire series of years are identical
with those contained in the summary table showing index percentages
for 21 commodities published regularly as a part of “ La Thermometre
des Affaires en France,” which forms the cover page of current
numbers.




200

BULLETIN OP TH E BUREAU OP LABOR STATISTICS.
IN D E X NUMBERS COMPUTED FROM
[Source: La R6,
(Base period,

Mar­
gin­
al
num­
ber.

Commodity.

1891

1892

1893

1894

1895

1896

1897

1898

109
109
107
117
93
92

92
95
97
101
79
73

83
84
89
84
101
90

76
78
84
77
102
81

75
79
85
66
81
71

75
77
82
71
79
74

100
97
97
96
86
81

101
102
103
101
101
89

Beef...................................................................
Veal...................................................................
Mutton..............................................................

103
98
105
101

94
98
101
100

94
103
94
99

100
105
103
118

95
103
104
117

76
89
97
78

78
90
94
80

70
85
90
98

1
2
3
4
5
6

Wheat................................................................

7
8
9
10
11

Barley...............................................................

Average, meats......................................

102

98

97

106

105

85

87

86

Rawhides.........................................................
Leather..............................................................
Wines:
14
Filtered......................................................
French.........................................................
15
Algerian......................................................
16
Foreign.......................................................
17

99
104

81
99

82
98

77
93

107
111

97
105

96
101

110
102

97
103
84
92

91
93
78
97

97
87
68
85

82
76
62
81

77
67
74
85

88
79
88
89

87
65
85
91

87
85
85
95

12
13

Average, w ines.......................................

91

94

88

80

76

85

81

89

Sugar:
Beet root, raw............................................
19
Beet root, refined.......................................
20

107
101

113
99

123
107

94
99

82
94

88
95

78
91

90
98

21

Average, sugar........................................

104

106

115

96

88

91

84

94

22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41

Alcohol..............................................................
Sirups................................................................
Starches (edible)1.............................................
Butter...............................................................
Oleomargarine..................................................
Rapeseed o il.....................................................
Linseed oil.........................................................
Tallow...............................................................

116
102
101
102

132
101
96
124

123
86
82
133

94
85
75
103

87
84
71
89

101
94
103
99
104
99
76
101
104
88
104
112
143
100
94

80
83
99
108
100
90
62
102
134
85
102
96
124
104
81

81
89
116
139
119
97
73
156
178
92
133
85
112
104
72

72
87
96
121
97
91
62
117
151
90
120
76
117
104
71

74
86
86
101
90
88
88
99
108
90
133
72
112
114
107

86
91
74
85
85
84
82
76
70
81
77
130
76
107
67
160
71
80
119
100

106
91
81
87
99
81
68
73
63
77
64
85
81
115
44
200
80
76
140
93

128
94
93
97
104
76
72
81
87
85
70
84
91
97
35
275
90
90
157
103

87
112
95
98
104

77
86
79
78
97

71
77
74
72
93

71
82
72
73
91

63
73
71
75
90

64
70
71
76
89

66
71
72
77
90

77
119
73
77
92

80

76

74

75

88

76
68i
63!
7S►
87
96>1
761

83
64
71
84t
88!
96i
781

86i
91
66i
75
75i
84
93! 108
97
97
95
9S>
8C>1 78

18

42
43
44
45
46

Stearic acid.......................................................
Oleic acid..........................................................
Glycerin.............................................................
H ay....................................................................

Rubber..............................................................
Petroleum.........................................................
Coal:
France.........................................................
Cardiff.........................................................
Mons............................................................
Charleroi......................................................
Saarbrucken...............................................

47

Average, coal..........................................

99

83

77

48
49
50
51
52
53
54

Copper................................................................
T in.....................................................................

90
97
99
93
100
100
84

81
99
90
80
94
100
83

78
92
75
74
91
10C
81

71
74
67
Lead...................................................................
72!
Iron (structural and merchant bar).................
93!
9S>
Sheet iron...........................................................
77
l The French term “ f6cule” includes such articles as potato flour, tapioca, sago,




arrowroot flour, etc.

201

INDEX NUMBERS---- FRANCE.
AVERAGE ANNUAL PRICES: 1891 TO 1913.
forme Economique.]
1890=100.)

1900

79
77
85.

79
77

87

90
83

1901

1902

1903

90

76
“ 93*

102

81

1904

87

1905

1906

93

92

1907

1908

94

1909

95

1910

102

1911

103

1912

1913

114

Mar­
gin­
al
num­
ber.

110

94.
91.
84.

73.
87 .
94.
104.
90

~1U~
103 .

9

87

90

91

90

100

96

100

101

100

n r 112 IS"

13

14
15
16
17

87 .
94 .

88.

102 .

93

10
11
12

85

71

90

82

61

78

120

129

122

123

18

73

"T18

20
21
22

141

40
41

19

91

94

117
96
95
105
130.
73
84
96
85.
95 .
84.
94.
69
97
33
243 .
94.
97
170 .
114

1ST

74

104
79
103

79
91
67
92

129
106

'*90*
126
104

85.
128.
119.

110

94
123
36

92 .
109.
34

45

109
'iih '

74
T l9~

.70

64

77

~122~~126~

118 "U 9 _

74

122

93

90

~U5~ ~145~ 173 "l65_

74.

32

121 133 .
'i02* 'ioe*

41

44

43

115

107

113

40

124

128

128

115

67

78

111

133

100.

42
43
44
45
46

101 .

79.

88.
97.
93
129
130

100

126
131
128
107

23
24
25
26
27
28
29
30
31
32
33
34
35
36
37

129

100

~128 ~ tst 94
125
140
126
76
90
79
108
140
97
152
110
99
144
115
105
127




102 101
loi"" 104
133
91
100

98
108

133
97
103
90
108

112

129

124 ~155~ ”l59~
182
149
188
110
118
107
144
116
160
95
115
120
111 118 118

121

113

118

106 ~104~ 102
140
161
141
91
104
114
118
113
114
103
105
111
105

101

102

119

127

47

~~124
213
106
161
125
119
124
118

48
49
50
51
52
53
54

122

"loo" ” l29
198 221
120
113
121 152
105
103

202

BULLETIN 03? T H E BUREAU OP LABOR STATISTICS.
INDEX NUMBERS COMPUTED FROM AVERAGE

Mar­
gin­
al
num­
ber.

Commodity.

Silk:
Organzine, French, first grade___
Organzine, French, second grade.
Tram, Japan.................................
Raw, first grade, C6vennes..........
Raw, second grade, Italy.............
Raw, China...................................
Canton...........................................

1891

85

87

Average, silk..
Flax, raw..................
Linen thread, No. 20.
Linen cloth................

95

100

Average, flax..

1893

1894

87

111

85
84
76
84

105
105
101
81
92

74
73
70
72
71
65
65

85

101

70

112

15T ~127 ~ ~128~

101
100
101

85

75

75

70

91
91
94

71
82
85
89

77
90
93
101

63
80
84
93

General percentages.




71

75

105
101

103 ~
105
101

101

106

103

63 ~ ~~70~
76
78
74
79
110 105
77
74

97
99
95
114
107
97
98
109

101

72
113
96
98

110
107
97
93
81

100

97
95
107
103
103
103

63
79
83
78

112

81
100

47

88
104
107
107
94
117
82
101

66~
72
70
75
71

94
100

79

86
90
89
74
99

95

99

75
75
76
75
70
75

76

90

Average, cotton.

75
75
78
75
74
77
72

72

101

Average, wool...

72
72
74
72
71
70
65

H i”

111

Cotton:
Raw......................
Warp (chaine 28)..
Woof (trame 37)...
Cloth.....................

76
76
75
74
72
65
67

75

117

85
87

1897

103 "

101
100

100

67~
70
73
85
71

Hemp..........................
Jute..............................
Sulphuric acid............
Chloride of lime..........
Carbonate of soda.......
Sal soda.......................
Nitrate of soda............
Sulphate of ammonia..
Superphosphates.........
Hydrochloric acid.......

78
78
77
78
78
67
71

106

77 ~
74
75
75
75

82

1896

119
104

77 ~
74
73
75
75

Wool, carded...........................................
Woolen yam, warp (chaine 40)..............
Woolen yam, woof (trame 60)...............
Woolen yam of 710 meters to the skein.
Merino cloth............................................
72

1892

71
82
85
80

66

_79

75

78
80
74

~45~ ~41
83
91
93
91
92
85
74
74
86 85
87
85
74
73
70
70

56
67
71
71

64
80
94
S3
73
83
83
85
74
97

99.60 94.20 97.60 89.40 84.40 82.20 83.40 87.60

203

INDEX NUMBERS---- FRANCE.
ANNUAL PRICES: 1891 TO 1913—Concluded.

1899

1900

1901

1902

91
91
96
95
94
91
95

87
87

74
73

83
82

93

87

74

82

98
97
91

160
130

175
135

145
123

95

137

147

130

98
101
100
101
88

80
93

70
79

79
88

100

88

79

86

61
72
75
83

90
99

79
95

80
89

73

98

91

87

98
90
93
82
76
84
85
101
80
96

109
105
92
91
86
88
93
101
76
96

91
107
92
99
92
92
100
96
76
96

111

1903 1904

1905

1906

1907

1908

1909

1910

1911

1912

1913

Mar­
gin­
al
num­
ber.

55
56
57
58

61
93

83

86

94

114

83

84

82

83

81

85

62
63
64

143

152

131

147

164

128

129

152

167

155

162
67
70
71

89

94|

98

107

112

93

105

112

105

97

103

72
73
74
75

76
100

102

90

100

117

115

104

118

109

105

113

92

86

80

80

97

106

106

106

109

i09

108

100

89
102
104
79

105

114

117

125

125

111

107

102

110

124

124

81

72

59

74

81

72

70

72

72

69

85

95.60 102.40 95.80 94.20 95.80 95.20 95.80 105.40 112.20 101.20 101.80 108.20 113.80 117.80 116.00




77
78
79
80
81
82
83
84
85

86
87

204

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

INDEX NUMBERS OF EMILE LEVASSEUR.
PUBLICATION.

This “ Inquiry into the price of food commodities for a period of
25 years in 70 high schools of France” was published in the Revue
ficonomique Internationale, Brussels, in May, 1909. Later in the
same year, under the title of “ Le colit de la vie,” it appeared as a sep­
arate pamphlet, which also was published by the Revue.1
HISTORY.

Toward the close of 1908 the minister of public instruction, at the
request of Mr. E. Levasseur, addressed to the principals of 70 high
schools of Paris and the Departments an inquiry concerning the
prices of certain food products and of coal as paid by the schools
since 1880.
The inquiry covered the years 1880, 1885, 1890, 1895, and each
year from 1900 to 1908, inclusive.
The schools selected, exclusive of those of Paris and its suburbs,
included some of the large and some of the small institutions in each
of the nine agricultural sections of France.
Mr. Lucien March, chief of the general statistical office (sta tistiq u e
g en er a te) of France, assisted in the work by assuming the responsi­
bility for the calculation of the index numbers from the figures
secured.
The author states that from the data received it was possible to
secure a sufficiently exact report of the variation which the prices of
commodities had undergone for a period of about 25 years.
SOURCE OF QUOTATIONS.

The prices considered are contract or semiwholesale prices (p r ix
obtained from 70 high schools. The articles are 21
in number—20 food commodities and coal.
d Ja d ju d ic a tio n )

BASE PERIOD.

The average price for the two years 1895 and 1900, taken as 100,
is used as the base. The index numbers for the period 1880 to 1908
derived from the use of this base are shown in the following table,
appearing on page 7 of the pamphlet:
Years.

Index
numbers.

1880....................... ...................... 111.9
1885....................... ...................... 104.2
1890...................... ....................... 101.4
1895.............................................. 100.2
1900....................... ...................... 99.3
1901....................... ...................... 99.8
1902............................................... 98.8

Years.

Index
numbers.

1903..................... ......................... 99.9
1904..................... ........................ 99.9
1905..................... ........................ 98.0
1906..................... ......................... 98.8
1907..................... ........................103.1
1908............................................... 106.5

i Le coftt dela vie. EnquSte sur le prix des denrees alimentaires depuis un quart de s M e dans 70
lycSes, par E. Levasseur, membre de PInstitut, administrateur du College de France.




INDEX NUMBERS---- FRANCE.

205

DESCRIPTION OF COMMODITIES.

The articles selected for which E. Levasseur computed index num­
bers are the following:
Bread.
Fresh meats (other than pork).
Fresh pork.
Smoked pork (charcuterie).
Poultry and game.
Red wine.
White wine.
Beer.1
Cider.2
Butter.
Drippings and lard.

Oil (table).
Eggs.
Milk.
Cheese.
Sugar.
Fresh fish.
Salted fish and canned fish.
Codfish.
Potatoes.
Coal.
TESTING.

Mr. Levasseur verified his index numbers by comparison with index
numbers for France, England, Germany, and the United States.
The variations as shown by the index numbers of prices in 70 high
schools were verified by comparisons with the following:
(a ) Index numbers relating to France only—
1. Statistics prepared by Mr. Lucien March, chief of the general
statistical office of France, and published in the Annuaire Statistique
de la France. These show a greater increase for all merchandise in
general than for food commodities alone (except in 1895). The index
numbers show a rapid increase since 1905.
2. Index numbers calculated by Mr. de Foville. These numbers
were based on the difference between the rates of duty, and show
prices to have been low in 1900, with an increase in all the following
years, especially since 1903.
3. Index numbers calculated by Mr. Levasseur from the duties on
32 food commodities, corresponding closely to those of the high
schools. These figures show a rapid rise in prices since 1903.
4. Index numbers calculated on the basis of current prices since
1882 by the purchasing agent of the southern railways. These index
numbers verify (or confirm) the index numbers of the high schools of
the southwest.
(b) Index numbers relating to foreign countries—
5. England—Sauerbeck’s index numbers for 45 articles of general
merchandise. These index numbers are published in the Journal of
the Royal Statistical Society. Sauerbeck’s index numbers for food
commodities show, like those of the high schools, a slight fall in prices
from 1900 to 1902 and also a marked rise in 1907 and 1908.
6. Index numbers computed in Germany for the city of Hamburg.
7. Index numbers computed by the United States Bureau of Labor
1 Included in the calculation of index numbers in 1 instance only.
2 Included in the calculation of index numbers in 3 instances only.




206

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

Statistics. These index numbers are for wholesale and retail prices,
respectively.
In conclusion the author states that “ these diverse statistics, despite
the differences of detail, confirm the statistics of the 70 high schools
and show clearly that the great changes in prices are not due to special
or local causes, but to general causes, the results of which are felt at
the time in all the great markets which are in constant commercial
communication with each other.” 1
TABLES OF RESULTS.

The following table, reproduced from page 15 of the publication,
shows the variations in the index numbers for food articles in the 70
high schools of Paris and suburbs and of the 9 agricultural sections of
France, by years, division into large and small schools being made:
IN DEX NUMBERS OF PRICES OF 20 FOOD COMMODITIES AND COAL IN 70 HIGH
SCHOOLS OF PARIS AND ITS ENVIRONS AND OF THE 9 AGRICULTURAL SECTIONS
OF FRANCE, W ITH DIVISION INTO LARGE AND SMALL SCHOOLS.
(Average o f prices for 1895 and 1900—100.)
Sections.

1880

1885

1890

1895

1900

1901

1902

1903

1904

1905

1906

1907

Paris.........................
Environs of Paris___
Northwest:
Large schools---Small schools___
North:
Large schools—
Small schools___
Northeast:
Large schools___
Small schools—
East:
Large schools___
Small schools---Southeast:
Large schools___
Small schools.. . .
South:
Large schools___
Small schools___
Southwest:
Large schools___
Small schools___
West:
Large schools___
Small schools—
Central:
Large schools—
Small schools___

115
114

112
106

99
100

101
98

99
102

99
99

97
99

100
105

99
99

95
97

96
99

98
102

99
105

117
111

111
100

105
106

100
100

100
100

96
98

96
100

99
101

103
102

93
99

95
102

95
104

104
109

111
114

98
99

102
97

100
105

100
95

100
98

98
98

99
99

98
99

97
100

96
107

99
110

102
109

111
122

106
106

103
98

101
100

99
100

99
101

98
98

97
97

97
98

95
99

96
100

103
102

107
109

108
109

103
108

101
101

100
99

100
101

97
99

99
100

99
100

101
100

99
99

99
99

102
104

106
106

105

104

100
106

100
102

100
98

100
99

100
94

99
95

101
100

99
97

98
101

101
106

106
107

108
116

102
102

95
98

94
99

98
101

95
103

93
99

96
101

90
103

90
100

88
100

99
107

98
112

106
115

102
112

98
104

101
102

99
98

101
98

100
101

98
101

98
103

96
100

98
102

104
107

105
109

108
112

99
103

102
109

99
100

101
100

105
104

102
101

104
99

102
96

98
96

97
97

101
102

108
107

111
113

101
108

101
100

104
100

96
100

97
103

93
100

98
102

98
102

95
103

102
104

104
113

104
117

P a r is ......................
Other schools:
Large schools (37)
Small schools (30)

114

109

99

100

100

101

97

100

99

94

96

98

98

110
113

103
105

100
102

100
101

99
99

99
101

98
99

100
100

99
101

96
99

97
101

101
105

105
109

Total...............

111

104

101

100

99

100

99

100

100

98

99

103

107

1908

Index numbers computed on the prices of 20 food commodities
and coal in 1908 are shown in the following table, the figures being
given separately for Paris, its suburbs, and the large and small
schools, respectively, of the 9 agricultural sections of France.2




i Le cotit de la vie, p. 11.

2 Ibid, pp. 22 and 23.

207

INDEX NUMBERS— FRANCE.

IN DEX NUMBERS OF PRICES OF 20 FOOD COMMODITIES AND COAL IN 20 GROUPS
(PARIS, ENVIRONS, LARGE AND SMALL SCHOOLS OF THE 9 AGRICULTURAL
SECTIONS OF FRANCE) IN 1908.
(Average of prices for 1895 and 1900=100.)

Sections.

Poul­
Red White Beer.
Smoked try
Bread. Fresh
pork. pork.
and wine. wine.
game.

But­
ter.

Drip­
pings
and
lard.

Fresh
meats
(exclu­
sive of
pork).

104.5
112.1

121.9

131.2
123.3

169.5
154.2

110.5
109.9

104.4

105.3
110.3

105.7

104.0
107.5

81.1

103.7
109.1

138.5

105.5
109.0

Cider.

115.3
120.2

117.8
174.4

116.4
135.6

106.9

47.2
62.2

73.1
102.1

117.6

121.6

166.0
114.2

164.1
113.2

111.7
124.0

65.1
57.9

64.7

113.3
119.2

113.1
106.3

189.6
97.1

105.8
113.0

53.3
73.3

80.0
97.8

114.5
119.8

106.2
106.1

136.8
116.8

105.7
106.2

59.3

76.3
72.9

111.1

119.9
116.6

116.5
113.7

121.1
116.3

117.6
112.9

70.4
70.1

71.6
88.1

110.1

122.0
120.7

123.8
113.5

115.7
128.1

101.3
97.9

65.2
81.2

77.7
84.5

116.1
102.8

135.2
134.6

105.2
110.9

106.7
118.7

97.8
116.0

95.7
107.7

110.7
134.7

63.7

103.6

111.5
104.4

150.6

101.1
109.3

122.1
118.4

112.8
116.6

128.6
101.3

117.1
108.7

74.3
72.6

112.1
116.1

110.5
103.1

122.0
128.3

110.1

136.5
116.7

130.2
110.3

106.5
135.5

130.8
131.3

61.2
61.8

72.0
82.2

119.3

95.2
106.7

109.5
105.2

121.5
132.5

111.0
113.1

110.5
140.1

118.6
108.2

65.0
66.7

56.4
84.6

116.7

125.0

112.0
104.1

General average. 119.0

113.0

116.5

111.3

63.3

110.0

120.3

110.7

Paris......................
Environs of Paris.,
Northwest:
Large schools..
Small schools..
North:
Large schools..
Small schools..
Northeast:
Large schools..
Small schools..
East:
Large schools..
Small schools..
Southeast:
Large schools..
Small schools..
South:
Large schools..
Small schools..
Southwest:
Large schools..
Small schools..
West:
Large schools..
Small schools.
Central:
Large schools.
Small schools.

68.1

91.5

113.8
109.4

111.1
112.6

Sections.

Oil
(table).

Paris............................
Environs of Paris.......
Northwest:
Large schools........
Small schools........
North:
Large schools........
Small schools........
Northeast:
Large schools........
Small schools........
East:
Large schools........
Small schools........
Southeast:
Large schools........
Small schools........
South:
Large schools........
Small schools........
Southwest:
Large schools........
Small schools........
West:
Large schools........
Small schools.......
Central:
Large schools........
Small schools........

91 3
112.2

110.3
107.5

95.6
100.0

108.0
108.9

59.3
60.7

71.0
98.4

108.3
87.9

116.1
127.9

112.4
96.9

116.2
111.2

61.0
61.0

98.5
108.6

93.4
114.1

108.2 I 106.5
104.7 1 104.5

114.8
114.5

117.2 i 107.2
125.0 i 99.0

107.4
105.1

120.6 !

102.9
109.3

115.3 l 100.4
123.7 | 125.0

120.6

74.2
111.6

General average.




100.2

Milk.

Cheese. Sugar.

110.6

Fresh
fish.

Salted
fish.

Cod­
fish.
81.6

127.2

112.5
118.8

Pota­
toes.
108.1
125.8

103.8
123.1

64.7
64.1

105.4
125.5

62.8
61.2

79.2
81.4

125.2
150.7

116.3

86.7
59.1

86.7
91.4

118.7
130.8

136.1

125.0

62.5
58.2

93.9

106.0
117.0

125.2

100.0

115.6 I 121.7
134.0 I 111.6

104.8
113.0

56.8

75.9
78.2

109.6
143.2

105.8

106.5
119.7
131.4

123.9

i

98.0
105.6

126.2

78.2

126.3
123.4

99.1
94.3

117.0
109.8

60.1
60.6

74.8
75.7

128.4

97.9
119.6

114.0
123.1

103.9
114.6

123.5
100.0

123.5
155.8

61.5
61.0

84.0
85.7

84.3
81.7

9.5

95.5
138.9

122.3
112.2

87.7
118.4

112.7
151.8

63.3
64.9

110.3
103.8

128.4

105.1

117.6

105.7

116.2

62.6

119.3

113.5
110.3

103.7
130.3

135.7
124.6

95.0

117.7
128.5

109.7

180.4
124.7
119.9
121.2

111.1

84.3

Coal.

122.3
117.6

130.5

112.6

109.2

107.1
148.8

156.0
128.3

109.1

115.3

107.5

110.5

144.9
129.5

208

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

GERMANY.
INDEX NUMBERS OF THE IMPERIAL STATISTICAL OFFICE.
PUBLICATION.

This series of index numbers, which represents wholesale prices of
commodities in German markets, is published yearly in the Vierteljahrshefte zur Statistik des Deutschen Reichs, a publication of the
Imperial Statistical Office, and appears regularly for each year in the
first quarter of the succeeding year.
The first report, including index numbers, was published in 1905
and covered the years 1899 to 1904. The table of index numbers
in late reports regularly covers the 10-year period ending with the
date of the publication of the report.
HISTORY.

Since the year 1879 the German Imperial Statistical Office has
published monthly average wholesale prices of commodities of
importance in German markets. These were shown in detail in the
Monatshefte zur Statistik des Deutschen Reichs up to the year 1891,
and from 1892 to the present time in the Vierteljahrshefte zur
Statistik des Deutschen Reichs.
The object of the price study, as stated at the outset, was the col­
lection of reasonably accurate and adequate average prices repre­
senting fixed grades of important articles of the wholesale trade,
with a view to the gradual assembling of really useful data for the
observation of the movement of prices. It was not until the year
1905 that the publication of relative prices was begun. The official
series of index numbers has been extended back only as far as the
year 1899.
SOURCES OP QUOTATIONS.

The number of markets represented in this study has been limited
to those with permanent arrangements for furnishing reasonably
accurate and representative quotations. The following sources of
information, representing 30 wholesale markets, are acknowledged in
the report for the year 1911: Chambers of commerce or boards of
trade in Augsburg, Berlin, Bielefeld, Brunswick, Bremen, Breslau,
Danzig, Dortmund, Frankfort on the Main, Halberstadt, Hamburg,
Cologne, Konigsberg (in Prussia), Krefeld, Landeshut (in Silesia),
Leipzig, Lubeck, Magdeburg, Mannheim, Muhlhausen (in Alsace),
Munich, Mtinchen-Gladbach, Nuremberg, Posen, and Stettin; the
administrations of municipal stockyards and slaughterhouses; the
null administration in Bromberg (for wheat flour from Berlin), the
stock exchange in Dusseldorf, the board of directors of the stock
exchange in the city of Essen, the United German Jute Manufac­
turers in Brunswick (for raw jute at Hamburg), the Konigsberg




INDEX NUMBERS---- GERMANY.

209

Commercial Association in Konigsberg (in Prussia) (for petroleum at
Danzig), the Merchants’ Association at Lindau, the Royal Adminis­
tration of Mines at Saarbrucken, and the Bureau of Trade Statistics
at Hamburg.
From the beginning ordinary published market quotations have
been avoided as representing fluctuations in quality and as not being
scientifically constructed. The study has been restricted to prices
secured currently from the above-named or similar sources.
BASE PERIOD.

The 10-year period 1889 to 1898 is taken as the base period.
reasons are assigned for this selection.

No

PRICES: H O W SHOW N AND COMPUTED.

Three tables show the prices involved in the computation of index
numbers. The first shows average monthly prices for the current
year, the second shows average yearly prices for the 20-year period
ending with the current year, and the third shows relative prices
for each year of the 10-year period ending with the current year.
All actual prices shown are averages. A tabular statement in the
first report (February, 1879) gave for each of the 26 markets then
included in the study the intervals at which prices for the Imperial
Statistical Office were determined and the methods of determining the
quotations. According to this statement the average actual prices
represent great variations from market to market in the number of
original quotations involved in the computation, some being based
on daily determinations while others are based on weekly or even
monthly determinations, and some representing a medium price or
quality while others are averages of the prices of the highest and
lowest or of the highest, medium, and lowest grades of the commodi­
ties reported.
A few series of index numbers represent interrupted series of actual
prices and a few others represent series of actual prices whose com­
parability is broken within the period involved in the table.
NUMBER AND CLASS OF COMMODITIES.

From the beginning (1899) index numbers have been given in sum­
mary form for 44 quotations. Of this number three represent iron and
two each represent coal and petroleum. Each of the other commodities
is represented by a single quotation. The number of commodities
is therefore 40. No index number for the total of the 44 quotations
is published, nor are index numbers shown for groups of commodities.
Index numbers are given for 40 articles, representing the 235
quotations of the table of actual prices. The number was originally
238, the two Stuttgart quotations for cotton yarn and the Stuttgart
94261°—Bull. 173—15------14




210

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

quotation for cotton goods having been dropped from both actual
and relative tables.
The table of actual average yearly prices as published in 1913 con­
tains 320 series of quotations, some of which are themselves calcu­
lations from more than one variety, as, for example, the first Mann­
heim quotation for barley. This table at present includes five com­
modities not represented in the tables of relatives, namely, raw
sugar, refined sugar, molasses, cocoa, and rubber. The compara­
bility of all series of sugar quotations has been interrupted by changes
in tariff laws; there is no continuous series of yearly average prices
on molasses for the base period, and the last two articles have been
added to the* list of commodities since the publication of index num­
bers was begun—cocoa in 1907 and rubber in the following year.
The commodities included are not classified into raw materials and
manufactured products. The great majority are raw materials but
a number are so-called semimanufactures ( H a lb fa b r ik a te n ).
Some commodities originally omitted from the list were considered
desirable but were not included because satisfactory data could not
be secured. The original number of articles (30) has been con­
siderably increased, but still certain important articles, as, for ex­
ample, lumber and flax, are not included even in the tables of actual
prices.
DESCRIPTION AND GROUPING OF COMMODITIES.

Index numbers are shown only for single commodities and not for
groups.
The description of commodities in the table of index numbers as
published in 1912 follows:
Rye (1,000 kilograms).

Berlin, good, minimum 712 grams per liter [51.3 pounds per bushel].
Breslau, medium grade.
Danzig, goods for free exchange (Ware z.freien Verkehr).
Frankfort on the Main, minimum 70 kilograms per hectoliter [54.4 pounds per bushel].
Hamburg, Russian, in bond.
Konigsberg, good, 714 grams per liter [51.5 pounds per bushel].
Leipzig, German, good.
Liibeck, Russian, 71.3 kilograms per hectoliter [55.4 pounds per bushel],
Mannheim, various origins, medium.
Munich, Bavarian, best.
Munich, Bavarian, good medium.
Wheat (1,000 kilograms).

Berlin, good, minimum 755 grams per liter [54.4 pounds per bushel].
Breslau, medium grade.
Danzig, goods for free exchange.
Frankfort on the Main, minimum 75 kilograms per hectoliter [58.3 pounds per bushel].
Hamburg, Holstein, Mecklenburg.




INDEX NUMBERS---- GERMANY.

211

Konigsberg, good, 749 to 754 grams per liter [54.0 to 54.4 pounds per bushel].
Leipzig, German, good.
Lindau, 78 to 79 kilograms per hectoliter [60.6 to 61.4 pounds per bushel], various
origins.
Mannheim, various origins, medium.
Munich, Bavarian, best.
Munich, good medium.
Oats (1,000 kilograms).

Berlin, good, minimum 450 grams per liter [32.4 pounds per bushel].
Breslau, medium grade.
Danzig, domestic.
Frankfort on the Main, good, native.
Konigsberg, good, 447 grams per liter [32.2 pounds per bushel].
Leipzig, German, good.
Lindau, Bavarian, 44 to 45 kilograms per hectoliter [34.2 to 35.0 pounds per bushel].
Mannheim, from Baden, from Wurttemberg, medium.
Munich, Bavarian, best.
Munich, Bavarian, good medium.
Corn (1,000 kilograms).

Bremen, American, best, in bond.
Breslau, Russian, medium grade.
Hamburg, American, in bond.
Leipzig, various origins.
Barley (1,000 kilograms).

Breslau, medium grade.
Danzig, brewing, domestic.
Frankfort on the Main, brewing.
Konigsberg, 647 to 652 grams per liter [46.6 to 47.0 pounds per bushel].
Leipzig, German, good.
Lindau. Hungarian, 65 to 66 kilograms per hectoliter [50.5 to 51.3 pounds per bushel].
Magdeburg, Chevalier, good medium. (Not in 1912.)
Mannheim, from Baden, from the Palatinate, medium.
Munich, Hungarian, Moravian, etc., best.
Munich, Bavarian, best.
Munich, Bavarian, good medium.
Hops (100 kilograms, without wrappings).

Nuremberg, market.
Nuremberg, Wurttemberg.
Nuremberg, Hallertauer.
Nuremberg, Hallertauer seal.
Nuremberg, Spalt.
Potatoes (1,000 kilograms, without sack).

Berlin, early red, for distilling,
Berlin, early red, for food, assorted.
Breslau, good, Silesian, food.
Magdeburg, Saxon, for food.
Magdeburg, distilling.
Stettin, sorted, red, for food.
Stettin, sorted, white, for food.




212

BULLETIN OF THE BUREAU OF LABOR STATISTICS.
Butcher's meat.

Beef (100 kilograms [220.5 pounds]) Berlin, slaughter weight.1
Pork (100 kilograms [220.5 pounds]) Berlin, slaughter weight.1
Veal (100 kilograms [220.5 pounds]) Berlin, slaughter weight.1
Mutton (100 kilograms [220.5 pounds]) Berlin, slaughter weight.1
Rye flour (100 kilograms with sack).

Berlin, No. 0/1, good average grade.
Danzig, No. 0/1, domestic price (Inlandspreis).
Cologne, No. 0/1.
Munich, No. 0.
Posen, domestic, No. 0/1.
Wheat flour (100 kilograms).

Berlin, No. 00, with sack.
Danzig, No. 00, with sack, domestic price
Cologne, Rhenish, No. 00, with sack.
Ltibeck, German, No. 0, without sack.
Munich, Bavarian, No. 2, with sack.
Posen, domestic, No. 00, with sack.
Butter (100 kilograms).

Berlin, I quality.
Berlin, II quality.
Munich, finest Swiss.
Munich, mountain.
Raw sugar (100 kilograms, net weight).2

Brunswick, 88 per cent centrifugal, without sack, 3 months’ time.
Halle, 88 per cent centrifugal, without sack, 3 months’ time.
Magdeburg, I product, 88 per cent centrifugal, without sack, 3 months’ time.
Stettin, 88 per cent centrifugal, without sack, 3 months’ time.
Refined sugar (100 kilograms).2

Brunswick, without container in paper.
Magdebuxg, I loaf (Brot), without container, in paper.
Stettin, I loaf (Brot), without container, in paper.
Molasses (100 kilograms, net weight).2

Magdeburg, for distilling.
Potato alcohol, crude (100 liters), alcohol.

Hamburg, with container.
Rapeseed oil (100 kilograms).

Berlin, crude, without container.
Danzig, crude, with container, export price.
Frankfort on the Main, with container.
1 Slaughter weight (Schlachtgewicht) is the presumptive weight of the four quarters on which the price
of the animal, without the deduction of the value of hide, head, feet, entrails, etc., has been apportioned.
Prior to July 1,1897, quotations were not on slaughter weight. A t the time the change was made it
was stated that according to information from authoritative sources the quotations on slaughter weight
are about 8£ per cent higher than on dressed weight.
2 Descriptions from table of actual prices. Article not included in table of relative prices.




INDEX NUMBERS---- GERMANY.

Hamburg, crude, with container.
Cologne, crude, good and clear, with container.
Konigsberg, crude, clear, without container.
Leipzig, crude, light color and clear, without container.
Mannheim, marketable, with container.
Herrings (1 cash, 150 kilograms).

Danzig, with container, Crown and full.
Danzig, with container, Crown, Ihlen.
Hamburg, with container, in bond, Norwegian.
Hamburg, with container, in bond, Scotch West Coast.
Stettin, clear, with container, Norwegian, commercial.
Stettin, clear, with container, Norwegian, large medium.
Stettin, clear, with container, Norwegian, fair medium.
Stettin, clear, with container, Norwegian, medium.
Stettin, clear, with container, Scotch, Crown, full brand.
Stettin, clear, with container, Scotch, Crown, Matfulls.
Stettin, clear, with container, Scotch, Crown, Ihlen.
Coffee (100 kilograms).

Bremen, clear, with sack, in bond, Sabanilla, fair ordinary.
Bremen, clear, with sack, in bond, Santos, good average.
Hamburg, net weight, in bond, Santos.
Hamburg, net weight, in bond, Rio.
Hamburg, net weight, in bond, La Guaira, unwashed.
Cologne, net weight, with sack, Java, good medium.
Cologne, net weight, with sack, Santos, good medium.
Mannheim, Santos, average quality.
Cocoa (100 kilograms, in bond).1

Hamburg,
Hamburg,
Hamburg,
Hamburg,
Hamburg,
Hamburg,

Akkra current.
St. Thome, fine.
Bahia, fair, fermented.
Trinidad current.
Samina current.
Arriba, choice, summer.
Tea (1 kilogram, in bond).

Hamburg, Kongo, Foochow.
Hamburg, Kongo, Shanghai.
Hamburg, Souchong.
Konigsberg, common Moning.
Konigsberg, fine Moning.
Konigsberg, finest Moning.
Rice (100 kilograms, in bond).

Bremen, Rangoon, shelled, 4 months ’ time.
Bremen, broken, No. 0, shelled, 4 months’ time.
Hamburg, Rangoon, shelled, highest price, 1 per cent discount.
Hamburg, Rangoon, shelled, lowest price, 1 per cent discount.
Hamburg, broken, shelled, lowest price, 1 per cent discount.

i Descriptions from table of actual prices. Article not included in table of relative prices.




213

214

BULLETIN OF THE BUKEATJ OF LABOR STATISTICS.
Pepper (100 kilograms, in bond).

Bremen, Singapore, 4 months’ time.
Hamburg, Singapore, 1 per cent discount.
Lard (100 kilograms, in bond).

Bremen, refined American, 4 months’ time.
Leaf tobacco (100 kilograms).

Bremen, with packings, in bond, Kentucky, ordinary, 6 months’ time.
Hamburg, in bond, Domingo, wrapper and filler leaves.
Hamburg, in bond, Brazil.
Mannheim, in bond, wrapper leaves, Palatinate.
Mannheim, in bond, wrapper leaves and filler leaves, Palatinate.
Mannheim, in bond, cut, Palatinate.
Hides and skins.

Bremen, 100 kilograms, ox hides, best dry, Buenos Aires, 6 months’ time.
Bremen, 100 kilograms, Buenos Aires, Saladero, 6 months’ time.
Bremen, 100 kilograms, kip hides, Durbunga, arsenic slaughtered, 6 months’time.
Bremen, 100 kilograms, kip hides, Hugli, slaughtered, 6 months’ time.
Bremen, 100 kilograms, kip hides, Dakka, best, 6 months’ time.
Hamburg, 100 kilograms, ox hides, Rio Grande, salted.
Hamburg, 100 kilograms, ox hides, dry, West Indian, Central American, etc.
Cologne, 100 kilograms, ox hides, best, green, Uruguay, 6 months’ time.
Cologne, 100 kilograms, kip hides, dry, East Indian, best Dakka, 6 months’ time.
Munich, 100 kilograms, ox and cow hides, best, green.
Frankfort on the Main, 100 kilograms, calfskins, 3 to 4 months’ time.
Frankfort on the Main, 100 kilograms, goatskins, 3 to 4 months’ time.
Frankfort on the Main, 100 kilograms, hare skins, 3 to 4 months’ time.
Leipzig, 500 skins, hare skins, German.
Leipzig, 500 skins, hare skins, Russian.
Wool (100 kilograms).

Berlin, North German sheep, medium.
Bremen, washed, Buenos Aires, I.
Cotton (100 kilograms).

Bremen, middling upland.
Bremen, good Oomrawuttee II.
Hamburg, New Orleans, middling.
Cotton yam (1 kilogram).

Augsburg, 36 warp, 42 woof.
Augsburg, 20 warp, 20 woof.
Krefeld, English Nos., 40-120, twofold, singed.
Krefeld, English Nos., 130-200, twofold, singed.
Miilhausen, in Alsace, metric Nos., warp No. 16, 30 days, 2 per cent discount; cash,
2J per cent discount.
Miilhausen, in Alsace, metric Nos., warp No. 28, 30 days, 2 per cent discount; cash,
2£ per cent discount.
Miilhausen, in Alsace, metric Nos., warp No. 40, 30 days, 2 per cent discount; cash,
2J per cent discount.




INDEX NUMBERS---- GERMANY.

215

Miilhausen, in Alsace, metric Nos., woof No. 16, 30 days, 2 per cent discount; cash,
2h per cent discount.
Miilhausen, in Alsace, metric Nos., woof No. 37, 30 days, 2 per cent discount; cash,
2J per cent discount.
Miilhausen, in Alsace, metric Nos., woof No. 50, 30 days, 2 per cent discount; cash,
2J per cent discount.
Munchen-Gladbach, Mule No. 8, f. o. b. factory.
Munchen-Gladbach, Water No. 12, f. o. b. factory.
Munchen-Gladbach, Water No. 20, f. o. b. factory.
Calico (1 meter).

Miilhausen, in Alsace, 90 centimeters [35.4 inches] wide.
Miilhausen, in Alsace, 78 centimeters [30.7 inches] wide, 16/16 thread.
Linen yam (1 kilogram).

Bielefeld, English Nos., average price for I and II, No. 30, flax yarn.
Bielefeld, English Nos., average price for I and II, No. 50, flax yarn.
Bielefeld, English Nos., average price for I and II, No. 10, tow yarn.
Bielefeld, English Nos., average price for I and II, No. 20, tow yam.
Landeshut, in Silesia, English Nos., average price for I, No. 30, flax yarn.
Landeshut, in Silesia, English Nos., average price for I, No. 50, flax yarn.
Landeshut, in Silesia, English Nos., average price for I, No. 10, tow yarn.
Landeshut, in Silesia, English Nos., average price for I, No. 20, tow yarn.
Raw silk (1 kilogram).

Krefeld, Italian organzine, 18-20, 9 months’ time, or cash 5 per cent discount.
Krefeld, Italian tram, 24-26, 9 months’ time, or cash 5 per cent discount.
Krefeld, Italian raw (grege.), 12-14, 9 months’ time, or cash 5 per cent discount.
Krefeld, Japanese organzine, 22-24, 9 months’ time, or cash 5 per cent discount.
Krefeld, Japanese tram, 34-40, 9 months’ time, or cash 5 per cent discount.
Krefeld, Chinese tram, 36-40, 9 months’ time, or cash 5 per cent discount.
Hemp (100 kilograms).

Liibeck, Petersburg dressed hemp.
Mexican fiber (100 kilograms).

Hamburg, in bales.
Raw jute (100 kilograms).

Hamburg, Brand RCF
Hamburg, Good I, native brands.
Hamburg, Good II, native brands.
Rubber, crude (1 kilogram).1

Hamburg,
Hamburg,
Hamburg,
Hamburg,
Hamburg,
Hamburg,
Hamburg,

South Kamerun.
Benguela II.
Upper Kongo I.
Kassai I, red.
Massai.
Mosambique I.
fine Para, hard.

1Descriptions from table of actual prices. Articles not included in table of relative prices.




216

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

Hamburg, Manaos, Negro, heads.
Hamburg, Cameta.
Hamburg, Peruvian balls.
Hamburg, Mexican gum.
Iron (1,000 kilograms).
German, pig:
Breslau, at the foundry, puddle.
Breslau, at the foundry, foundry pig.
Dortmund, at the foundry, Bessemer.
Dortmund, at the foundry, puddle-1.
Dortmund, at the foundry, Thomas.
Dusseldorf, at the foundry, puddle.
Dusseldorf, at the foundry, foundry pig.
Dusseldorf, at the foundry, Luxemburg No. 3.
English, pig:
Hamburg, Scotch No. 1.
Hamburg, Middlesboro No. 1.
Swedish, bar:
Liibeck, I Stockholm.
Lead (100 kilograms).

Berlin, various German brands.
Frankfort on the Main, Rhenish, double refined.
Halberstadt, refined, Harz, soft.
Halberstadt, refined, Silesian, soft.
Hamburg, Harz, soft, double refined.
Cologne, Rhenish, soft, double refined.
Copper (100 kilograms).
Berlin, Mansfeld.
Berlin, foreign I, Bede brand.
Frankfort on the Main, German double refined, in sheets.
Hamburg, English, best selected.
Zinc (100 kilograms).

Breslau, good, Silesian.
Frankfort on the Main, refined, zinc blend.
Halberstadt, Rhenish Westphalian, crude.
Hamburg, Silesian, in sheets.
Cologne, Rhenish, crude, “ W II und S S.”
Tin (100 kilograms).

Frankfort on the Main, Banca.
Hamburg, Banca, in blocks.
Anthracite coal (1,000 kilograms).

German:
Breslau, pit price, Lower Silesian, gas.
Breslau, pit price, Upper Silesian, gas.
Dortmund, at the mine, fallen, lump (run of mine).
Dortmund, at the mine, puddle.
Dusseldorf, at the mine, open-burning.
Dusseldorf, at the mine, anthracite.
Dusseldorf, at the mine, uninflammable.
Dusseldorf, at the mine, gas.
Essen, at the mine, open-burning.
Essen, at the mine, anthracite.
Essen, at the mine, uninflammable.



INDEX NUMBERS---- GERMANY.

217

German—Concluded.
Essen, at the mine, gas.
Saarbriicken, at the mine, open-burning.
Saarbrticken, at the mine, anthracite.
English:
Danzig, f. o. b., English, pea.
Danzig, f. o. b., Scotch, machine.
Hamburg, f. o. b., West Hartley.
Hamburg, f. o. b., Sunderland.
Petroleum (100 kilograms), with container.

A.merican:
Standard white, Berlin.
Standard white, Danzig.
Standard white, Hamburg, in bond.
Standard white, Magdeburg.
Standard white, Mannheim.
Standard white, Posen.
Standard white, Stettin.
Russian:
Breslau.
Liibeck, “ Nobel.”
SUBSTITUTIONS, ADDITIONS, AND OMISSIONS.

Within the period of 13 years covered by index numbers three
series of relatives have been changed, namely, those for hides and
skins, cotton yam, and cotton cloth. The change in the series for
hides and skins was occasioned by the substitution in 1909 of a new
set of quotations for Frankfort hare skins (German and Russian)
with no alteration in the description of the article and with some of
the earlier actual average prices identical with the old figures. The
data for six years of the base period are incomplete for the new
series. No reason is assigned for the substitution of the new series
and no explanation is given in regard to its source. The later series,
being published in the 10-year table 1900-1909, does not include the
first year for which index numbers have been regularly shown.
Minor substitutions of varieties or brands which apparently do not
affect prices are occasionally indicated in footnotes to the tables
(e. g. the Mannheim quotations on oats as given in the tables pub­
lished in 1912). Actual price series printed in “ old-style” type on
account of a break in the comparability of the figures are in a num­
ber of cases represented by relative series (e. g. the Cologne coffee
quotations).
The changes in the series of index numbers for cotton yam and for
cotton cloth were occasioned by the discontinuance of quotations
from the Stuttgart market. In this case new series of relatives with
the Stuttgart quotations eliminated were constructed for the whole
period covered by the index numbers. Series of actual prices with
data lacking for one or more years are in several cases represented
by series of relative prices.



218

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

Within the period which affects the index numbers only two new
commodities, cocoa and rubber, have been added. In these cases,
as in cases where new quotations are added for commodities already
included, the actual price series alone is given.
TABLE OF RESULTS.

The following table of index numbers is taken from more than one
report, because no table of relatives as published covers more than
10 years. In cases where the earlier and later series of index num­
bers are not the same both sets of figures have been copied.
RELATIVE PRICES OF ARTICLES OF WHOLESALE TRADE.
[Source: Vierteljahreshefte 7.ur Statistik des Deutschen Reichs: 1912 and preceding years.)
(Base period, 1889-1898=100.)

Commodity.

1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912

96
R ve...................................... 100 97
88
91
94
Wheat..................................
101
98
96
Oats.....................................
91 103 106
Corn (maize)........................
99
96 98
Barley.................................
81
Hops.................................... 118 90
93
103
87
Potatoes..............................
101. 104 102
Beef.....................................
Pork...................................
91
92 107
Veal..................................... 0)
0)
0)
Mutton............................... 106 108 109
94
Rye llonr.............................
97 95
Wheat flour.........................
91
86 92
Butter.................................. 100 100 105
Potato alcohol..................... 102 97
81
90 109 107
Rapeseed o i l .......................
Herrings.............................. 129 144 116
Coffee...................................
53
63
56
Tea......................................
92
84
93
Rice..................................... 106 104 104
Pepper................................. 147 168 164
Lard....................................
82 105 127
Leaf tobacco........................ 103 108 108
1
0
) 2113 2111
Hides and skins.................. \8107
8110 8107
W ool.................................... 127 117
94
Cotton..................................
79 120 102
4105
Cotton yarn......................... / 492 <117 5104
\6 91 5116
8 6 6107 6 89
Calico................................... /«
\5 86 H08 5 89
Linen yarn..........................
96 118 120
Raw silk.............................. 109 102
91
Hemp..................................
112 124 135
Mexican fiber...................... 105 112
97
Raw jute.............................
98 114 107
Iron, German, pig............... 122 153 115
Iron, English, pig...............
128 145 113
Iron, Swedish, oar.............. 120 148 118
Lead..................................... 130 149 112
Copper................................. 149 148 141
Zin c...................................... 131 108 91
Tin....................................... 153 166 148
Coal, anthracite, German... 106 120 123
Coal, anthracite, English... 113 159 125
Petroleum, American......... 106 110 106
Petroleum, Russian............ 102 106 97

97
94
108
116
95
88
81
106
114
0)
117
95
92
101
71
100
128
55
83
95
161
152
101
2122
3118
104
105
4105
5104
694
5 94
102
99
128
98
103
106
115
115
97
108
98
150
116
118
103
91

91
90
93
106
92
138
102
113
96
0)
128
89
89
105
86
90
106
51
87
104
167
126
93
2117
8115
117
128
4121
5121
6110
5111
110
107
121
117
111
105
108
113
100
120
110
158
112
114
110
99

90
98
92
108
94
166
133
115
94
0)
123
87
94
107
127
85
99
60
90
101
159
105
91
2121
8118
118
144
4131
5131
«118
5118
120
93
114
128
113
104
102
112
102
119
118
158
111
108
104
98

98
98
102
115
101
107
126
120
123
(l)
134
91
94
112
107
87
136
62
81
100
155
109
92
2132
8129
123
114
4108
5117
6108
5109
116
99
114
122
149
104
108
114
118
141
133
180
113
108
99
96

107
100
114
119
103
72
83
129
128
0)
146
101
97
115
91
105
146
62
87
103
143
130
103
2146
8142
134
130
\l34

127
116
125
131
114
78
122
128
106
(l)
144
122
112
114
119
135
115
59
94
115
123
135
121
2137
8134
138
135
159

}127
J
130
107
125
121
194
119
117
118
149
176
141
225
118
113
106
102

145
157
133
132
117
185
136
126
118
165
179
126
216
127
137
108
106

115
130
119
139
112
97
115
115
128
0)
137
110
126
120
108
107
107
62
87
107
90
171
112
\l43
/
132
137
131

101
116
107
127
103
142
98
127
127
0)
143
97
113
123
107
107
121
74
94
105
103
183
129
156
135
172
149

113
114
120
127
121
195
146
134
110
167
146
106
111
129
104
118
124
97
102
116
122
136
141
144
129
159
142

125
121
134
147
129
184
165
145
141
181
160
115
115
134
141
123
145
107
102
142
143
156
123
180
131
142
136

119 113
132 110
96
99
128 130
113 100
152 128
119 101
112 110
115 115
116 112
121 119
107 119
168 169
133 129
129 118
113 110
112 106

121
124
96
138
104
131
107
112
124
111
116
124
193
127
120
105
103

119
134
95
145
114
180
108
110
123
120
114
135
238
125
118
105
104

123
134
93
167
113
191
123
129
123
154
148
141
259
130
139
125
120

123
118
114
140
118
54
121
121
112
0)
136
120
115
119
143
129
92
60
88
113
87
133
118
2125
8122
121
122
137

1 No index number published.
2 New series occasioned by substitution in Frankfort quotation for German and Russian hare skins.
8 Old series.
* New series including Stuttgart quotations.
5 Old series including Stuttgart quotations.
« New series excluding Stuttgart quotations.




INDEX NUMBERS---- GERMANY.

INDEX NUMBERS

219

OF THE JAHRBUCHER FUR NATIONALOKONOMIE
UND STATISTIK.
PUBLICATION.

Three general “ indexes” and one table of index prices for which no
general relative is computed, all based on German wholesale price
statistics, are found in the Jahrbucher fur Nationalokonomie und
Statistik, which is published monthly at Jena, Germany. They are,
as now published, the work of Dr. Johannes Conrad, a professor of
the University of Halle, although in the years through which the
index numbers have been carried, different persons have assisted in
their compilation.
No distinctive name, so far as known, has become identified with
any of the three, though they are variously spoken of as "Conrad’s
indexes” and the "Jahrbucher indexes,” and are now and then
referred to as the "Hamburg indexes.”
These index numbers appear about two years late, and with some
irregularity. Thus the figures for 1911 appeared in the August, 1913,
issue and those for 1910 in the issue of July, 1912. Since 1887, how­
ever, presentation of the figures at some time during the year has been
made. Each issue of the Jahrbucher reproduces the figures from the
beginning, though those for the earlier years are grouped by periods
in the later issues.
HISTORY.

The following is a translation of the history of these price studies ^as
given by Dr. Conrad in volume 17 of the Jahrbucher, third series,
1899, page 642:
In these Jahrbucher, in volume 3,1864, appeared for the first time
the results of an investigation into the course of prices, based upon
the Hamburg quotations on imported goods subject to taxation.
Prof. Laspeyres was the author. This study was based upon the one
already made by Soetbeer which brought the data up to 1856, Las­
peyres carrying the figures to 1862 for 48 articles. He compared the
years 1851-1862 with Doth of the two preceding decades, taking the
arithmetical mean of the prices, in order to demonstrate the alteration
of prices in consequence of the gold importation resulting from the
discovery of gold in California.
In the year 1874, in volume 23, Prof. H. Paasche, then a student, at
my suggestion carried the investigation further upon the basis of the
same materials but by a different method. Instead of computing the
arithmetical mean he reckoned the quantities of 22 articles consumed
by the population in the current year and multiplied the quantity of
each article by the average price for the basic period (1847-1867) and
by that for the current year, in order to get a more accurate relative.
Again in the year 1882 Richard van der Borght, now Prof. van der
Borght, took up this computation according to the same principles
and for the same articles and printed the results of his investigations
in volume 5 (new series), 1882. However, he used other figures for the




220

BULLETIN OF THE BUBEATT OF LABOB STATISTICS.

consumption quantities, since statistics on this subject in the mean­
time had improved.
In the year 1887 we ourselves in volume 15 (new series) utilized the
Hamburg quotations for an investigation of the price reduction in the
eighties, and carried the data forward in the same manner. Since
then we have annually compared the prices of the current year with
those for preceding years, in order to determine how the movement
of prices has further developed.
SOURCE OF QUOTATIONS.

Two of Dr. Conrad’s indexes and one table of 47 articles for which
a general index is not printed are based on the actual yearly average
prices appearing in the publication entitled “ Hamburg’s commerce
and shipping” (H a m b u rg s H a n d e l u n d S c h iffa h r t), issued by the
Hamburg Bureau of Trade Statistics (H a n d e lss ta tistis ch es A m t ) .
This publication contains actual prices, per 100 kilograms, net,
stated in marks. The report for the year 1911 contains prices for 174
articles and subdesignations of articles. The price is that of sea­
borne commerce declared at entry at the port of Hamburg.
In the first study published in the Jahrbucher, that appearing in
volume 3, 1864, by Dr. E. Laspeyres, professor in the University of
Basel, and entitled “ Hamburg prices from 1851 to 1863, and the Cali­
fornia and Australian gold discoveries since 1848” ( H a m b u rg er W a a r en p r e is e 1 8 5 1 -1 8 6 3 u n d d ie c a lifo rn iscJ i-a u stra lisch en G o ld en td eck u n g en

848), Dr. Laspeyres stated that the prices for the 48 articles used
by him in preparing his relative had appeared every Friday since the
year 1736 in the official “ General price-current” (A llg e m e in e r P r e i s
C o u r a n t).
However, so far as known, no use of them for purposes of
a relative had been made prior to Prof. Soetbeer’s compilation
beginning with 1831. In volume 23, 1874, Prof. Paasche stated that
the current report on Hamburg’s commerce and shipping for that year
contained prices for more than 300 articles for the years 1847-1872.
Prof. Soetbeer, in a study entitled “ The movement of prices in the
years 1886-1890” (D a s N iv e a u d er W a r e n p r e is e i n d e n J a h r e n 1 8 8 6 1 8 9 0 ), published in the Jahrbucher, volume 58 (3d series, vol. 3), 1892,
made the same statement for the years 1886-1888. In other words,
no material change had occurred in the method of quoting Hamburg
prices from 1847 to 1888.
During all these years Hamburg had been a free port, collecting
duties on all goods entering the city, even if they came from other
States of Germany, and therefore the prices declared at entry on all
goods, whether received into the city by sea or river, by rail or
wagon, appeared in the official price statistics. On October 15,1888,
however, Hamburg entered the German customs union ( Z o llv e r e in ).
The following statement as to the effect of this change on the Ham­
burg quotations is abstracted from Prof. A. Soetbeer’s statement in
s e it I




INDEX NUMBERS---- GERMANY.

221

the Jahrbucher fur Nationalokonomie und Statistik, vol. 3, 3d series,
1892, pp. 590, 591:
With the entry of Hamburg into the German customs union the
Hamburg trade statistics underwent a fundamental alteration which
no longer permitted of a direct comparison with former years. The
quotations upon articles of domestic commerce—i. e., articles received
from neighboring German States—ceased, and quotations upon im>orts entering by sea only were available from that date. Thereore, for only those articles which had previously been exclusively or
in overwhelming proportion brought in bv sea were the figures after
1888 comparable. If the long series o f preceding calculations of
average yearly prices was not to be finally terminated and an entirely
new series started, it would be necessary to make a complete revision
and recomputation of the preceding tables and a new computation of
average actual prices to include through the entire period from 1847
to 1888 only those articles which had been importea by sea.
The great importance of the Hamburg quotations as a basis for
price statistics being generally recognized, and the great desirability
of a continuous series from 1847 being evident, the director of the
Hamburg bureau of trade statistics decided to make this recomputa­
tion. By means of subsidiary material, by use of the price declara­
tions on exports by sea for the period 1847-1888, by the utilization
of trustworthy market reports for the period, and in some degree by
the use of expert estimates, the average actual prices were brought
to the more restricted basis for the entire period.
Prof. Soetbeer concludes his discussion in these words: “ These com­
bined means—i. e., of recomputation—have cost much work and
time, but one may with all good confidence put trust in them that in
spite of all difficulties success has been attained in securing the con­
tinuity of this most important work on the movement of prices,
which is not only desirable but necessary.”
The effect of this change is discussed by Dr. Conrad in the Jahr­
bucher for 1893 (3d series, vol. 6, p. 695):

{

Unfortunately the first source—i. e., Hamburgs Handel und
Schiffahrt—has undergone an alteration which has rendered necessary
a reconstruction of the tables based upon it. Up to the year 1888
the prices were reckoned upon the customs declarations of goods
entering by all means of transportation. As a result of the entry of
Hamburg into the German customs union the customs declarations
now embrace only goods imported by sea. As a result of this a
comparison with preceding years could not be made with any accu­
racy. Now, however, the yearly average actual prices for 1847-1888
have been recomputed on the basis of 1888. This has been done by
restricting the prices for 1847-1888 to those articles which were im­
ported by sea. This reduced the number of articles upon which
quotations were given from 320 to 180.
No changes of importance have taken place since Dr. Conrad’s
article was written. An average actual quotation is now carried for
each of 171 articles and subdesignations of articles,




222

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

The fourth series of relative prices published in the Jahrbucher and
prepared by Dr. Conrad is not based upon the famous Hamburg
wholesale price quotations but upon official statistics of the German
customs union. The history and source of this series are sep­
arately treated oh page 232 of the present bulletin.
From this point it will be necessary to consider separately the
three series of index numbers based on the Hamburg trade statistics
which Dr. Conrad now presents annually in the Jahrbucher.
Index numbers computed from prices of 39 selected articles.

The first series contains relatives for each of 39 articles from 1871 to
date and appears in current issues of the Jahrbucher. No general
relative is now computed for this group. This is the table used in
the Report from the Committee on Finance of the United States Sen­
ate, 1893.1
BASE PERIOD.

This table appears in the first study of this series, vol. 3 of the
Jahrbucher, 1864, by Dr. E. Laspeyres in practically its present form.
For 42 of the 48 articles then included in it the period 1831-1840 was
used as a base. Relatives for 3 articles had 1841-1850 as a base,
while for 3 articles 1851-1853 was the base. A general relative, not
weighted, was computed with 1831-1840 as the base.
When the study of relative prices was resumed in the Jahrbucher by
Prof. Paasche, volume 23, 1874, the period 1847-1867 was used as a
base and the relative for each of 47 articles was computed, but as at
present Prof. Paasche made no general relative for this table.
Of the three articles for which Dr. Laspeyres had been forced to use
1851-1853 as abase, Dr. Paasche dropped two (soda and Java coffee),
while rapeseed appeared under a slightly different designation. He
makes no comment as to how he adjusted it to the basis of 1847-1867.
Dr. Richard van der Borght, by whom the work was continued in
1882, did not print this table at all. Its publication was resumed,
however, by Dr. Conrad in volume 1, 3d series, of the Jahrbucher,
1891, with 1847-1870 as the base. This base has remained unaltered
since that date.
PRICES: H O W SHOW N AND COMPUTED.

In the latest available presentation of the table, that published in
June, 1914, relatives are shown by articles and decades from 1871 up
to and including 1900. Relatives appear for 1901-1905, 1906-1910,
and for each of the years separately from 1906 to 1912, inclusive.
By the use of the tables in preceding issues, beginning with volume 1,
3d series, 1891, it is possible to get the decennial relative for each
article since 1851-1860 and the yearly relative from 1886 to date, on
i Report from the Committee on Finance of the United States Senate on Wholesale Prices, Wages, and
Transportation. Mar. 3,1893. 52d Congress, 2d session, Report No. 1394. Pt. I, pp. 297-301.




INDEX NUMBERS---- GERMANY.

223

the present base. No yearly relatives are printed for the years
1864-1885, inclusive, and the yearly relatives for 1851 to 1863, ap­
pearing in the Jahrbucher (vol. 3, 1864) are, as already noted, on a
different base.
Their reduction to the present base has not been
made except by decades.
Actual average prices are also shown in the June, 1914, issue of the
Jahrbucher by zentners (50 kilograms) in marks, for 1847-1870, for
the decades 1871-1880, 1881-1890, and 1891-1900, for the five-year
periods 1901-1905 and 1906-1910, and by single years from 1906 to
1912. By the use of preceding issues, actual average prices by years
are available, beginning with 1886.
A comparison with the official figures shows that from 1891 to
date the actual prices of the table under consideration are those of
Hamburgs Handel und Schiffahrt as now published in its current
issue, reduced from the doppelzentner (100 kilograms) to the zentner
(50 kilograms). The data for years prior to 1891, however, are not
those shown in the current official publication but those published
previous to the complete recomputation of the official figures, which
was made after Hamburg entered the German customs union; in
other words, Dr. Conrad's table of actual prices for years prior to
1888 has not been readjusted to the basis of 1888, as are the official
quotations which are now published for Hamburg.
Two relatives are given for each article for the current year,
namely, the per cent which the average price for that year is of the
average price for 1847-1850 and for 1871-1880, respectively.
No general relative has ever been printed for this table except
as it first appeared in 1864, although the Committee on Finance of the
United States Senate in 1893 computed a relative for its own use.1
NUMBER AND CLASS OF COMMODITIES.

The table shown in the Jahrbucher for June, 1914, the latest avail­
able, embraces 39 articles. As originally computed by Dr. Laspeyres
the table contained 48 articles, but Prof. Paasche in 1874 made a num­
ber of changes in his selection and reduced the number to 47. He
brought the table from 1847 to 1872 to the basis of his selection,
however, and since he computed no general relative, the changes
were of no especial significance. As already stated, this table was
not printed from 1874 until Dr. Conrad resumed it in 1891. (Jahr­
bucher, 3d series, vol. 1, pp. 916, 917.) In it he used the same 47
articles that Dr. Paasche had used.
In later issues of the Jahrbucher actual prices as well as relatives
appear for all articles up to the year 1891, inclusive. In 1892, how­
ever, neither actual prices nor relatives were shown for raw sugar,
i Report from the Committee on Finance of the United States Senate on Wholesale Prices, Wages, and
Transportation, Mar, 3,1893. 52d Congress, 2d session, Report No. 1394. Pt. I, pp. 297-301.




224

BULLETIN OF THE BUREAU OF LABOR STATISTICS,

silk, flax, hops, rapeseed oil, horsehair, or butter, and no prices or rela­
tives have appeared for these articles since that year, although they were
not finally eliminated from the table until the issue of the Jarbucher
for July, 1911. No explanation of the change is made.
All the articles embraced in this table are either raw materials or
semimanufactured materials.
DESCRIPTION AND GROUPING OF COMMODITIES.

No description of the articles in this table has accompanied it after
its first appearance in 1864. At that time Dr. Laspeyres gave a
detailed description, but his list of articles is not the same as is used
in later tables, and the time that has elapsed since its appearance
makes his description of no great present significance. It is there­
fore not reproduced. It may be found in the Jahrbucher, vol. 3,
pp. 89-92, inclusive.
As the table now stands, the description of articles is an exact copy
of the description found in Hamburgs Handel und Schiffahrt with
the following exceptions:
C otton .—The table as published in the Jahrbucher merely says
“ cotton” (B a u m w o lle ) , but in the original table the boxhead reads
“ Cotton and cotton waste” ( B a u m w o lle u n d B a u m w o ll-a b fa U ) . The
actual prices are those of “ Commerce and shipping” reduced to a
basis of 50 kilograms.
C oa l. —Dr. Conrad’s designation is “ Coal and coke” ( S teirikoh len
u n d K o k s ), but the official designation is “ Coal” (S tein J coh len ). Dr.
Conrad’s actual prices are the same as the official figures, reduced to
zentners.
I r o n , E n g lis h ba r. —This designation does not appear at all in the
official Hamburg figures. i ‘ Bar iron from all sources ’ ’ (S ta n g e n -E is e n
im G a n z en ) and “ English strap iron” (S ta n g e n -E is e n , e n g lisc h ) are
quoted. Comparison of the actual prices, as quoted by Dr. Conrad
in the Jahrbucher, shows that from 1881 to 1890 the quotation appear­
ing under the heading “ English bar” is “ English strap iron.” From
1891 to 1905 it is the quotation of the official figures for “ bar iron
from all sources.” From 1906 to 1910 it is again the quotation for
“ English strap iron.” So far as can be ascertained, no explanation
of this irregularity is offered.
No grouping of commodities is made in the table. No general
relative has been published since 1864. The relative for 1864 was
not weighted.
Index numbers computed from prices of 19 articles in 6 groups.

A second table of weighted relatives for six groups containing 19
articles, together with a combined weighted relative for all articles,
appears in current issues of the Jahrbucher, The weighted relative




INDEX NUMBERS-----GERMANY.

225

for all articles is in each issue compared with a simple relative based
on all the articles quoted in the official Hamburg prices for which the
continuity of the quotations is such as to make them usable. This
latter relative is discussed on pages 228-232 of this bulletin.
BASE PERIOD.

The weighted relative as now published is calculated on two bases
for purposes of comparison: 1847-1880 and 1871-1880. The table also
contains a relative for the years 1871-1880 with the period 1847-1867
as a base.
This table first appeared in Prof. Paasche’s study published in vol.
23,1874. At that time 1847-1867 was used as the base. This same
period was used as the base period by Dr. van der Borght in 1882, but
he also made a relative for the years 1876-1880, with 1847-1875 as a
base. When Dr. Conrad calculated this table in 1887 he reduced it
to its present form.
PRICES: H O W SH OW N AND COMPUTED.

As the table is now published relatives are shown for the 10-year
periods 1881-1890 and 1891-1900, for the 5-year periods 19011905, and 1906-1910, and for each year since 1905. By the use of
preceding issues a yearly relative for the present 19 articles is avail­
able beginning with 1888 and also a relative by 5-year periods begin­
ning with 1881-1885.
NUMBER AND CLASS OF COMMODITIES.

The first issue of the table embraced 22 commodities, all raw ma­
terials. As it now appears it contains 19 raw materials. The re­
duction from 22 to 19 articles really occurred in 1892, since in the
table of actual prices from which the relatives were made no quotation
has appeared for sugar, silk, or unforged zinc since 1891. The table,
however, was preserved in its original form until vol. 31 of the Jahr­
bucher, 1906 (containing the relative for 1904), when the table was
recast to include only the 19 articles now quoted. No explanation of
the reason for the omission of the three articles noted above was made,
but since they are not now quoted in Hamburgs Handel und
Schiffahrt, their omission from the official statistics undoubtedly
forced their omission also from Dr. Conrad's weighted relative.
DESCRIPTION AND GROUPING OF COMMODITIES.

The articles used by Dr. Conrad are described in his table exactly
as they appear in their original source (Hamburgs Handel und
Schiffahrt), with four exceptions:
1.
R i c e .—The quotation on rice is that for all rice, without dis­
tinction of kind.
94261°— Bull. 173—15------15




226

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

2. C otton .—In the Jahrbucher the quotation under the heading
“ Cotton” (B a u m w o lle ) is the price appearing in Hamburgs Handel
und Schiffahrt under the heading “ Cotton and cotton waste”
( B a u m w o lle u n d B a u m w o ll-a b fa ll).

3. F i s h o i l —This article appears in the Hamburg tables merely as
“ Tran,” in the Jahrbucher as “ Fischtran.”
4. C oa l. —The quotation in Hamburgs Handel und Schiffahrt is
for “ coal and coke” (S tein koT ih n u n d KoTcs), while in the Jahrbucher
the heading used is merely “ Coal” (S te in k o h le n ) . The price, how­
ever, is that for coal and coke.
Actual average prices are also shown in current issues of the
Jahrbucher by zentners (50 kilograms) in marks from 1847 to date.
The grouping of articles is as follows:
I. Coffee, cocoa, tea, pepper, and rice.
II. Cotton.
III. Indigo, saltpeter, fish oil, and palm oil.
IV. Iron (pig and cast), tin, copper, and lead.
V. Coal.
VI. Wheat, rye, barley, and oats.
WEIGHTING.

The index number under consideration was first weighted by Prof.
Paasche according to consumption in the German Empire in 1874.
He states that he determined the consumption quantities which he
used by various means. For those articles which were not pro­
duced in Germany it was possible to get fairly accurate data from the
import and export statistics of the German customs union. The arti­
cles which came to Germany exclusively by importation were coffee,
cocoa, tea, pepper, rice, cotton, indigo, saltpeter, fish oil, and palm oil.
Domestic production of sugar being at that time subject to a tax, it was
possible to get accurate statistics of the consumption of sugar from the
official imperial statistics (R eich ssta tistiJ c) . The same was true of the
mineral products, iron, zinc, tin, copper, lead, and coal. Prof. Paasche
stated that the least satisfactory were his statistics for the consump­
tion of grain. He found that it was absolutely impossible to make
accurate tables of grain production, and he therefore used the figures of
Hausner in his “ Comparative European Statistics” ( V erg leich en d e
S ta tis tic E u r o p a s ) , 1864.
The weighted relative was computed according to the method
of Drobisch. To illustrate, the relative for each group for the year
1868 was computed as follows:
German consumption for that year of each article in the group was
ascertained as indicated above. It was then multiplied by the aver­
age price for the base period, 1847-1867, and by the average price




INDEX NUMBERS---- GERMANY.

227

for 1868. The sum of the products for 1847-1867 was then ascer­
tained and likewise the sum of the products for 1868. The relative
for 1868 was the ratio of the sum of the products for 1868 to the sum
of the products for 1847-1867.
The relative for the entire table of 22 articles (now 19 articles) in
six groups was made in the same way. That is, the sum of all the
products for 1847-1867 was found and the sum of the products for
1868 was compared with it to get the general relative. Thus, the
textile group consisted of cotton and silk. The average price of
cotton for 1847-1867 was 26.83 thalers per zentner; for 1868, 25.92
thalers per zentner. The consumption of cotton in 1868 was 1,509,961 zentners. The average price of silk for 1847-1867 was 616.31
thalers per zentner; for 1868, 858.12 thalers per zentner. The con­
sumption of silk for 1868 was 22,088 zentners. The relative for the
group was therefore secured as below:
26.83X1,509,961 equals................................
616.31X22,088 equals...................................

40,512,254
13, 613,055
54,125,309

25.92X1,509,961 equals
858.12X22,088 equals..

39,138,189
18,954,155
58,092,344

Relative for 1868 equals...............................^4^125^309^" e<luals 107

The sum of the 22 products for 1847-1867 was 954,341,370; for
1868 it was 1,130,430,232. The relative for the 22 articles for 1868
113,043,023,200
,
therefore was — ’ 0^ -0~
4?r~ equals 118.5.
yo4,o41,o7U

The table was reweighted for each year from 1868 to 1872, inclu­
sive, according to the consumption for that year.
When Dr. Richard van der Borght computed this table in 1882,
he followed the same method as Prof. Paasche, weighting anew for
each year the figures for the years 1873 to 1880, inclusive. As to
the source of his statistics of consumption he states that for grainy,
instead of using Hausner’s results for the years 1873-1877, he sub­
stituted the figures of X. von Neumann-Spallart which had appeared
in his “ Review of International Traffic” in 1878. For the years
1878-1880 he used the imperial statistics of domestic grain produc­
tion which had become available.
The estimates of consumption of other articles were all based on
official imperial statistics (R e ic h s sta tis tik ), including statistics of
domestic production, and of export and import.
When Dr. Conrad computed the table for the years 1880 to 1886,
inclusive (Jahrbucher, vol. 15, pp. 322-331), he used the consumption




228

BULLETIN OP THE BUREAU OF LABOR STATISTICS.

statistics of 1880 for the entire table, having abandoned Dr. van
der Borght’s method of weighting each year separately. Dr. Conrad
has continued to weight this table according to the consumption
statistics of 1880 to the present time.
So far as can be ascertained, the general relative for the entire
19 articles was weighted for all years prior to 1906, inclusive, as
indicated in the preceding paragraphs. From 1907 to date, however,
the general relative has been obtained by computing the simple
arithmetical average of the relatives for the six groups.
TESTING.

It can not be said that any method of testing was applied to Dr.
Conrad’s weighted relative prior to 1889. From that year, however,
the weighted relative has been compared to the unweighted relative
for all articles with continuous quotations included in Hamburgs
Handel und Schiffahrt. For 1889 this simple relative embraced 318
articles. It now includes but 157 articles. Dr. Conrad has also
included in his periodical price studies since that date a relative
based on the official imperial quotations (R e ic h s s ta tis tik ), and in addi­
tion has reprinted Sauerbeck’s index as found in the Journal of the
Royal Statistical Society, London. These various relatives he has
utilized for comparison with his own work.
Index numbers computed from prices of 157 articles.

Beginning with volume 1, 3d series, of the Jahrbiicher for 18.91,
Dr. Conrad’s price studies have also included, as stated in the preced­
ing paragraph, a simple relative for all the articles contained in
Hamburgs Handel und Schiffahrt for which the quotations have
been continuous. The relative is printed in the same table as the
weighted relative described above and is used for purpose of compari­
son with it.
BASE PERIOD.

The base periods are the same as for the weighted relative already
discussed. A relative for the years from 1871 to 1880, with 1847-1867
as the base, and two relatives for the years from 1881 to date are
shown. Of the latter the first has as its base the period 1847-1880,
while the second is computed upon 1871-1880 as a base.
PRICES: H O W SHOW N AND COMPUTED.

In the latest available issue of the Jahrbucher (June, 1914), rela­
tives are shown for each of the decades 1881-1890 and 1891-1900,
for the five-year periods 1901-1905 and 1906-1910, and for the years
1906 to 1912 separately. By the use of preceding issues it is possible
to get a yearly relative beginning with 1886 and a relative by fiveyear periods beginning with 1881-1885.




INDEX NUMBERS---- GERMANY.

229

NUMBER AND CLASS OF COMMODITIES.

As now printed the index number includes 157 articles, largely raw
materials, but including also some manufactured and semimanufac­
tured articles, as varied in character as possible. A yearly relative
on the present basis of 157 articles is available only from 1902 to date.
This is occasioned by the fact that when the relative was recomputed
on its present basis the years prior to 1902 were grouped.
When first computed for 1889 and preceding years 318 articles were
included. The relative for 1891 and preceding years, which appeared
in volume 5, third series, embraces 320 articles. However, by the time
that Dr. Conrad prepared his relative for 1892 the recomputation of the
Hamburg prices necessitated by the entry of Hamburg into the Ger­
man customs union, to which repeated reference has been made, had
taken place, and Dr. Conrad found it necessary to recompute his rela­
tive on the reduced basis of 163 articles.
Dr. Conrad's statement concerning this is as follows:
It became necessarv for us to reduce the number of articles included
in our computation of the arithmetical mean from 320 to 163 articles,
and on this basis we have available figures from 1847 to date. This
fact is bound to have an appreciable effect on our results. The de­
crease in prices as a result of this in recent times is rendered less impor­
tant, while the rise in prices in the seventies compared to 1847-1867
becomes greater. Thus with the old list of articles the relative for
1871-1880, with 1847-1867 as a base, is 104. With the restricted
list it is 111. With 1847-1880 as a base the average of 320 articles
for the decennium 1871-1880 has a relative 79, but the relative for the
new list of 163 articles is 95.
The difference is yet greater for the years 1890-91, for which the rela­
tive according to the old method on the base 1847-1880 was 74,
but is now 95, and on the base 1871-1880 was 72 and is now 85 plus.
The year 1892, so far as comparison is possible, with the base 18711880 would have shown a decrease of from 70 for 1891 to 68 for 1892.
With the present list of articles, however, the relative for 1892 is 82.7
as against 87.4 for 1891.
This is a difference of about 15 per cent, arising from the fact that
the prices of domestic products had decreased much more than the
prices o i those imported by sea. This difference will serve to warn
against the putting of too great faith in these tables, for they show
how results differ according to the number of articles considered and
show how necessary it is to use as large a number of articles as pos­
sible.
With the publication of the relative for 1906 and preceding years in
volume 34, third series, of the Jahrbucher, 1907, the number of articles
was further reduced to 158 and the relative was recomputed from the
beginning on the new basis. Dr. Conrad does not state what articles
were dropped. When the index for 1907 was printed the number was
still further reduced to 157, without recomputation, where it now
stands.



230

BULLETIN OP THE BUREAU OF LABOR STATISTICS.
DESCRIPTION AND GROUPING OP COMMODITIES.

No list of included articles is printed in the Jahrbucher. The 157
articles, however, are from the following list of 174 articles for which
average annual prices from 1850 to date appear in “ Hamburgs Handel
und Schiffahrt.” The quotations for the articles marked with an
asterisk (*) are in some degree imperfect, and it seems safe to con­
clude that the excluded articles are among those so designated:
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.

Aloes.
Antimony.
Oranges.
Arrack.
Asphalt.
Balsam of copaiba.
Peruvian balsam.
Cotton and cotton waste.
Bay leaves.
Tin plate.
Sheet iron.
Lead.
White lead.
Borax.
Bristles.
Cinchona bark.
Cochineal.
Divi-divi.
Iron wire.
Pig and cast iron.
Bar iron from all sources [im

22.
23.
24.
*25.
26.
27.
*28.
*29.
*30.
31.
32.

Strap iron, English.
Sheet billets, Swedish.
Iron rails and fishplates.
Elephant’s tusks and ivory.
Peas.
Extract of logwood.
Extract of redwood.
Extract of yellowwood.
Extract of quercitin (dyer’s oak).
Figs.
Deerskins, doeskins, and reindeer
skins.
Calfskins.
Sheep and goat skins.
Dried fish.
Meats, fresh and cured.
Meat extracts.
Nutgalls.
Cotton yam.
Coconut fiber yarn.
Jute and manila hemp yam.
Linen yarn.

Ganzeri).

33.
34.
35.
36.
*37.
*38.
39.
*40.
*41.
42.




43.
*44.
45.
46.
47.
48.
49.
50.
*51.
*52.
53.
54.
55.
56.
57.
58.
59.
60.

Woolen and half-woolen yarn.
Casings (i. e., for sausages).
Yellow metal and brass.
Gin.
Barley.
Plate glass.
Natural guano.
Gum arabic.
Gum benzoin.
Dammar.
Raw rubber.
Copal.
Gutta-percha and chicle.
India-rubber shoes.
Dried and salted hides.
Oats.
Hemp.
Resin and galipot.

61. Herring.

62.
63.
64.
65.
66.
67.
68.
*69.
70.
71.
72.
73.
74.
75.
*76.
77.
78.
79.
80.
81.
82.
83.
84.
*85.

Logwood.
Yellowwood.
Redwood.
Ebony wood.
Rosewood ( Jacaranda).
Corkwood.
Mahogany.
Walnut wood.
Cedar wood.
Honey.
Ox and cow horns.
Indigo, natural and manufactured.
Raw ginger (zinziber).
Iodine and iodine preparations.
Jute.
Cheese.
Coffee, raw, without designation of
kind (im Ganzen).
Coffee, raw, Brazil.
Coffee, raw, San Doiningo.
Coffee, raw, Java.
Coffee, raw, La Guaira.
Coffee, raw, Porto Rico.
Cocoa.
Potassium monochromate.

INDEX NUMBERS---- GERMANY.

86.
87.
88.
89.
90.
*91.
92.
93.
94.
95.
96.
97.
98.
99.
100.
101.
102.
*103.
104.
105.
106.
107.
*108.
109.
110.
*111.
112.
113.
114.
115.
116.
*117.
*118.
119.
*120.
121.
122.
123.
124.
*125.
126.

Camphor.
Cinnamon.
Cassia lignea and cassia vera.
Catechu, brown and yellow.
Bones.
Bone ash.
Bone charcoal and bone meal.
Cognac.
Currants.
Corks.
Madder.
Copper.
Licorice.
Leather.
Candles.1
Maize.
Almonds.
Manila hemp, sisal, etc.
Nutmeg flowers.
Nutmegs.
Nails, iron.
Cloves.
Corozo nuts and coconuts (for use
in making buttons, etc.).
Walnuts and hazelnuts.
Castor oil.
Cottonseed oil.
Coconut oil.
Linseed oil.
Olive oil.
Palm oil.
Turpentine.
Oil cake.
Paraffin and vaseline.
Mother-of-pearl shells.
Refined petroleum.
Pepper.
Phosphorus.
Piassaba.
Allspice.
Quicksilver.
Rice, kind not specified (i. e., im

*127.
128.
129.
130.

Rice, Japanese.
Rice, Java.
Rye.
Raisins.

Ganzeri).

131.
132.
133.
134.
135.
*136.
137.
138.
139.
140.

141.
142.
143.
144.
145.
146.
147.
148.
149.
150.
151.
152.
153.
*154.
155.
156.
157.
158.
159.
160.
161.
162.
163.
164.
165.
166.
167.
168.
169.
170.
*171.
172.
173.
174.

231

Rum.
Grass seed.
Clover seed.
Flaxseed.
Rapeseed (Raps Undrubsaat).
Sesame seeds.
Sago and tapioca.
Saltpeter.
Anchovies.
Sardines.
Shellac, and gum lac.
Slate.
Grease.
Sulphur.
Sail twine (canvas yarn).
Soda, calcined and crystallized.
Steel.
Stearine.
Coal.

Rattan.
Sumac.
Tobacco, without designation as to
origin (im Ganzeri).
Tobacco, San Domingan.
Tobacco, Cuban.
Tobacco, Porto Rican.
Tallow.
Cordage, new.
Tea.
Fish oil.
Wax.
Baleen.
Spermaceti and margarine.
Wines without designation of origin
or kind (im Ganzeri).
Wines, exclusive of champagne.
Champagne.
French wines.
Portuguese wines.
Spanish wines.
Tartar.
Wheat.
Woolen waste and shoddy.
Sheep’s wool.
Tin.
Lemons.

1 This term is not further defined in current issues. The word “ Lichte,” which in technical usage is
equivalent to “ Kerzen,” appears without qualification. In the issue for 1889 it was defined as
“ Stearin-Lichte,” i. e., composite candles; but it can not be safely inferred that at present composite
candles only are included,since the technical term “ Lichte” covers all sorts of candles, as tallow,
paraffin, wax, etc.




232

BULLETIN OF THE BUREAU OF LABOR STATISTICS.
TESTING.

The only test of this relative so far as is shown in the Jahrbucher is
its comparison with Dr. Conrad’s weighted index, with Sauerbeck’s
index for England, and with that which Dr. Conrad bases on the
German imperial statistics (ReichsstatistiT c ).
Index numbers based on prices of the German customs union.

In addition to the three series of index numbers described in the
foregoing paragraphs Dr. Conrad regularly includes in his annual
study of prices a table of actual and relative prices derived from
official quotations of the German customs union. This table first
appeared in the Jahrbucher when Dr. Conrad took up the work of
studying German prices, as published in the fifteenth volume of the
new series, 1887 (forty-ninth volume of entire series). This index
has usually appeared with those based on Hamburg prices, but in a
few cases has been issued separately. In general it has appeared
more regularly than the Hamburg index.
SOURCE OF QUOTATIONS.

In the volume of the Jahrbucher in which this table first appeared
Dr. Conrad merely states that “ Table III, which follows, presents the
movement of prices from 1871 to 1886, as they are so satisfactorily
published in the official imperial statistics, and we compare the periods
1879-1882 and 1883-1886.”
In the next issue, however, the table is credited to the “ Monthly
Statistical Journal of the German Empire” ( M o n a ts h e fte d er S ta t is t ic
d es D e u ts c h e n R e i c h s ), and it continued to be so credited until the
publication of the table for the year 1892 (Jahrbucher, 3d series, vol.
61, entire series, 1893). From that date to the present it has been
from the “ Monthly Statistical News of Foreign Commerce in the Ger­
man Customs Union” ( S ta tistis ch e N acJ iw eise u b e r d e n a u sw d rtig en
H a n d e l d es d eu tsc h e n Z o llg e b ie ts ).
BASE PERIOD.

As at present issued the index is computed upon two bases—
1879-1883 and 1879-1889. The period 1879-1882 was the base as
the table was originally printed. It was changed to 1879-1883 with
the presentation of the table for the year 1888 (Jahrbucher, vol. 18,
whole series vol. 52, 1889). The second index, with the base 18791889, was added when the table was printed for 1890 (Jahrbucher,
3d series, vol. 1, whole series vol. 56, 1891).




INDEX NUMBERS---- GERMANY.

233

PRICES: HOW SHOW N AND COMPUTED.

On the base 1879-1883 the table as now published1shows relatives
by five-year periods from 1884 to 1913, inclusive, and annual rela­
tives for years since 1908. By the use of preceding issues of the
Jahrbucher an annual relative is available from 1888 to date, except­
ing for 1889 and 1902. For these years relatives were not computed,
although the actual prices were available.
On the base 1879-1889 the table shows relatives by five-year
periods from 1889 to 1913 and annual relatives from 1908 to date.
By the use of preceding issues an annual relative is available from
1890 to date, except for 1889 and 1902, as noted above.
Actual prices are presented in the same way, and are available for
every year of the period covered.
NUMBER AND CLASS OF COMMODITIES.

The table as constructed embraces quotations on 33 articles. In
this number are included two designations each for sugar, coffee,
tobacco, and cotton yarn. However, for the years since 1909 the
quotation on herring is lacking, while from 1907 to date no quotation
on copper is shown. Therefore the relative as now printed is actu­
ally based on 33 articles from 1884 to 1906, 32 articles from 1907 to
1909, and 31 articles for years since 1909.
The relative, however, is not computed on 31 quotations but on 103,
as for many articles several quotations are utilized in making up the
average annual price on which the relative is based. There has been
a slight variation in the number of quotations from time to time, but
no change of any significance. As first issued the relative was based
upon 96 quotations.
Of the 33 articles in the list some are raw materials, others are fin­
ished manufactures, while a few are semimanufactured products.
DESCRIPTION AND GROUPING OF COMMODITIES.

The articles are not divided into groups. They are described as
follows in the current issue of the Jahrbucher:2

1.
2.
3.
4.

Wheat, 14 quotations (1892 to date).
Eye, 14 quotations (1892 to date).
Barley, 15 quotations (1892 to date).
Maize, 5 quotations (1892 to date).

i Jahrbucher fiir Nationalokonomie und Statistik, 102. Band (III. Folge, 47. Band), Heft 6. Juni, 1914,
p. 800.
Idem, pp. 799-801.




234

5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.

BULLETIN OF THE BUREAU OF LABOR STATISTICS,

Oats, 14 quotations (1892 to date).
Wheat flour, 6 quotations (1892 to date).
Rye flour, Berlin.
Rapeseed oil, Berlin.
Potato alcohol, Berlin (1892-1903), Hamburg (1904 to date).
Raw sugar, Magdeburg.
Refined sugar, Magdeburg.
Coffee, Rio, good ordinary, Bremen (Sabanilla, 1896 to date).
Coffee, plantation, Ceylon, medium, Frankfort on the Main.
Rice, Rangoon table, Bremen.
Pepper, Bremen.
Herring, Norwegian, Hamburg (Scotch, 1904 to date).
Leaf tobacco, ordinary Kentucky, Bremen.
Leaf tobacco, second-grade Brazil, Bremen.
Cotton, middling upland, Bremen.
Wool, Berlin.
Hemp, Liibeck.
Raw silk, Milan organzine, Krefeld.
Cotton yarn, Nos. 40-120, Krefeld.
Cotton yam, warp 16, Miihlhausen, in Alsace.
Cotton cloth, Miihlhausen, in Alsace.
Linen yam, No. 30, flax yam, Bielefeld.
Lead, 6 quotations.
Copper, Mansfeld, Berlin
Zinc, 5 quotations.
Tin, 3 quotations.
Pig iron, Scotch No. 1, Hamburg (up to 1900, inclusive, Berlin).

32. Petroleum, Hamburg (up to 1900, Bremen), in bond.

33. Coal, Westphalian, Berlin.

SUBSTITUTIONS AND ADDITIONS.

Such substitutions as have been made are with respect to grade or
place of quotation and are indicated in the description above. No
additions to the list of articles have been made.
TESTING.

This relative is presented for purposes of comparison with the two
other general relatives published currently in the Jahrbucher, namely:
The weighted relative of 19 articles (based on Hamburg quotations)
and the simple relative of 157 articles (also based on Hamburg
quotations).
Dr. Conrad also reprints Sauerbeck’s index for comparison with
his own, and in addition compares them with other studies which
appear from time to time.




235

INDEX NUMBERS---- GERMANY.
TABLES OF RESULTS.

The following tables, compiled mainly from the issue of the
Jahrbucher for June, 1914, show the principal index numbers com­
puted by Dr. Conrad:
RELATIVE PRICES OF 39 SELECTED ARTICLES (BASED ON HAMBURG TRAD E STA­
TISTICS), B Y SPECIFIED PERIODS, 1871 TO 1910, AND B Y YEA R S, 1906 TO 1912.1
(Base period, 1847-1870=100.)

Article.

18711880

18811890

Coffee (Brazil).
Cocoa...............
Tea...................
Currants..........
Raisins.............
Almonds..........
Pepper.............
Coconut oil......
Palm oil...........
Indigo..............
Mahogany........
Cotton..............
Hemp...............
Rice.................
Wheat..............
Rye..................
Barley..............
Oats.................
Clover seed......
Rape and rapeseed...............
Linseed oil.......
Calfskins..........
Bristles............
W ax.................
Tallow..............
Fish oil............
Lard.................
Herrings..........
Iron, pig..........
T i n . . . . . ..........

155.13
116.53
86.57
89.50
99.51
110.91.
140.35
80.13
100.46
111.41
92.56
81.03
98.01
81.43
104.38
106.26
127.79
109.97
115.02

120.29 125.62 75.90
130.98 119.07 109.48
67.20 52.96 53.98
80.33 63.44 79.07
8 8 .1 0
79.43 94.19
1 11 .22 101.34 101.56
182.93 104.65 162.61
68.54 61.26 70.88
71.04 57.61 62.71
93.36 77.48 54.23
83.14 68.59 56.97
62.53 45.95 48.24
85.55 80.37 90.13
68.15 59.36 57.78
76.26 61.73 60.73
82.54 71.34 66.46
89.95 57.76 51.34
89.34 78.07 82.24
98.00 82.46 85.84

18911900

19011905

19061910

1906

1907

1908

1909

1910

1911

83.37 82.99 73.75 81.84 84.55 102.28 130.44
111.84 103.48 154.58 118.15 97.66 95.80 97.05
58.83 60.86 59.90 51.04 61.75 60.48 63.50
96.39 90.88 97.40 95.50 94.32 104.34 109.77
102.87 99.51 118.89 102.76 89.10 108.44 119.52
122.96 89.96 125.70 114.17 126.09 128.26 135.48
127.35 144.41 135.27 115.40 114.99 126.64 143.51
76.93 71.92 77.30 71.48 72.40 85.11 78.08
74.67 69.47 78.41 66.63 69.15 85.31 83.16
40.85 47.15 51.89 38.24 33.75 34.58 43.95
42.86 48.53 44.29 44.13 37.11 43.28 57.06
49.70 48.57 48.87 50.52 45.23 55.55 29.43
91.69 83.95 81.88 97.46 96.73 101.17 97.15
65.39 62.24 68.46 69.84 64.47 62.62 74.60
73.24 63.65 67.31 73.97 84.11 73.97 71.69
80.10 70.84 82.60 90.61 89.96 73.34 76.10
67.84 63.35 74.15 71.36 69.90 62.62 72.69
88.93 68.47 101.64 86.48 91.53 86.89 8 8 .6 6
93.58 78.51 86.80 94.09 90.02 108.48 114.58

97.88 84.99 72.13 68.52 74.68 86.16
90.54 69.48 6 6 .2 2 78.36 77.26 68.06
103.00 74.42 59.58 70.38 87.89 94.10
155.22 145.76 88.75 74.42 81.07 82.12
76.98 54.12 60.99 74.41 75.45 81.98
89.74 74.04 57.52 6 6 .1 1 74.78 68.82
82.17 66.65 49.24 52.95 55.64 49.77
86.92 78.25 62.50 75.88 87.92 76.23
121.94 109.15 103.77 110.82 117.54 123.41
117.71 76.57 78.20 83.38 97.55 94.55
96.32 84.72 84.43 95.06 108.85 105.33
8 8 .0 2
63.91 57.20 67.35 68.90 70.80
.............. 112.19 83. 77 86.44 95.06 102.94 112.18
Quicksilver----- 129.54 83.40 87.16 102.43 91.37 87.74
Coal and coke.. 109.88 77.78 85.85 87.65 87.65 83.95
Saltpeter.......... 96.71 73.71 56.30 65.62 69.05 75.14
Bar iron (Eng­
lish).............. 113.53 72.37 13.31 8 6 .6 8 81.37 76.69
Cotton yarn— 115.60 105.39 87.32 95.66 113.48 114.12
Woolen a n d
half-woolen
7 am .............. 101.73 69.98 61.72 54.88 65.61 65.92
Lmen yam ...... 80.55 98.17 109.91 116.56 117.10 116.30

78.00 73.82 70.56
75.24 76.86 76.57
82.66 86.48 87.31
78.52 77.91 80.76
83.06 72.42 67.14
74.76 78.47 73.51
57.55 51.97 54.30
83.36 84.71 98.89
126.40 111.75 110.35
97.28 97.55 96.46
111.65 100.27 107.18
81.95 65.36 65.77
116.76 102.59 96.23
84.08 92.43 94.67
93.83 92.59 83.95
75.70 68.98 65.90

76.61 73.75
96.50 115.13
88.76
85.73 93.92
72.72 80.79
78.69 78.67
62.10 65.61
108.59 89.45
115.11 117.26
104.09 96.46
118.15 143.34
63.89 63.46
91.14 86.49
95.65 98.04
86.42 82.72
62.11 67.37
8 6 .77

1912

150.41
102.30
57.56
110.71
1 2 2 .8 8

138.64
164.27
72.83
81.78
35.94
7.78
51.85
53.27
69.84
76.53
87.48
86.53
103.69
111.24
85.78
102.08
85.20
98.92
82.48
84.69
56.68
99.65
176.49
103.81
153.85
72.31
98.24
92.25
95.06
74.58

83.39 88.35 90.74 74.92 92.51 90.74
109.64 110.96 133.41 121.13 127.38 132.44
68.72 63.18 62.90 66.38
121.75 117.93 111.49 117.35

67.65
121.11

63.32
125.60

i Jahrbucher fur Nationalokonomie und Statistik, 102. Band (III. Folge, 47. Band), Heft 6 . Juni, 1914.
p. 796.




236

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

RELATIVE PRICES OF CERTAIN GROUPS OF ARTICLES W EIGHTED IN PROPORTION
TO THEIR CONSUMPTION (BASED ON HAMBURG TRADE STATISTICS), B Y SPECI­
FIED PERIODS, 1871 TO 1910, AND B Y Y EA R S, 1906 TO 1912.1
(Base period, 1847-1880=100.)

Prices
18711880
Article.

on
18471867=

18811890

18911900

19011905

19061910

1906

1907

1908

1909

1910

72.31

87.35

75.47

77.31

80.41

53.54

47.93

58.88

62.9

54.95

39.91

36.26

38.09

46.26

39.06

91.68

96.96 107.71 117.90

1911

1912

100).

Coffee, Brazil..
Cocoa...........
Tea..............
141.66 100.46 101.50
Pepper.........
Rice............

66.82

79.69

92.32

n.
81.84

66.28

•101.65

78.87

111.!

72.97

Coal..

109.88

Wheat.
Rye___
Barley.
Oats. . .

Cotton............

48.70

51.07

m.
Indigo___
Saltpeter..
Fish o il...
Palm oil..

61.15

42.73

63.36

73.26

79.51

94.48

75.90

83.74

86.03

85.54

112.51

78.01

63.90

63.40

Average.. 105.54
A rith m etical
mean, com­
puted from
157 Hamburg
average prices 111.31

77.43

68.44

67.93

71.31

70.55

91.70

84.10

76.37

83.03

80.52

IV.

Pig iron...
T in...........

19.50 102.72

Le:
81.93

84.34

I

80.72

92.77

72.52

72.60

76.02

67.14

76.42

70.10

18.52

70.79

76.04

89.47

82.07

80.22

82.87

84.80 2 86.51

79.29
79.38

1 Jahrbucher fur Nationalokonomie urid Statistik, 102. Band (III. Folge, 47. Band) Heft 6. Juni, 1914,
p. 798.
since the separate listing of several qualities of an article,
2 Computed from prices of only 134I articles,
i
such as tea, coffee, rice, wool, oil, etc.,, has been discontinued.




237

INDEX NUMBERS---- GERMANY.

RE LA T IV E PRICES OF CERTAIN GROUPS OF ARTICLES W EIGH TED IN PROPORTION
TO TH EIR CONSUMPTION (B A SE D ON HAMBURG TR AD E STATISTICS), B Y SPECI­
FIED PERIODS, 1871 TO 1910, AND B Y Y EA R S, 1906 TO 1912-Concluded.
(Base period, 1871-1880=100.)

Article.

Prices
18711880
(based 1881on
1890
18471867=
100).

18911900

19011905

19061910

1906

1907

1908

1909

1910

1911

1912

I.
Coffee, Brazil..
Cocoa...............
Tea................... 1
Pepper.............. j
Rice................. I

.

79.43

88.23

52.62

74.09

57.16

81.21

70.16

71.87

74.76

82.72

85.82

79.63

56.69

59.44

61.39

59.92

60.29

62.32

55.79

68.61

73.40

64.04

78.97

61.69

61.22

40.28

93.43

49.57

37.64

34.19

35.91

43.63

36.83

67.19

67.48

73.21

96.76

82.49 105.18

90.83

93.90

99.30 110.31 120.74

70.79

78.09

80.23

78.76

76.40

85.39

84.27

76.40

78.65

75.28

86.52

76.37

62.65

62.07

68.88

65.65

75.48

71.56

74.94

66.18

68.60

78.16

74.86

66.19

65.56

70.30

68.12

75.52

69.46

07.85

70.57

75.66

78.69

85.18

78.47

71.07

77.30

74.87

83.19

76.51

74.74

77.21

78.88 1 80.47

n.
Cotton..............
in.
Indigo________
Saltpeter..........
Fish o il............
Palm oil...........
IV.

P% iron... T,,, .
Tin...................
Copper..............
Lead.................
V.

Coal..................
VI.

Wheat..............
1-.....
Rye................... J
Barley.............. f .........
J
Oats.................
Average
A rith m etica l
mean, com­
puted. from
157 Hamburg
average prices

i Computed from prices of only 134 articles, since the separate listing of several qualities of an article,
such as tea, coffee, rice, wool, oil, etc., has been discontinued.




BULLETIN OF THE BUREAU OF LABOR STATISTICS.
riV E PRICES OF SELECTED ARTICLES 1890 TO 1913(BASED ON ACTUAL A VEI
[Source: Jahrbucher fiir Nationalokon®mie und
(Base i

araal

Article.

im-

1890

1891

1894

1896

1897

er.

1
2

3
4
5
6
7
8
9
10
11

12
13
14
15
16
17
18
19
20

21
22

23
24
25
26
27
28
29
30
31
32
33

75.78 83.37 92.24 104.15
Wheat—141 quotations............... 102.59 118.95
82.70
79.09 80.01 85.35 97.77
Rye—141 quotations................... 110.27 136.65
89.15 95.43 99.96 107.92
Barley—151 quotations............... 113.03 112.12
84.84 69.69 68.09 76.
Maize—51 quotations.................. 91.53 116.60
88.61 94.01 101.85 110.00
Oats—131 quotations.................. 116.57 115.49
74.62 78.
Wheat flour—61 quotations....... 100.60 115.14
88.20 96.88
79.75 78.78
Rye flour, Berlin3....................... 112.95 140.41
89.28
79.
Rapeseed oil, Berlin.................... 120.15 110.99
105.63 107.75
Potato alcohol, Hamburg4......... 112.79 140.46
39.85 44.27 36.48 39.10
Sugar, raw, Magdeburg.............. 63.99 67.51
83.14
66.49 72.69 69.74 70.36
Sugar, refined, Magdeburg.........
83.98
Coffee, Rio,5 good ordinary, Bre­
men........................................... 153.66 140.79 124.89 141.96 140.79 139.07 124.39 96.73 70.19
Coffee, plantation, Ceylon,6 me­
dium, Frankfort on Main......... 120.70 120.37 115.75 120.03 112.78 108.38 105.20 98.87 85.33
Rice, Rangoon, table, Bremen.. 103.13 102.01 95.58 86.51 83.56 76.85 79.76 86.73 104.47
61 52.92 49.07 39.90 39.90 40.20 54.04 73.26
Pepper, Bremen.......................... 88.57
Herring, Norwegian, Hamburg.. 84.82 111. 79 86.46 61.16 77.64 107.30 76.28 96.85 92.94
Leaf tobacco, Kentucky, ordi­
85.62 101.31
76.98 67.64 66.34 81.16
nary. Bremen........................... 68.78 75.
Leaf tobacco, Brazil, second
grade, Bremen......................... 136.45 120.16 95.56 83.24 71.93 67.63 88.72 107.52 120.67
Cotton, middling upland, Bre­
men .......................................... 99.64 76.99 107.38 76.76 63.15 63.64 71.44 65.55 54.78
Wool, Berlin............................... 94.37 89.50 81.30 76.14 70.81 73.16 77.04 72.34 76.88
Hemp. Liibeck........................... 91.67 85.61 83.43 95.08 106.58 107.66 105.72 101.53 107.05
77.73 83.56 102.88 68.44 77.30 73.81 70.48 71.63
Raw silk, K refeld...................... 95.
Cotton yarn, Nos. 40-120,Krefeld. 100.94 91.61 81.55 88.47 79.87 81.13 90.57 80.29 76.73
Cotton yam, warp, 16, Mulhausen, in Alsace......................... . 90.50 81.56 78.21 84.92 78.77 77.00 81.56 75.20 64.80
89.56 106.
92.61 95.00 94.78 79.56 76.52
Calico, Miilhausen, in Alsace___ 110.13
Linen yam, No. 30, flax yarn,
Bielefeld.................................. 88.63 89.57 90.52102.84 100.57 85.31 84.55 82.84 82.84
Lead—61 quotations................. 100.89 95.14 81.73 75.97 73.64 80.13 85.41 93.80 99.33
77.27 82.78 83.86 89.18
Copper, Berlin.......................... 101.48 93.45 83.95 79.44 70.
Zinc—51 quotations.................. 143.22 143.32130.50 107.19 96.05 91.04 101.25 108.20 127.04
62.66 63.47 73.46
1.40 94.76 97.68 92.34 73.86 66.
Tin—31quotations....................
Pig iron, Scotch, No. 1, Ham­
94.32 94.04 92.64
burg........................................ 120.35 106.08 95.04
.12 88.92 82.00 70.32 79.88
1.76 84.86 79.02 62.93
Petroleum, Hamburg, in bond.
Coal, Westphalian, Berlin......... 132.60 129.40 117.34 114.13 114.68 114.58 113.64 114.41 117.56
Arithmetical average......... 105.71 98.14 95.32 91.52 82.54 81.75 81.82 82.65 84.04




1 Present number has varied from time to time.
2 Relative not computed for this year.

239

INDEX NUMBERS-----GERMANY.

PRICES AS SHOWN B Y OFFICIAL STATISTICS OF THE GERMAN CUSTOMS UNION),
tik, 3. Folge, 1, 5, 11,17, 24, 31, 34, 44, 46, and 47.]
1 87 9 -1 8 8 9 = = 1 0 0 .)

1900

1901

85.51 90.04
93.13 96.12
104.48 106.36
89.67 91.84
102.30 107.34
76.57 81.32
93.33 91.16
111.15 104.80
96.71 80.99
41.71 35.94
78.55 85.57

1902

1903

1904

1905

1907

1909

1910

1911

1912

1913

112.99
121.16
113.57
133.98
120.34
102.10
114.90
126.60
66.90
38.70
60.30

124.32
113.67
109.11
125.37
126.51
111.52
107.54
103.54
50.50
39.79
61.56

111.66
99.61
99.63
111. 12
113.67
100.32
92.80
104.98
50.12
46.27
69.97

108.85
110.41
116.25
110.82
137.02
98.90
103.04
117.13
48.85
47.34
68.37

115.66
122.20
126.23
127.72
142.78
100.04
110.10
120.94
65.92
46.21
71.41

108.33
108.47
106.13
110.40
120.96
98.58
100.77
121.67
70.94
34.83
57.73

62.61

66.60

78.95 110.59 128.14 109.97

12

75.93 75.81 76.85 76.34 77.41 73.33
93.07 94.62 95.98 104.96 104.46 97.01
99.64 96.57 89.15 76.56 54.20 56.83
97.71 116.43 148.26 112.83 101.11 128.38

82.15 96.35 104.64 99.83
97.94 110.63 129.71 114.66
64.11 76.35 89.37 83.29

13
14
15
16

(2)
<2;
(2)
(2)
(2)
(2)
(2)
(2)
(2)
(2)
(2)

85.55 92.55 89.92
89.70 90.31 93.85
96.69 100.96 102.97
91.99 94,35 97.26
97.71 97.68 105.57
79.30 83.66 82.02
86.85 84.82 98.88
87.63 82.45 85.45
33.90 37.40 40.26
78.74 58.72 63.01

110.62
124.08
111.52
113.08
132.77
99.15
122.52
133.95
55.89
31.33 31.63
54.42 56.51

78.54 58.70 (2)

51.62 66.55 69.55

70.56

83.89 76.30 (2)
94.59 96.38 (2)
104.88 102.41 (2)
137.80 125.04 (2)

72.90
99.46
104.49
109.75

Mar­
ginal
num­
ber.

1908

1906

95.04
102.16
105.99
103.14
119.44
85.68
101.50
103.85

63.83

1
2
3
4
5
6
7
8
9
10
11

(2)

112.54 93.82 92.86 109.51 136.56 156.13 124.47 145.62 155.41 153.98 141.31

17

125.75 100.09 (2)

84.28 89.81 98.86 109.99 137.73 133.58 108.65 108.49 141.08 146.98 151.24

18

99,34 104.59

89.89 77.19 (2)
92.10 75.84 <2)
121.33 132.05 (2)
83.40 73.24 (2)
166.77 89.73 (2)

100.20 108.47 85.61 99.15 106.08 93.57
98.11 97.56 102.23 114.23 114.78 103.71
118.53 111.54 113.29 122.07 129.57 124 . as
87.62 74.56 78.56 86.45 112.71 79.48
108.39 112.37 107.34 129.77 163.10 126.62

113.00
121.20
164.99
81.52
138.57

19
20
21
22
23

75.42 65.92 (2)
101.30 81.74 (2)

78.77 98.88 82.77 101.12 118.44 97.76 94.41 109.50 108.38 102.23 107.26
103.91 107.39 113.86 118.70 132.61 109.56 104.35 106.52 107.40 113.91 110.87

24
25

105.92 109.43
131.94 98.55
125.26 121.12
126.86 106.55
135.95 121.76

(2)
(2)
(2)
(2)
(2)

96.87 106.35 104.73
89.64 90.42 105.08
102.07 99.63 118.29
129.52 138.60 148.72
129.78 129.21 145.10

147.48 124.93 1*39.13 145.33 158.40 165.i6 144.91
177.16 137.42 138.28 157.32 194.08 212.03 207.38

26
27
28
29
30

97.76
127.72 134.51

(2)
(2)
(2)

105.30 98.11 100.57 110.17 120.02 105.42 103.18 104.64 103.44 119.23 125.60
97.89 89.58 84.44 92.88 94.33 102.64 93.21 81.20
116.89 116.62
122.86 122.86 129.06 125.95 131.14 132.52 128.38 127.00 *i27*00 111. 14 132.52

31
32
33

99.21 93.13

(2)

93.82 94.05 96.46 108.01 113.56 104.94 102.13 104.17 112.20 120.31 115.83

115.16
131.45
147.42
165.89
184.41

139.57 115.92
146.12 102.51

104.51
114.23
126.85
82.69
128.09

117.66
108.22
141.93
79.34
141.51

104.94
112.87
163.20
74.68
131.45

99.38 111.23 122.56 119.29 121.04
82.79 98.30 98.55 159.56 166.13

4 From 1892 to 1894, No. 00 with sack; from 1895 to date, No. 0/1.
8 Berlin, 1879-1903, Hamburg, raw, 1904 to date.




132.41
110.41
135.31
81.22
151.36

5 Coffee, Sabanilla from 1896 to date.
6 Java, 1879-1890.

240

BULLETIN OF THE BUREAU OF LABOR STATISTICS,

INDEX NUMBERS OF OTTO SCHMITZ.
PUBLICATION.

The data embodying the results of Mr. Schmitz’s original study
of the course of wholesale prices were published in book form in Berlin
in 1903.1 “ Those from 1903 to 1906 are extracted from manuscript
figures supplied by the author himself and in the possession of the
Royal Statistical Society. As regards later years the index numbers
have been taken from the supplement to the Zeitschrift fur Sozialwissenschaft (Monatliche Ubersichten liber die allgemeine Wirtschaftslage); only the general index numbers for all articles and for one
group (the metal group) are there given, however.” (Great Britain.
Report of an Enquiry by the Board of Trade into Working-class
Rents and Retail Prices, 1912. p. 354.)
HISTORY.

This study was undertaken in order to furnish a convenient means
of measuring the fluctuations in German prices. It was considered
highly desirable to construct an index covering a long period of years
and reliable data from a single source were not available. By using
two sources the author was able to cover the second half of the nine­
teenth century and to include six price waves, whose duration he
fixes as follows:
From 1849 to 1858.
From 1858 to 1870.
From 1870 to 1879.
From 1879 to 1886.
From 1886 to 1895.
From 1895 to 1912.
SOURCE OF QUOTATIONS.

The earlier series of index numbers (1851-1885) is based on quota­
tions of the bureau of trade statistics at Hamburg.2 Mr. Soetbeer,
who used figures from the same source, says in regard to the way in
which the quotations were obtained:
“ Throughout this period (1847-1885) Hamburg was an important
market for almost all raw materials. Moreover, it has been a free
port, without duties or differential taxes. Commodities imported are
declared in writing, with a statement of their weight and of their
ordinary trade designations. Their value is stated separately for
each commodity, either according to its price on ’Change that day,
or if there were no quotations, according to the probable price plus
the cost of importation. For consigned goods a careful estimate of
i Die Bewegung der Warenpreise in Deutchsland von 1851 bis 1902; nebst zwei Erganzungen: Bankdiskont, Goldproduktion und Warenpreisstand, der Weizenpreis, von 400 v. Chr. bis 1900. Von Otto
Schmitz. Berlin, 1903. 443 pp.
a In regard tp Hamburg Bureau ol Trade Statistics see also pp. 220 and 221.




INDEX NUMBERS---- GERMANY.

241

the prices sufficed, sometimes supplemented with a statement of their
insured value. These declarations, which were carefully supervised,
were then collected by the bureau of trade statistics, and tables were
made out of the quantity and value of goods exported and imported.” 1
The later index number (1879-1902) is based on data of the Imperial
Statistical Office, which publishes for each of the more important com­
modities several series of quotations representing a number of im­
portant markets and varieties. From these the author has selected
as a basis for his index number a single series from a representative
market, and in connection with the detailed description of the article
has stated the manner in which prices are obtained for the Imperial
Statistical Office at that market. These statements are retained with
the descriptions as given herewith. Eleven markets are represented:
Berlin and Bremen by 8 commodities each; Breslau and Hamburg
by 3 each; and Danzig, Magdeburg, Krefeld, Bielefeld, Dortmund,
Cologne, and Essen by 1 each.
In the few cases in which a hiatus occurs in the official series the
source of the substituted data is stated in connection with the descrip­
tion of the commodity.
BASE PERIOD.

The average of the 10 annual prices as given by the Imperial Sta­
tistical Office for the first 10 years covered by its reports (1879-1888)
was taken as the base for both series of index numbers. On the ground
that the period was a time of quiet and normal business development
and contained a period of falling prices, beginning in 1880, and a period
of rising prices, beginning in 1885, the average price was considered
comparatively normal and reasonably satisfactory as a basis.
The base period for lard is evidently the 9-year period, 1880 to 1888.
The Imperial Statistical Office did not publish quotations on lard for
1879 and no other quotations were substituted.
PRICES: HOW SHOWN AND COMPUTED.

The various methods of determining monthly averages are given
in the section on description and grouping of commodities. The
simple average method is used in computing the various averages for
the tables. All index numbers are printed with two decimals.
Data of the Imperial Statistical Office are in almost all cases
shown separately from Hamburg data and in much greater detail.
The principal table of the study shows for each article and group
of articles the index numbers for each month, quarter, half year,
and year from 1879 to 1900, while the corresponding table for Ham­
burg gives the index number for articles and groups by years only.
Actual prices of the later series are shown by months and of the
earlier by years.
i Bimetallism in Europe, by Edward Atkinson, p. 226.

94261°—Bull. 173—15----- 16



242

BULLETIN OF TH E BUREAU OF LABOR STATISTICS.
NUMBER AND CLASS OF COMMODITIES.

The index number based on the Imperial Statistical Office prices
(1879-1902) represents 29 commodities, while the earlier number
based on Hamburg prices (1851-1885) includes only 24 commodi­
ties—corn, linen yarn, petroleum, and one quotation each for iron
and coal being lacking. The list of 29 commodities is the original
list of the Imperial Statistical Office, with the following modifica­
tions: (1) Wheat flour, rye flour, refined sugar, cotton yarn, and
cotton cloth are omitted as not being strictly raw materials. Linen
yarn, however, is not omitted because the list does not include flax.
(2) Two kinds of iron and two of coal are carried as separate com­
modities instead of one description for each group. (3) Lard and
butcher’s meat are added as being important commodities introduced
early in the period covered by the figures of the statistical office.
For the years 1909 to 1912, inclusive, however, copper and one
description of pig iron are excluded.
DESCRIPTION AND GROUPING OF COMMODITIES.

The descriptions of commodities are based on the descriptions
of the selected varieties as given by the Imperial Statistical Office
and cover the years 1879-1902.1 Concerning the descriptions for
the period 1851-1885 the general statement is made that Hamburg
quotations are for standards of quality similar to those of the Impe­
rial Statistical Office.
The grouping of commodities appears in the list which follows.
The description of the first commodity is given in full, but the descrip­
tions of the others are somewhat abbreviated.
Group I.—Grains.

1. Wheat.

Official Berlin quotations. The average monthly price is ascertained by the com­
mittee of senior merchants of Berlin on the basis of daily quotations based on hear­
ings of the brokers. The early quotations are for 1,000 kilograms good, sound, yellow
wheat of all origins, 71.3 kilograms per hectoliter. On October 1,1887, the standard
was fixed at a minimum of 71.5 kilograms per hectoliter, and since January 1,1893, the
quotation is for good, sound, dry wheat, free from musty odor (hard wheat excluded),
of all origins, minimum 72.5 kilograms per hectoliter. In the author’s judgment
this repeated raising of the standard has exercised no demonstrable influence on
prices.
A further rise in the standard grading weight to 75.5 kilograms per hectoliter on
January 1,1893, was caused by the introduction of the new grain tester and was of
merely formal nature.
On account of the closing of the Berlin grain exchange there are no official quota­
tions for wheat, rye, and oats from January 1, 1897, to April 1,1899. Therefore the
official Breslau quotations, suitably adjusted (unter entsprechender Anpassung), are used.
For January, February, and March, 1899, information founded upon the daily publi­




1 No detailed description is available after 1902.

INDEX NUMBERS---- GERMANY.

243

cations of the central quotation office of the Prussian agricultural chambers was used
since the standard of quality was the same. From April, 1899, the official Berlin
quotations are again given. The index numbers calculated on the Breslau figures are
printed in italics.
1879, duty free.
1880-1885, duty per 1,000 kilograms gross, 10 marks.
1885-1887, duty per 1,000 kilograms gross, 30 marks.
1887-1891, duty per 1,000 kilograms gross, 50 marks.
1891 and following, duty per 1,000 kilograms gross, 35 marks.
2. Rye.

Official Berlin quotations. The monthly average price is ascertained as in the
case of wheat up to October, 1887. The quotations are for 1,000 kilograms good,
sound rye of all origins. The standard grading weight rose during the period from
65.9 to 67.8 kilograms per hectoliter without demonstrable influence on the price.
In the absence of Berlin quotations for the period January, 1897, to April, 1899, the
procedure was the same as in the case of wheat.
1879, duty free.
1880-1885, duty per 1,000 kilograms gross, 10 marks.
1885-1887, duty per 1,000 kilograms gross, 30 marks.
1887-1891, duty per 1,000 kilograms gross, 50 marks.
1891 and following, duty per 1,000 kilograms gross, 35 marks.

S.

Barley.

Breslau quotations. The price is ascertained every week day by the municipal
market commissioner. The monthly average price is derived from the Breslau
chamber of commerce. The quotation is for 1,000 kilograms medium heavy barley.
1879, duty free.
1880-1885, duty per 1,000 kilograms gross, 15 marks.
1885-1887, duty per 1,000 kilograms gross, 23 marks.
1887-1891, duty per 1,000 kilograms gross, 23 marks.
1891 and following, duty per 1,000 kilograms gross, 22.50 marks.
4 . Oats.

Official Berlin quotations. The monthly average price is ascertained as in the case
of wheat. The quotation is for 1,000 kilograms good, sound oats of all origins. The
standard grading weight has increased from 38.6 to 41.5 kilograms per hectoliter
without demonstrable influence on the price. In the absence of Berlin quotations
from January, 1897, to April, 1899, the procedure was the same as in the case of wheat.
1879, duty free.
1880-1885, duty per 1,000 kilograms gross, 10 marks.
1885-1887, duty per 1,000 kilograms gross, 15 marks.
1887-1891, duty per 1,000 kilograms gross, 40 marks.
1891 and following, duty per 1,000 kilograms gross, 28 marks.
5. Com (maize).

Bremen quotations. The price is determined every Saturday by the chamber of
commerce through licensed brokers, and the monthly average price is derived from
the medium prices .actually paid. The quotations are for 1,000 kilograms mixed
American corn of prime quality, in bond. On account of a shortage in corn, occa­
sioned by poor crops in America, there are no quotations from April, 1882, to March,
1883. There are likewise no quotations from September to December, 1884, because




244

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

only La Plata and Danube varieties were in the market. Although corn is quoted in
bond, duty rates are given for the sake of*completeness.
1879, duty free.
1880-1885, duty per 1,000 kilograms gross, 5 marks.
1885-1887, duty per 1,000 kilograms gross, 10 marks.
1887-1891, duty per 1,000 kilograms gross, 20 marks.
1891 and following, duty per 1,000 kilograms gross, 16 marks.
Group II.—Other products of agriculture and products of fishing.

6. Herrings.

Danzig quotations. The monthly average price is ascertained on the basis of the
prices actually paid every Saturday or on the last exchange day of every week as
determined by the bourse commission. The quotations are for 1 cask of 150 kilo­
grams gross, “ Crown and full,” in bond. The custom rate, which is given although
the article is quoted in bond, is 3 marks per cask.
7. Rapeseed oil.
Berlin quotations. The monthly average price is ascertained as in the case of wheat.
The quotations are for 100 kilograms good, raw rapeseed oil. Owing to the closing
of the Berlin produce exchange, January, 1897, to December, 1898, Konigsberg quo­
tations were used for these two years and continued through 1899 and 1900. Since
1901 Berlin quotations have been used again. The relatives for 1897 and 1898 are
printed in italics.
1879, duty per 100 kilograms gross, including container, 3 marks.
1880-1885, duty per 100 kilograms gross, including container, 4 marks.
1885 and following, duty per 100 kilograms gross, including container, 9 marks.
8. Alcohol.

Hamburg quotations. The average price is ascertained through the brokers by the
chamber of commerce on the basis of prices actually paid and the average of the
medium prices of every month. The quotation is for raw potato alcohol (per 100 liters
pure alcohol) in bond. For the sake of completeness the customs rates so far as could
be ascertained are appended.
January, 1879, to July, 1879, duty per 100 kilograms net, 36 marks.
July 5,1879 to 1885, duty per 100 kilograms net, 48 marks.
1885-1891, duty per 100 kilograms net, 80 marks.
July 1, 1891, to July 14, 1900, duty per 100 kilograms net, 125 marks.
July 14, 1900, and following, duty per 100 kilograms net, 160 marks.
9. Raw sugar.

Magdeburg quotations. The average price is determined on the basis of the highest
and lowest prices fixed on Friday of each week by commissions and commissioners
of senior merchants. The quotations are for 100 kilograms first quality. The descrip­
tion was 96 per cent polarization until October, 1887, 92 per cent yield until Feb­
ruary, 1897, and 88 per cent yield since that date. The first two descriptions are
considered as corresponding fairly well. The difference in value between 92 per
cent yield and 88 per cent yield is estimated at £ mark, but this difference is not taken
into account in calculating the index number. From September, 1888, the quota­
tion includes customs duty and excludes excise tax, and the index number has been
revised accordingly.
10. Butcher's meat.

Berlin quotations. Prices have been quoted regularly by the month since May,
1887. For the years 1881 to 1886 the yearly average prices were subsequently ascer­




INDEX NUMBERS---- GERMANY.

245

tained and communicated by the Imperial Statistical Office in December, 1889.
The price for 1881 is the average price for the months March to December. For 1879
and 1880 Hamburg prices form the basis of the index numbers.
The index number is based on the average of the quotations for beef, veal, pork,
and mutton. The earlier quotations were for 100 kilograms.
Beef, dressed weight, average of the prices for second grade.
Pork, live weight, with 20 per cent tare, highest quotation for second grade.
Veal, dressed weight, lowest quotation for first grade.
Mutton, dressed weight, medium of the prices for first grade.
Since July, 1897, the prices for beef, veal, and mutton are for 100 kilograms,
slaughter weight. Both dressed weight and slaughter weight are estimated on the
presumptive weight of the four quarters on which the price of the animal has been
apportioned, dressed weight with and slaughter weight without the deduction of
the value of the hide, head, feet, entrails, etc. The price based on slaughter
weight, according to information obtained by the Imperial Statistical Office from
authoritative sources, is about 8J per cent higher than the price based on dressed
weight. In computing the index number, allowance has been made for this change
in the manner of quoting.
11. Lard.

Bremen quotations. The monthly average price is ascertained as in the case of
corn. The quotations are for 100 kilograms refined American lard, Wilcox brand, in
bond. The Imperial Statistical Office did not publish lard quotations until 1880.
The prices for September, October, and November, 1888, and for January, 1892, were
merely nominal, owing to lack of the commodity, but were included in the yearly
average. Although lard is quoted in bond, the rate of duty, 10 marks throughout the
period, is given.
Group III.—Colonial goods, etc.

12. Leaf tobacco.

Bremen quotations. The monthly average price is ascertained as in the case of
corn. The quotations are for 100 kilograms Kentucky ordinary, container included,
in bond. Although the article is quoted in bond, the rates of duty are given.
In 1879 the duty on 100 kilograms net was 24 marks.
At present the duty on tobacco leaves, stems, and ribs, as well as tobacco sauce, is
85 marks.
At present the duty on stripped leaves and smoking tobacco is 180 marks.
IS. Coffee.

Bremen quotations. The monthly average price is ascertained as in the case of
com. The quotations are for 100 kilograms net, including sack, in bond. Prior to
1896 the quotations were for “ Rio, good ordinary,” and since January, 1896, for
“ Sabanilla, fair ordinary.” The change was made because Rio, good ordinary, had
become relatively of small importance in the Bremen market. Although quoted in
bond, the duty rates are given.
Prior to July, 1879, duty per 100 kilograms net, 35 marks.
Since July, 1879, duty per 100 kilograms net, 40 marks.
14* Rice.

Bremen quotations. The monthly average price is ascertained as in the case of
com. The quotations are for 100 kilograms Rangoon table rice, shelled, in bond.
Although quoted in bond, duty rates are given.
Prior to July, 1879, the duty per 100 kilograms gross, 3 marks.
Since July, 1879, the duty per 100 kilograms gross, 4 marks.



246

BULLETIN OF THE BUREAU OF LABOR STATISTICS.
15. Pepper.

Bremen quotations. The monthly average price is ascertained as in the case of
com. The quotations are for 100 kilograms, black Singapore pepper, in bond.
Although quoted in bond, duty rates are given.
Prior to July, 1879, duty per 100 kilograms net, 39 marks.
Since July, 1879, duty per 100 kilograms net, 50 marks.
Group IV.—Raw materials of the textile industry.

16. Cotton.

Bremen quotations. The monthly average price is ascertained as in the case of
com. The quotations are for 100 kilograms middling upland, in bond, duty free.
17. Wool.

Berlin quotations. The monthly average price is ascertained on the basis of weekly
quotations by the committee of senior merchants of Berlin after having heard the
brokers. The quotations are for 100 kilograms North German sheep wool, medium
grade, duty free.
18. Hemp.

Hamburg quotations. Quotations every Friday. The monthly average price is
ascertained as in the case of alcohol. The quotations are for 100 kilograms Mexican,
net, in bales of about 350 pounds, 7 pounds tare, in bond, duty free.
19. Raw silk.

Krefeld quotations. Quotations are ascertained on the second Monday of every
subsequent month by a committee of merchants and manufacturers on the basis of
local transactions and of the medium price. The quotations are for 1 kilogram Milan­
ese organzine, classique, 18-20, duty free.
20. Linen yam.

Bielefeld quotations. Quotations represent prices determined on the first and
middle of every month by the secretary of the chamber of commerce on the basis of
the prices obtained at the sales occurring at the spinning mills. The quotations are
for 1 kilogram linen yarn, No. 30 (English number), medium price, between grades
la and Ila, but for the sake of accuracy the author has given the yearly average price
per 100 kilograms.
1879, duty per 100 kilograms gross, 3 marks.
Group V.—Metals.

21. Foundry pig iron.

Breslau quotations. Prices are obtained through private persons at the end of the
month from the average of all the quotations for delivery on the last day of the month.
The quotations are for 1,000 kilograms Silesian foundry pig iron at the foundry. The
grade of Silesian foundry pig iron quoted at Breslau is lowei than that quoted at Diisseldorf. Since 1897 other lower prices have been quoted for pig iron sold to points in
Lower Silesia, but these have not been included in the index number.
Until June, 1879, duty free.
Since June, 1879, duty per 1,000 kilograms gross, 10 marks.




INDEX NUMBERS---- GERMANY.

247

22. Bessemer pig iron (from the Ruhr districts, Rhenish Westphalia).

Dortmund quotations. Prices are determined at the end of every month by a com­
mittee of the chamber of commerce or the secretary of the chamber of commerce on
the basis of schedules filled out by producers, consumers, and dealers and on the basis
of the lowest and highest prices. The quotations are for 1,000 kilograms Bessemer
pig iron from the districts of the Ruhr at the foundry.
Until June, 1879, duty free.
Since June, 1879, duty per 1,000 kilograms gross, 10 marks.
28. Lead.

Berlin quotations. The monthly average price is ascertained as in the case of wool.
The quotations until May, 1899, are for 100 kilograms Tamowitz lead, Saxonia brand,
and since May, 1899, are for 100 kilograms Tarnowitz and Harz lead. For November
and December, 1887, on account of the great fluctuation in prices, averages could not
be given. Duty free.
24. Copper.

Berlin quotations. The monthly average price is ascertained as in the case of wool.
The quotations are for 100 kilograms Mansfeld copper. For the year 1887 the Imperial
Statistical Office, owing to the great fluctuation in prices, gives a nominal average
covering only the months January to May. The author, convinced that this nominal
figure was too low, computed an average for the greater part of the second half of the
year on the basis of information which he himself obtained. Taking the average of
the Imperial Statistical Office for the first five months and his own average as repre­
senting the last seven months, he obtained the average which he has used for the year.
He states that this is the only case in which he has deemed it expedient to depart from
the official figures, and in this case the low nominal figure of the statistical office would
have affected the average for the base period sufficiently to have raised the total index
number for the period 1889-1900 about 10 points (i. e., one-tenth of a unit. Index
numbers are printed with two decimals). In March, 1899, when prices fluctuated
greatly the monthly average price given in the tables is merely approximately correct.
Duty free.
25. Zinc.

Cologne quotations. Quotations are obtained every Wednesday by Cologne whole­
sale establishments or the chamber of commerce. The information is based on the
booking of sales and represents the prices paid by buyers—for the indicated grade of
the commodity—settling their accounts regularly. The quotations are for 100 kilo­
grams Rhenish crude zinc, brand “ W H und S S.” The price for March, 1889, is
nominal, since there were no sales. Duty free.
26. Tin.

Hamburg quotations. Quotations are obtained every Friday. The average price
is ascertained as in the case of alcohol. The quotations are for 100 kilograms Banca
tin in blocks. Duty free.

Group VI.—Coal and petroleum.
27. Ruhr anthracite coal.

Essen quotations. Quotations are obtained once a month at the industrial exchange
in Essen by an exchange committee of the chamber of commerce on the basis of the
sales at all the mines in the district and on the basis of the highest and lowest prices.
The quotations are for 1,000 kilograms anthracite coal at the mine. From May to Au­
gust, 1889, there are no quotations on account of the strike. Duty free.




248

BULLETIN OF THE BUBEAU OF LABOB STATISTICS.
28. Upper Silesia gas lump coal ( Gas-Stiichkohle).

Breslau quotations. The price is obtained through private persons at the close of
the month from the average of all quotations for delivery on the last day of the month.
The quotations are for 1,000 kilograms. Upper Silesian lump coal for gas, f. o. b. mine.
Duty free.
29. Petroleum.
Bremen quotations. The monthly average price is ascertained as in the case of corn.
The quotations are for 100 kilograms American white refined, including container, in
bond. From January 1,1892, to July 1,1893, hogshead duty amounting to about 0.95
marks per 100 kilograms net was charged. After the lifting of the hogshead duty, July
1, 1893, the quotation was again for the commodity in bond. Although quoted in
bond, the duty rates are given.
From January, 1879, duty free.
From 1891, duty, 6 marks.
SUBSTITUTIONS AND ADDITIONS.

The report includes parallel data from Hamburg and from the
Imperial Statistical Office for the 7-year period 1879 to 1885. Ham­
burg index numbers were obtained by finding the average actual
Hamburg price for the 7-year period and the average relative Imperial
Statistical Office price for the same period. It was then calculated
what Hamburg price corresponded to the index number 100 and on
the basis of the resulting figure index numbers for the Hamburg
quotations from 1851 to 1885 were computed.
In the case of a few commodities, breaks occur in the series of
quotations. In the absence of Berlin prices Breslau prices have been
substituted from January, 1897, to April, 1899, in the series for wheat,
rye, and oats. These are said to have been “ suitably adjusted”
( u n te r en tsprecT iend er A n p a s s u n g ) but the process is not described.
Neither is the method of changing from Hamburg to statistical office
prices of butcher’s meat at the end of 1880 described. No actual
prices are given for lard for the year 1879, and the index numbers
for all months of that year are given as 100. This procedure is
equivalent to the substitution of the average of the actual prices
from 1880 to 1888 as the actual price for 1879. The base period for
this commodity is therefore the 9-year period 1880 to 1888.
WEIGHTING.

The total index is the simple arithmetic mean of the index num­
bers of the 29 articles. No system of weighting is used. The author
holds that the simple average of wholesale prices of important raw
materials roughly indicates the course of prices and that this is its
only purpose. He does not think that manipulation on the basis
of estimated consumption makes it a satisfactory index of the stand­
ard of life or increases its value as an index of price movements.
He may be considered, however, to have weighted his own index, in
a loose sense, for two of the 29 commodities are different varieties of
coal and two others are separate varieties of iron.



249

INDEX NUMBERS---- GERMANY.
TESTING.

No test of the index number is made. A comparison is shown,
however, with an index number derived from Soetbeer’s figures for
the total of the 24 articles in question, as follows:
Period.
1851-1855....................
1856-1860....................
1861-1865....................
1866-1870....................
1871-1875....................

Soetbeer.

Schmitz.

117.18
123.35
121.46
120.98
135.56

117.32
122.06
120.58
119.62
130.75

TABLES OP RESULTS.

The following table shows the index numbers for the total of all
articles. The numbers from 1851 to 1878 represent Hamburg quota­
tions for 24 commodities, while the numbers from 1879 to 1902 rep­
resent Imperial Statistical Office quotations for 5 additional com­
modities, or 29 in all. The statement is made that the addition of
the 5 articles affects the index number only slightly.
IN D E X NUMBERS OF WHOLESALE PRICES IN GERM ANY, 1851 TO 1912, ACCORDING TO
OTTO SCHMITZ.
Total
index
num­
ber:
Year. 18791888
equals
100.
1851
1852
1853
1854
1855
1856
1857
1858
1859

100.60
103.33
118.91
131.79
131.96
132.59
132.76
112.08
114.16

Total
Total
Total
Total
Total
index
index
index
index
index
num­
num­
num­
num­
num­
ber:
ber:
ber:
ber:
ber:
Year. 1879Year. 1879Year. 1879Year. 1879Year. 18791888
1888
1888
1888
1888
equals
equals
equals
equals
equals
100.
100.
100.
100.
100.
1860
1861
1862
1863
1864
1865
1866
1867
1868

118.73
117.24
120.31
122.68
125.28
117.37
119.88
120.69
120.09

1869
1870
1871
1872
1873
1874
1875
1876
1877

120.14
117.32
123.02
136.12
141.56
130.60
122.41
119.52
119.86

1878
1879
1880
1881
1882
1883
1884
1885
1886

110.62
100.87
111.71
109.26
106.52
104.06
99.62
92*88
88.00

1887
1888
1889
1890
1891
1892
1893
1894
1895

90.98
96.07
100.87
107.54
104.75
95.46
92.21
83.79
83.55

1896
1897
1898
1899
1900
1901
1902
1903
1904

83.91
85.79
90.65
98.43
106.49
100.11
99.19
100.64
100.22

Total
index
num­
ber:
Year. 18791888
equals
100.
1905
1906
1907
1908
1909
1910
1911
1912

103.50
112.08
119.43
112.87
111.65
113.65
118.95
130.41

N otes.—1. Index from 1851 to 1878 inclusive based on actual wholesale prices of the Hamburg Bureau
of Trade Statistics; from 1879 to 1912 on actual wholesale prices of the Imperial Statistical Office of
Germany.
2. The index numbers for the years 1909,1910,1911, and 1912 do not include two articles, copper and pig
iron (one kind).
3. Index for 1851 to 1902 inclusive from “ Die Bewegung der Warenpreise in Deutschland von 1851 bis
1902” ; index for 1903 to 1912 inclusive from the British “ Report of an Enquiry by the Board of Trade
into Working-class Rents and Retail Prices, 1912.”

The two series of index numbers for the 7-year period for which
parallel data were available are:




Year.

1879.............................
1880............................
1881.............................
1882.............................
1883.............................
1884.............................
1885.............................

Hamburg.

104.47
109.94
110.19
106.21
104.52
98.31
90.69

Imperial
Statistical
Office.
100.87
111.71
109.26
106.52
104.06
99.62
92.88

250

BULLETIN OP THE BUREAU OP LABOR STATISTICS.

A table of index numbers for each of the 29 articles, and a total
index number, has been prepared for a series of 5, 10, and 25 year
periods, as well as for the 22-year period 1879 to 1900. These are
calculated on the basis of 1879-1888 equals 100. It is obvious that
any one of these period indexes could be made the basis (100) of a
new series of calculations for the different commodities if so desired.
IN DEX NUMBERS, B Y PERIODS.
1851-1855
1856-1860
1861-1865
1866-1870
1871-1875

117.32
122.06
120.58
119.62
130.75

1876-1880
1881-1885
1886-1890
1891-1895
1896-1900

112.52
102.47
96.69
91.95
93.05

1851-1860
1861-1870
1871-1880
1881-1890
1891-1900

119.69
120.10
121.63
99.58
92.50

1851-1875

122.07

1876-1900

99.34

1851-1900

110.70

1879-1900=96.97

In summary the author shows the price level of the latest decade
included in the study, by means of cumulative figures, as follows:
The index number for the period 1851 to 1900 is 110.70; for 1879 to
1900, 96.97; for 1891 to 1900, 92.50.
INDEX NUMBERS OP ADOLF SOETBEER.
PUBLICATION AND HISTORY.

Dr. Adolf Soetbeer, a German economist, published the results
of a study of wholesale prices in Hamburg as early as 1858,1 but his
main contribution to the study of prices and the one that contains the
index number which he continued until his death, in 1892, appeared
in 1885, under the title “ Materials toward the Elucidation of the
Economic Conditions affecting the Precious Metals and the Question
of Monetary Standards” ( M a te r ia lie n z u r E r ld u t e r u n g u n d B e u r t d lu n g
d e r w ir ts c h a jtlic h e n E d e lm eta llv e rh a ltn is s e u n d d e r W a h ru n g s fra g e .
B e r l i n , 1 8 8 5 ; 2 . A u f g a b e , B e r l i n 1 8 8 6 ).
This publication has been
translated into English, in full, by Prof. F. W. Taussig and the trans­
lation is included in the United States Senate Document No. 34
(pp. 57-258), 1st session, 50th Congress, 1887. Dr. Soetbeer added
the index numbers for 1886 in the Hamburg Borsen-Halle, Nos. 181 and
182 (a translation of which also appears in Senate Document No. 34,
pp. 271-276), and published the indexes for succeeding years up to
1890, inclusive, in his article on “ The course of prices from 1886 to
1890” (.D a s N iv e a u d e r W a r e n p r e is e i n d e n J a h r e n 1 8 8 6 - 1 8 9 0 ) , which
appeared in the Jahrbucher ftir Nationalokonomie und Statistik,
1892, 3. Folge, 3, pp. 588-596. The original tables compiled by
Soetbeer show average prices and indexes for each individual article,
i Beitrage zur Statistik der Preise: I. Uebersicht der Durchscbuitts-Preise verschiedener Handelsartikel
nach den Angaben im Hamburger Bdrsen-Preiscourante in den Jahren 1851-1857 unter Vergleichung mit
den Durchschnittspreisen der Jahrzehnte 1831-1840 und 1841-1850; II. Zusammenstellung der jahrlichen
Durchschnitts-Preise ftir Weizen in Hamburg, Hannover, Braunschweig, Berlin, Frankreichund England
wahrend der Jahre 1851-1857. Hamburg, 1858.




INDEX NUMBERS---- GERMANY.

251

for groups, and for five-year periods from 1851 to 1885. The table
for 1886 omits some articles of minor importance but continues the
data for the rest and for the groups—-with the exception of the group
of British exports. The article in the Jahrbucher continued the data
to include 1890, by groups only, but stated that the figures for the
individual articles were in the writer’s hands in manuscript but
would have to be reserved for future publication on account of lack
of room in the Jahrbucher. Soetbeer’s death in 1892 prevented the
realization of his plan. Although his index numbers were not con­
tinued beyond 1891, two other important indexes have been based
on Hamburg prices, namely, those of Dr. Heinz, published in Hamburgs Handel und Schiffahrt by the bureau of trade statistics of
Hamburg and those of Prof. Conrad, of Halle, published in the Jahr­
bucher fur Nationalokonomie und Statistik.1
SOURCE OP QUOTATIONS—BASE PERIOD.

Soetbeer used the average wholesale prices published by the
bureau of trade statistics of Hamburg. These prices do not go back
farther than 1847, since prior to that time no such statistical bureau
existed. For that reason Dr. Soetbeer found it necessary to select
1847-1850 for the base period, although he himself states that he
would have much preferred to use 1841-1850.
PRICES: H O W SHO W N AND COMPUTED.

The annual average prices were computed by a simple arithmetical
process from the total quantity and total value of each article im­
ported as recorded by the Hamburg bureau. Since the bureau of
trade statistics entered the weight and kind of each article imported
into Hamburg, and the price of each on the Hamburg exchange
on the day of importation, this was most easy. When the price of
an article was not quoted on the exchange, then the invoice value,
plus freight, insurance, and other charges, was entered. As long
as Hamburg was a free harbor and all goods entering the city in any
manner whatever were recorded, these average prices were of the
utmost importance. But when, in 1888, Hamburg joined the German
customs union, and only articles entering by sea were required to be
recorded, the figures for land importations could no longer be ob­
tained with any accuracy, nor could those of articles entering by
both land and sea. In consequence the quotations of the bureau
of trade statistics, which up to that time had numbered over 300,
now dropped to 163. Dr. Heinz, director of the Hamburg Bureau
of Trade Statistics, undertook the task of going back over the records
of the bureau to separate sea importations from land importations
for Soetbeer’s list of articles, so as to furnish reliable average prices
on them for years up to 1891, inclusive. But after the death of
1 For a description of these index numbers see pp. 219-232 of the present bulletin.




252

BULLETIN OP TH E BUREAU OF LABOR STATISTICS.

Soetbeer, in 1892, Dr. Heinz carried on this investigation for a
different list of articles, selecting only such as presented data pre­
vious to 1888 that could also be quoted subsequent to 1888. This
list contains only 70 of Soetbeer’s 114 articles, but adds 110 new quo­
tations. The average prices of these 180 commodities were carried
back to 1850.
DESCRIPTION AND GROUPING OF COMMODITIES.

The 114 articles on which Soetbeer computed his index numbers
were grouped as follows:
I. Products o f agriculture, etc. (20 articles).

Wheat.
Wheat flour.
Rye.
Rye flour.
Oats.
Barley.
Malt.
Buckwheat.
Peas.

White beans.

Potatoes.*
Hops.
Clover seed.
Rapeseed.
Rapeseed oil.
Linseed oil.
Oil cake.
Raw sugar.
Refined sugar.
Spirits from corn or potatoes.
II. Animal and fish 'products (22 articles).

Beef*
Veal.*
Mutton.*
Pork*
Milk *
Butter.*
Cheese.
Tallow.
Lard.
Hides.
Calfskins.

Leather.
Horsehair.
Bristles.

Bed feathers.
Bones.
Buffalo horns.
Glue.
Herrings.
Dried fish.
Fish oil.
III. Southern products (7 articles).

Raisins.
Currants.
Almonds.
Dried prunes.

Olive oil.
Wine in casks.
Champagne.
IV . Colonial products (19 articles).

Coffee.
Cocoa.
Tea.
Pepper.
Allspice.
Cassia bark.
Rice.
Sago.
Arrack.
Rum.




Tobacco.
Indigo.
Cochineal.
Logwood.
Redwood.
Mahogany.
Cane.
Palm oil.
Ivory.

INDEX NUMBERS---- GERMANY.

253

V. Minerals and metals (14 articles).

Copper.
Quicksilver.
Sulphur, raw.
Saltpeter, raw, Chile.
Salt.
Lime.
Cement.

Coal.
Pig iron.
Bar iron.
Steel.
Lead.
Zinc.
Tin.

VI. Textile materials (7 articles).

Silk.
Cordage.

Cotton.
Wool.
Flax.
Hemp.

V II. Miscellaneous (11 articles).

Potash.
Soda.
Tallow candles.
Tar.
Wax.

Guano.
India rubber.
Gutta-percha.
Rosin.
Pearl ash.
Pitch.

V III. British articles o f export (14 articles).

Cotton yam.
Piece goods, plain.
Cotton piece goods, printed.
Cotton stockings and socks.
Thread for sewing.
Common glass bottles.
Linen yam.

Linen, plain.
Linen sail cloth and sails.
Woolen and worsted yam.
Woolen cloths, etc.
Flannels, etc.
Worsteds.
Carpets, etc.

Note.—The prices of articles marked with an asterisk are the average of the prices paid by Hamburg
institutions (hospitals, etc.) for large purchases.
SUBSTITUTIONS AND ADDITIONS.

Although no substitutions or additions are mentioned specifically
as having been made, the procedure of the Hamburg Bureau of Trade
Statistics in securing its average annual wholesale prices, as described
by Dr. Soetbeer, would readily admit of such being done. Dr.
Soetbeer states 1that since the quantities and kinds of many important
articles undergo changes in the course of decades, it had seemed
proper to the bureau to take no account of the different kinds of each
article but to treat all kinds as one in order to get a general indication
of the changes in prices. He also adds that to meet objections to
his selection of articles for his compilation, he has carefully revised
the list, so as to exclude those for which the Hamburg wholesale
prices are not to be considered fairly indicative of prices in the
general trade; that a number of very important articles were included
for which the official statements give no figures and for which whole­
sale prices have been ascertained from the yearly accounts of large
1 U. S. Senate Ex. Doc. No. 34 (p. 226), 50th Cong., 1st sess., 1887.




254

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

institutions at Hamburg, as in the case of meat, butter, milk, and
eggs; that the prices of yarns and cloths which appeared in the earlier
compilation were later excluded, since the indirect influence of the
German import duties on the importation of cheaper grades prevents
the prices from indicating the general range of prices of such articles,
and that in their place have been given corresponding average prices
of yarns and cloths exported from England, as well as the prices of some
other articles of manufacture, all derived from the British trade
statistics. Likewise, to incorporate the results of a thorough and
detailed revision made by the Hamburg Bureau of Trade Statistics
of its earlier price tables, it was necessary to make some slight changes
in the figures for some articles in the second edition of the Materialien
as compared with the figures for those same articles in the first edition.
Since Soetbeer’s tables were discontinued only six years after their
first publication there was no occasion to resort to interpolation, and
no trace of any is evident.
WEIGHTING.

Soetbeer's index numbers were not weighted, although the problem
was recognized and met half way by his discriminating selection of
articles and by his including more than one variety of a commodity
in the list, as in the case of wheat, rye, sugar, etc.
TESTING.

To test his index Dr. Soetbeer constructed a comparative table
showing the successive annual average prices from 1871 to 1890 of
three different groups of Sauerbeck's articles as compared with three
similar groups of his own, recomputing for this purpose his own index
numbers on Sauerbeck's base period of 1867-1877. The comparisons
are made, first, between the general index number for Sauerbeck's
complete list of 45 articles 1 and his own index number for 100 articles;
second, between the index numbers for their respective groups of
agricultural products, which include 7 articles in Sauerbeck's list and
20 in Soetbeer's; and, third, between the index numbers for their
respective groups of minerals and metals, consisting of 8 articles for
Sauerbeck and 14 for Soetbeer. In the same article he makes a
further test by contrasting his index number for 100 articles for the
separate years from 1881 to 1890 with the index numbers for exports,
for imports, and for both, which are published by the Imperial Statis­
tical Office For this comparison the base period 1881 of the bureau
is used.




1 See pp. 271-273 of this bulletin.

255

INDEX NUMBERS---- GREAT BRITAIN.
TABLE OF RESULTS.

In his article on “ The course of prices from 1886 to 1890” 1 Dr.
Soetbeer presents his last published table, which gives his computations
of index numbers for the 114 articles by groups, as follows:
SUMMARY OF RELATIVE PRICES OF COMMODITIES FOR THE YEARS 1847-1890.
(Base period, 1847-1850=100.)

Year.

1847-1850..
1851-1855..
1856-1860..
1861-1865..
1866-1870..
1871..........
1872..........
1873..........
1874..........
1875..........
1871-1875..
1876..........
1877..........
1878..........
1879..........
1880..........
1876-1880..
1881..........
1882..........
1883..........
1884..........
1885..........
1881-1885..
1886..........
1887..........
1888..........
1889..........
1890..........
1886-1890..

VH.
I.
II.
V.
IV.
VI.
III.
Miscel­
Agricul­ Animal Southern
Textile laneous
Colonial Minerals
tural products, products. products.
and
metals. materials. articles.
products.
etc.
100.00
129.99
131.84
124.46
137.74
144.76
144.17
146.21
150.99
138.16
144.90
141.06
145.34
132.50
132.92
138.11
138.12
137.50
138.45
143.33
123.85
110.75
130.77
101.31
96.28
98.18
102.06
107.53
101.06

100.00
114.79
132.31
128.24
136.35
144.14
155.82
156.72
157.76
158.59
154.57
155.79
152.51
141.53
137.60
147.30
146.76
151.21
155.17
156.40
150.26
140.45
150.65
133.53
129.93
128.97
130.95
129.85
130.41

100.00
110.43
134.72
114.13
121.54
122.99
125.36
132.15
145.02
131.35
131.50
128.69
140.55
134.34
139.10
154.65
138.91
146.57
139.23
142.38
120.16
123.78
134.41
122.44
121.81
120.09
127.57
138.61
126.08

100.00
110.97
122.61
118.64
118.32
120.22
130.25
134.32
136.74
132.11
130.72
129.74
130.29
125.61
123.34
122.92
126.38
122.60
122.47
120.17
117.90
116.39
119.91
115.45
116.59
116.41
118.82
119.35
117.32

100.00
107.03
113.59
102.11
95.47
101.85
121.63
140.60
116.70
107.49
116.90
106.27
98.87
94.14
84.28
88.33
94.35
84.87
86.99
82.93
78.69
74.23
81.55
70.52
72.50
75.57
78.55
83.54
76.12

100.00
105.20
107.12
131.83
129.17
119.23
122.79
119.58
112.80
111.47
117.17
105.54
108.33
102.33
98.76
96.72
102.33
99.29
95.10
95.93
97.02
95.89
96.65
89.76
81.42
82.17
89.05
81.92
84.86

VIII.
British
exports.

100.00
106.65
108.21
144.33
105.90
117.48
128.54
119.14
112.21
98.74
114.98
101.78
99.80
97.24
90.21
95.23
96.79
94.89
99.10
95.38
84.82
81.35
91.11
78.75
77.30
74.31
86.41
91.70
81.70

100.00
98.47
102.41
127.56
130.55
122.64
130.07
128.52
126.06
124.96
126.44
119.23
114.04
111.03
105.93
108.15
111. 70
103.08
104.72
104.72
103.36
100.48
103.28
97.03
95.98
94.91
96.60
94.96
95.90

I-VIII.
Total
articles,
114.
100.00
112.22
120.91
123.59
123.57
127.03
135.62
138.28
136.20
129.85
133.29
128.33
127.70
120.60
117.10
121.89
123.07
121.07
122.14
122.24
114.25
108.72
117.68
103.99
102.02
102.04
106.13
108.12
104.41

GREAT BRITAIN.
INDEX NUMBERS OF THE BOARD OF TRADE.
PUBLICATION.

The first report of this series contains the results of an investiga­
tion conducted by the Labor Department for years prior to 1902.2
Since that year an annual report on the subject has been prepared
and published by the Board of Trade in the January issue of The
Labour Gazette, London.
HISTORY.

The inquiry concerning the subject of prices had occupied the
Labor Department for several years, and, on account of the great
amount of public attention devoted to all questions affecting prices
of commodities, it was decided in 1903 to publish the results up to
1Jahrbucher fur Nationalokonomie und Statistik, 1892, p. 593.
2 Report on Wholesale and Retail Prices, 1902. Great Britain. Board of Trade.




256

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

1902 without delay. The first report in 1903 consisted of a series of
comparative tables of actual wholesale prices covering the years from
1871 to 1902.
The index of wholesale prices was computed upon 45 selected arti­
cles. In respect to most of the articles the actual prices were carried
as many years back of 1871 as was regarded safe, considering the
nature of the data available. However, the index was not computed
back of 1871, even where the actual prices of individual articles were
secured for earlier years, as in the instance of bread in the city of
London, where the price was carried back to 1758.
To show the average change of general prices, not only from 1871
to 1902 but extending over the whole of the nineteenth century, a
chart was published covering the period from 1801 to 1902.
The statement showing the course of prices from 1801 to 1846 is
based on Jevons* index number, from 1846 to 1871 on that of Sauer­
beck, and from 1871 to 1902 on the Board of Trade index as shown
in these reports.
SOURCE OF QUOTATIONS.

The data used in the original report were import and export aver­
age values, contract prices at hospitals and institutions, prices at
markets, ascertained values of coal and iron in different districts used
for the determination of wage rates, prices from private firms, asso­
ciations, etc. The import values were based on the declarations of
the importer, those for exports also being declared values, but the
report states that various difficulties were encountered “ in tracing
back the average value of the same article throughout so long a period,
arising to a large extent from changes in classifications.” 1 The same
system of declared values was in force throughout the period. The con­
tract prices of certain articles for hospitals and asylums of the London
County Council represent the prices paid, throughout the period, by
a somewhat similar class of consumers. Market prices were compiled
from official reports, newspapers, and market quotations. The ascer­
tained values of coal and iron were from reports made by accountants
for use in the determination of the general rate of wages by sliding
scale or otherwise. In a few cases it was found necessary to secure
quotations from the original sources and from private corporations.
A memorandum states that it was proposed to use either import or
export values according as the article was chiefly one of import or
export, except for British corn, milk, potatoes, beef, mutton, and
brick.2
BASE PERIOD.

The year 1871 was originally adopted as the base period, and from
1871 to 1906 the index was computed upon this base for the average
1Report on Wholesale and Retail Prices, 1902, p. 427. Great Britain. Board of Trade.
2 Idem, p. 439.




INDEX NUMBERS---- GREAT BRITAIN.

257

price of all the articles. This was used as a standard until 1906, when
1900 was established as the base and the index for the 45 articles as
a whole was recomputed on the new basis from 1871 to 1906. Since
then 1900 has been used as the base period.
PRICES: H O W SHOW N AND COMPUTED.

In the original report the prices were shown as yearly actual aver­
ages for the separate articles. In succeeding reports up to 1905 the
actual average prices were not shown, but the index numbers for the
four groups and the general index number were published. Since
1905 no price data other than the general index number have been
shown.
NUMBER AND CLASS OF COMMODITIES.

The list of articles covered by the reports numbers 45 and includes
principally raw materials or materials at an early stage of manufac­
ture. The January, 1914, issue of the Labour Gazette states that, in
compiling the general index number, the index numbers for 47 sepa­
rate articles were weighted in accordance with their estimated con­
sumption. Counting milk, butter, and cheese as separate articles
(heretofore counted one) probably accounts for the new number;
however, no explanation is vouchsafed.
DESCRIPTION AND GROUPING OF COMMODITIES.

Below is shown the list of the 45 articles included in the group and
general index, the weight allotted to each article, and the source of
the quotations.1
Group I .— Coal and metals (6 articles).

Source of price quotations*

Article.

Export values of coal.
Export values of pig iron.
Import values of copper regulus.
Import values of crude zinc.
Import values of block tin.
Import values of pig and sheet lead.

Coal..................................
Pig iron............................
Copper (ore and regulus).
Crude zinc........................
Block tin..........................
Lead.................................
Total.

Group I I .— Textiles (raw materials— 6 articles).
Cotton, raw...................................................................
Wool, British...............................................................

38
6

Wool, foreign................................................................

13

Jute, raw.......................................................................
Flax, raw......................................................................
Silk, raw.......................................................................

3
4
9

Total...................................................................

73

Import values of raw cotton.
Export values of sheep and lambs’
wool.
Import values of sheep and lambs’
wool.
Import values of jute.
Import values of flax.
Import values of silk.

i Report on Wholesale and Retail Prices, 1902, pp. xxxv-xxxvii,

94231°—Bull. 173—15----- IT




Great Britain, Board pf Trade,

258

BULLETIN OF THE BUREAU OF LABOR STATISTICS.
Group I I I .— Food and drink (23 articles).
A . Corn, etc.

Article.

Wheat, British
Barley, British
Oats, British...
Wheat, foreign.
Barley, foreign.
Oats, foreign...
Maize...............
Hops................
R ice.................
Potatoes...........

Source of price quotations.

Gazette average of British wheat.
Gazette average of British barley.
Gazette average of British oats.
Import values of wheat.
Import values of barley.
Import values of oats.
Import values of maize.
Import values of hops.
Import values of rice.
Contract price, potatoes at St. Thomas’
Hospital.

Total----B. Meat, fish, and dairy products.
Beef...............................

Beef (Uve),^lst class, Metropolitan Cat-

Mutton..........................

Mutton (live), 1st class, Metropolitan
Cattle Market.
Import values of bacon.
Average price of milk at Bethlam Royal
Hospital and St. Thomas’s Hospital.
Import values of eggs.
Export values of herrings.

Bacon............................
Milk, butter, cheese, etc

Eggs-.......................

Herrings........................
Total...................
C. Tea, tobacco, wine. and sugar.
Sugar.........
Tea............
Coffee.........
Cocoa.........
Rum ..........
W ine.........
Tobacco....

Import values of refined sugar.
Import values of tea.
Import values of coffee.
Import values of cocoa.
Import values of rum.
Import values of wine.
Import values of unmanufactured to­
bacco.

Total

Group I V .— Miscellaneous (10 articles).
Cotton seed.........
Linseed...............
Olive oil......... .
Palm oil..............
Paraffin...............
Petroleum...........
Bricks.................
Hewn fir..............
Caoutchouc.........
Hides...................

Import values of cotton seed.
Import values of linseed.
Import values of olive oil.
Import values of palm oil.
Import values of paraffin.
Import values of petroleum.
Price of stocks at Glasgow.
Import values of hewn fir.
Import values of caoutchouc.
Import values of hides.

Total.........
Grand total

SUBSTITUTIONS AND ADDITIONS.

Various difficulties, as was previously stated, were met in tracing
average values throughout so long a period. Changes in classification
were the cause to a large extent. The methods adopted in making
substitutions are not fully explained. Apparently no new articles
have been added since the publication of the first report.



INDEX NUMBERS---- GREAT BRITAIN.

259

INTERPOLATION.

In the discussion of sources of information the statement is made
that where the data related only to some of the earlier years of the
period covered, or could not be continued to the present, they were
omitted, and that when large gaps existed in the records it was the
general practice as far as possible to start the table from dates sub­
sequent to the gaps in order to preserve continuity.1
WEIGHTING.

The method of weighting used in computing this index number was
that based on the amount of consumption of the various articles in
the United Kingdom. The consumption of an article is defined to
mean any process by which the commodity is substantially changed
in character. The original report in its explanation of the “ consump­
tion standard” states that “ the theoretical basis of the consumption
standard is the proposition that the true measure in the change of the
value of money is the change in the amount of gold that must be paid
by consumers throughout the country for all commodities in their
finished state consumed by them per unit of time.” 2 The value of
the national consumption of the 23 raw materials which were derived
almost entirely from foreign sources was taken to be the declared
value of the imports less the declared value of the exports. The
value of the consumption of the 22 remaining articles was the value
of the quantity produced plus the value of the amount imported, if
any, minus the value of the exports, if any. The results thus ob­
tained represent the estimated value in millions sterling of the annual
consumption of the articles. The millions sterling constitute the
weights allotted.
The weights assigned to the various articles were placed against the
percentage variations in prices. The percentage variations were com­
puted for each year by using 1871 as the base, or 100. For example,
the percentage price of coal in 1872 was 161.1 per cent, the price in
1871 being 100. This percentage multiplied by 34—the weight
allotted to coal—produced 5,477.4, or what was termed the weighted
percentage. The sum of the weighted percentages of all the articles
in a group divided by the sum of the weighted percentages for the
base period produces the index number for the group in the specified
year. For example, the weights for the group of coal and metals were
5,950.0 for the base year, and for 1872 the total was 9,173.2, which
divided by the figure for the base year equals 154.1, the published
index number of the group for 1872. A continuation of this process
produces the other group indexes and the general index for the 45
commodities is computed in like manner.
* Report on Wholesale and Retail Prices, 1902, p. 426. Great Britain. Board of Trade.
p. 432.

2 Idem,




260

BULLETIN OF THE BUREAU OF LABOR STATISTICS.
TESTING.

Some comparison of results was made with the results of other
indexes. The principal test was made by using certain articles as given
by Sauerbeck. These articles covered only 28 of the 45 price quota­
tions in his report, but they formed nine-tenths of the total weight
of the Board of Trade index number. These articles were used to
form a special index number, making use of the weights allotted as
above. The results are shown below.
COMPARISON OF A SPECIALLY WEIGHTED NUMBER FROM THE BOARD OF TRAD E
AND SAUERBECK’S NUMBER.i
Specially
weighted Sauerbeck’s
number.
number.

Year.

1867.........................
1868.............................
1869.............................
1870.............................
1871.............................
1872.............................
1873.............................
1874.............................
1875.............................
1876.............................
1896.............................

100
100
96
94
96
107
113
104
97
96
62

100
99
98
96
100
109
111
102
96
95
61

1 Report on Wholesale and Retail Prices, 1902, p. 449. Great Britain.

Board of Trade.

TABLES OF RESULTS.

The principal table of the original report shows the index numbers
for the four general groups, three subgroups under food and drink,
and the index for all the 45 commodities.1 This table is reproduced
below and carried forward to 1905, the last year that the index was
computed upon the basis of 1871 as 100.
BOARD OF TRAD E IN D E X NUMBERS, B Y GROUPS, 1871 TO 1905.
[The index numbers are the sum of the weighted percentages divided by the sum of the weights.]

Index
number
for all
Year.
the 45
commodi­
ties.

1871..

1872..
1873..
1874..
1875..
1876..
1877..
1878..
1879..
1880..
1881..
1882..
1883..

III. Food and drink.
I.
Coal and
metals.

II.
Textiles
(raw
materials).

I l l A.
Com, etc.

III B.
in c .
Meat, fish, Sugar, tea,
and dairy wine, and
tobacco.
produce.

Total,
group III.

IV.
Miscella­
neous.

100.0

100.0

100.0

100.0

100.0

100.0

154.1
194.8
158.8
126.3
107.2
99.9
92.9
86.8
94.9
91.1
91.8
90.3

114.0
110.2
102.6
100.2
93.6
91.7
88.5
84.1
88.6
87.0
84.1
82.0

104.0
109.7
110.8
99.9
98.7
107.9
98.9
97.4
98.0
94.9
95.8
93.9

102.0
109.3
110.1
116.6
117.2
114.9
112.2
106.7
106.6
106.7
112.0
113.7

100.0

110.6
118.8
113.6
107.8
104.2
105.3
99.3
94.9
97.4
95.7
97.3
96.5

102.0
98.1
94.9
93. g
90.7
96.9
88.4
85.2
86.2
84.4
83.6
80.9

100.0

102.7
108.2
108.6
107.3
106.8
110.3
104.2
100.5
100.8
99.4
102.3
102.1

105.1
109.5
108.4
99.3
97.9
97.3
88.3
81.5
89.2
88.5
89.1
87.5

1Report on Wholesale and Retail Prices, 1902, p. 34. Great Britain. Board of Trade,




261

INDEX NUMBERS— GREAT BRITAIN.
B OARD OF TR AD E IN D E X NUMBERS, B Y GROUPS, 1871 TO 1905-Concluded.

Year

Index
number
for all
the 45
commodi­
ties.
88.3
83.0
78.5
76.7
79.3
80.8
82.8
84.1
80.1
78.7
75.1
72.2
69.8
71.3
73.6
74.5
83.2
79.2
78.8
78.6
78.7
77.7

1884..
1885..
1886..
1887..
1888..
1889..
1890..
1891..
1892..
1893..
1894..
1895..
1896..
1897189818991900190119021903..
19041905..

III. Food and drink.
I.
Coal and
metals.

II.
Textiles
(raw
materials).

I l l A.
Corn, etc.

86.4
82.1
78.8
80.1
83.0
94.1
113.6
106.6
98.8
89.4
91.7
85.8
83.3
84.4
92.7
107.5
151.9
124.7
114.9
111.2
106.1
105.6

79.8
75.7
69.0
70.7
70.0
72.4
72.9
70.1
66.1
66.6
60.8
57.7
64.0
59.7
54.8
57.2
70.0
65.7
65.0
71.3
78.7
73.5

81.1
76.7
71.8
71.2
71.7
70.6
72.0
83.2
73.4
68.3
63.1
62.2
57.6
62.7
73.1
63.7
62.4
64.0
63.7
63.8
66.7
64.8

III B.
in c .
Meat, fish, Sugar, tea,
Total,
and dairy wine, and group III.
produce.
tobacco.
104.7
96.3
92.7
88.2
94.0
92.8
91.7
91.1
91.8
95.2
92.2
88.2
81.9
84.6
81.8
85.8
90.3
89.8
94.4
92.1
89.0
88.4

91.4
85.1
81.1
78.3
81.8
81.1
80.6
84.9
81.3
81.1
76.9
74.2
69.4
72.4
75.1
73.2
74.9
75.3
76.7
75.7
75.5
74.8

70.7
66.7
63.7
61.4
65.0
68.0
63.8
64.8
63.9
65.0
59.5
56.2
57.0
54.9
54.2
52.5
52.4
50.1
46.1
47.0
48.2
52.1

IV .
Miscella­
neous.

82.2
80.2
73.8
69.3
71.0
74.3
72.9
70.1
68.1
66.2
62.7
62.6
63.6
63.9
66.5
67.9
74.3
71.7
69.2
68.4
66.0
68.7

A second table, reproduced below, shows the index as now published,
the price in 1900 being used as the base, or 100. No group index
figures have been regularly published since the revision. The exact
method of computation upon the new base period is not stated, but a
note to the report for 1906, issued in January, 1907, states that “ the
index number has, however, now been recalculated with the year
1900 as its base year instead of 1871.” *■ It is presumed that the cal­
culation is made in the same way as in the original report.
BOARD OF TRAD E IN D E X NUMBERS, 1871 TO 1913.
[Source: The Board of Trade Labour Gazette, January, 1914, p. 5.J

Year.

1871___
1872....
1873....
1874....
1875....
1876....
1877....
1878....
1879....

Index
number.
135.6
145.2
151.9
146.9
140.4
137.1
140.4
131.1
125.0

Year.

1880...
1881...
1882...
1883...
1884...
1885...
1886...
1887...
1888...

Index
number.
129.0
126.6
127.7
125.9
114.1
107.0
101.0
98.8
101.8

Year.

1889...
1890...
1891...
1892...
1893...
1894...
1895...
1896...
1897...

Index
number.
103.4
103.3
106.9
101.1
99.4
93.5
90.7
88.2
90.1

Year.

Index
number.

1898...
1899...

93.2
92.2

1900...

100.0

1901...
1902...
1903...
1904...
1905...
1906...

96.7
96.4
96.9
98.2
97.6
100.8

Year.
1907...
1908...
1909...
1910...
1911...
1912...
1913...

Index
number.
106.0
103.0
104.1
108.8
109.4
114.9
116.5

INDEX NUMBERS OP THE ECONOMIST.
PUBLICATION.

This index represents the course of wholesale prices of commodi­
ties in the United Kingdom.
i The Board of Trade Labour Gazette, vol. 15,1907, p. 4.




262

BULLETIN OF TH E BUREAU OF LABOR STATISTICS.

It is compiled and published each month in the Economist,
London, the general results for each year appearing in the first issue
of January of the following year.
HISTORY.

The object of this compilation originally was to throw some light
on the relation between the gold supply and prices. In 1849 gold had
been discovered in California and in 1850 in Australia, and the pouring
of this gold into Europe seemed to be accompanied by a general
upward movement of prices. It was to ascertain whether there had
been such a movement and, if so, its extent that the Economist index
numbers were developed.
In 1859 William Newmarch, then editor of the Journal of the Royal
Statistical Society, published an article in that journal on the prices
of the previous year, in which the prices of 19 commodities in the London
market were expressed as percentages of the average of the prices of
1845-1850. These commodities were as follows: Coffee, sugar, tea,
tobacco, wheat, butcher’s meat, cotton, silk, flax and hemp (average),
wool, indigo, oils (average of 3 varieties), timber, tallow, leather,
copper, iron, lead, and tin. In 1860 and 1861 similar articles appeared
in the journal, when in addition to these 19 commodities 3 others were
added: Raw cotton, cotton yarn, and cotton cloth—all at Manchester
prices. The prices of these 22 commodities were expressed in the form
of percentages, but no general index number was constructed from
them.
The Economist stated in its issue of February 20, 1864, in which it
published for the first time its commercial history and review of the
past year, that in the table of actual wholesale prices it was following
the arrangement and method which were adopted by Tooke and New­
march in their history of prices, and continued by Newmarch in the
Journal for 1859, 1860, and 1861.
The first table in the Economist report of 1864 presented the actual
prices in pounds sterling of 45 articles for the base period of 1845-1850
and for succeeding years down to 1862. The prices were for a given
date, being either those for January 1 or July 1 for all years previous
to 1863, for which year they were the prices for the 1st of each month.
The 45 articles were coffee, sugar (3 kinds), rum, tea, tobacco,
butter, wheat, beef (2 kinds), mutton (2 kinds), pork, cotton, silk,
flax, hemp, wool (4 kinds), dyes (2 kinds), oils (3 kinds), timber (2
kinds), tallow, leather, saltpeter, ashes, copper, iron (2 kinds), lead,
steel, tin, raw cotton (3 kinds), cotton yam, and cotton cloth (2 kinds).
For a few of the articles the prices were not continuous throughout
the period.
A second table was printed entitled “ Proportionate results,” being
the percentage that the actual price of each article for the given date




INDEX NUMBERS— GREAT BRITAIN.

263

was of the actual average price for 1845-1850. In this table, instead
of 45 series of percentages, the number was reduced to 22. This num­
ber was made up of 37 of the 45 series of quotations under the heads of
coffee, sugar, tea, tobacco, wheat, butcher’s meat, cotton, raw silk,
flax and hemp, sheep’s wool, indigo, oils, timber, tallow, leather,
copper, iron, lead, tin, raw cotton (cotton wool), cotton yarn, and
cotton cloth. Of these 22 series of percentages, the 9 composed of
more than one description or grade of the article were sugar (2 kinds),
butcher’s meat (2 kinds of beef and 2 of mutton), flax and hemp (2
articles), sheep’s wool (4 kinds), oils (3 kinds), iron (2 kinds), raw
cotton (3 kinds), cotton cloth (2 kinds), and timber (2 kinds).
The articles were divided into five groups as follows:
I. Colonial and tropical produce (food).
II. Wheat (England and Wales) and butcher’s meat (New­
gate market).
III. Raw materials of manufacture.
IV. Metals.
V. Manchester markets.
The articles under the fifth head were raw cotton, cotton yam, and
cotton cloth.
During the years 1864 to 1867 the composition of these tables
remained the same, except that in 1865 the percentage for raw cotton
was computed upon one grade instead of three, as formerly.
The commercial history, published by the Economist, for 1868 gave
for the first time the total index number. However, this was simply
the total at each date of the 22 percentage columns, no general index
being computed, and it was not until 1869 that the numbers were
added together and divided by 22, the result thus becoming the
“ Economist” index number, which has been published year by year
since that date.
It was announced in the Economist of February 4,1911, that it was
deemed desirable to change the basis upon which the index number
had been calculated. This statement recited the intention of the
publication to make this review of prices more far-reaching by
embodying in the index quotations of some important articles which
play a large part in modern commerce, and at the same time to retain
its character as a wholesale market index number. On account of the
inclusion of new articles, it became necessary to adopt a base period
sufficiently recent to include standardized quotations of modern
commodities.
It was stated that, owing to the fact that many commodities are
now important in the business life that were not so regarded at the
time of beginning the index, the list of commodities had been re­
vised and the number increased. The result of this recasting was
published in the issue of November 18, 1911, showing how* the new




264

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

index number was made comparable with the old figures by dividing
the total index number for the 44 commodities by two, thus reducing
it to that of 22 articles as used formerly.
The chief change made is in respect of the coal and iron trades,
which were formerly represented by one quotation only, but are now
given a weight of 5 quotations out of 44. Quotations are added, for
the first time, for barley, oats, potatoes, and rice among foodstuffs;
Egyptian cotton and jute among textiles; iron bars, steel rails, and
coal among minerals; and petroleum, oil seeds, rubber, and soda
crystals in the miscellaneous group.
In order to show the relation between the percentage index as com­
puted by the old method and the index number obtained by the new
plan, the latter has been reduced to the same basis. The index num­
ber, however, is based on the prices of 44 articles, while the old per­
centage number was computed on but 22. They have been made
comparable, as stated, by dividing the index number for the 44 articles
by 2.
SOURCE OP QUOTATIONS.

The quotations used in compiling this index are market prices as
published weekly in the Economist, which represent those of the
London or Manchester markets.
BASE PERIOD.

As has been explained, the base period formerly was 1845-1850, but
in November, 1911, it was announced that the base period had been
changed to 1901-1905.
PRICES: H O W SHOW N AND COMPUTED.

As was previously stated, the quotations used are those published
weekly in the Economist. The weekly prices for the selected articles
are published each month in the discussion of the index number.
NUMBER AND CLASS OF COMMODITIES.

The original number of commodities, as has been stated, was 22.
In 1911 the number was increased to 44. Raw, or what might be
termed primary, commodities only are included in these quotations.
The following table, appearing in the Economist of November 18,
1911, shows the number of quotations for each commodity, comparing
the old with the new base period:




265

INDEX NUMBERS— GttEAT BHITAIN.

NUMBER OF COMMODITIES: SERIES OF QUOTATIONS UNDER THE OLD BASIS OF
1845-1850, COMPARED W ITH THAT UNDER THE NEW BASIS OF 1901-1905.

Commodities.

Wheat and flour......................
Barley......................................
Oats..........................................
Potatoes...................................
Rice..........................................
Beef..........................................
Mutton.....................................
Pork.........................................
Sugar........................................
Conee.......................................
Tea...........................................
Tobacco...................................
Butter......................................
Cotton (raw. yarn, cloth).......
W ool.......... ............................
Flax.........................................
Hemp
Jute..........................................
Silk..........................................

Old basis New basis
quotation quotation
number. number.
1

3
1
1
1
1

1

2
1
2
1
1
1
1
4
2

1
1
1
1
4
1
1

2
1
1

1

Commodities.

Old basis New basis
quotation quotation
number. number.
1

1
1
1

Tin...........................................
Lead.......................................
Timber.....................................
Leather....................................
O il...........................................

1
1
1
1
1
1

1
1
1

Petroleum................................
Rubber....................................
Tallow.....................................
Indigo......................................
Soda crystals...........................

i
1

Total..............................

22

Pig iron...................................
Steel rails.................................
Iron bars.................................

1
1
1
1
1
1
1
1
44

DESCRIPTION AND GROUPING OF COMMODITIES.

The detailed table of prices week by week as published includes 33
series of quotations. The articles are arranged in four groups as
follows:1
Minerals (8 articles).

Iron, Cleveland, No. 8, G. M. B.
Iron, common bars.
Iron, steel rails.
Coals, best steam Newcastle.
Copper, standard.
Tin, standard.
Lead, English pig.
Saltpeter, Bengal.
Textiles (7 articles).

Cotton, middling American.
Cotton, yarn, 32’s, twist.
Wool, N. S. Wales, greasy, average.
Silk, Canton.
Flax, Riga ZK.
Hemp, manila.
Jute, native firsts.
Food products (9 articles).

Wheat, Gazette averages (English grain).
Barley, Gazette averages (English grain).
Oats, Gazette averages (English grain).
Flour, town-made, household.
Beef, inferior.
Beef, prime.
Mutton, prime.
Potatoes, good English.
Rice, Rangoon.




i The Economist, July 4,1914, p. 8.

266

BULLETIN OF TH E BUREAU OF LABOR STATISTICS.
Miscellaneous (9 articles).

Sugar, West India sirups.
Sugar, beet, German.
Tea, Congou, middling, common.
Tea, Congou, medium, good.
Coffee, Santos, good average.
Oils, petroleum.
Oils, olive, Levant.
Tallow, town.
Rubber, fine hard Para.
SUBSTITUTIONS AND ADDITIONS.

It is stated that over so long a period of time some variations have
inevitably arisen in the mode of quoting prices, but in all such cases
the nearest approach possible has been made to a uniform quotation.
The articles cited are raw cotton, tea, sugar, flax, and wool. It is
further stated that in some cases, where it has been considered
desirable to introduce a commodity to replace one no longer actively
dealt in, the current price of the substituted article has been taken as
equivalent to the same percentage of the basis price as was repre­
sented by its predecessor.
INTERPOLATION.

The supplying of missing data, if such has been found necessary,
has not been noted.
WEIGHTING.

The index is computed by means of simple arithmetical average.
Indirect weighting is attained by the selection of articles.
This method, as applied to the present index, has been frequently
criticized because of the small number of articles included. An
attempt to correct the fault of giving each article an equal weight
was made by Mr. R. H. Inglis Palgrave, in 1886, in a memorandum
to the Royal Commission on the Depression of Trade and Industry.
The method used by him was to give each relative price an importance
proportional to the consumption of the article, which was ascer­
tained by adding to the production the imports and deducting the
exports. He thus obtained a series of figures representing the im­
portance, in each year, of the consumption of each commodity, and
used these in connection with the Economist figures for the years 1865
to 1885 upon the basis, 1865-1869 equals 100. The data prepared by
Mr. Palgrave in 1886 have not been continued for subsequent years
(see Report of United States Senate Finance Committee on Whole­
sale Prices, Wages, and Transportation, 1893, Pt. I, pp. 228, 229).
TESTING.

The testing of the accuracy of the results secured in this index is
made by comparison with the results in other index compilations.




267

INDEX NUMBERS---- GREAT BRITAIN.

The table below shows a comparison by decades of the weighted
and unweighted indexes of the Economist with those of Sauerbeck (an
unweighted index) and the Board of Trade (a weighted one). The
weights for the Economist index have been calculated on the basis
of consumption in the country as estimated by the Board of Trade.1
The decade 1891 to 1900 is considered as the base or 100.
IN D E X NUMBERS OF THE LEVEL OF PRICES IN THE UNITED KINGDOM, B Y DECADES,
1861 TO 1910.
(Base period, 1891-1900=100.)

Economist.
Decade.

Sauerbeck.
Weighted. Unweighted.

1861-1870....
1871-1880....
1881-1890....
1891-1900....
1901-1910....

146
131
107
100
110

152
131
108
100
108

Board of
Trade.

151
144
113
100
110

138
138
112
100
111

The Economist of August 26, 1911, presents as a test of the accu­
racy of its index a table to show that retail prices have pursued much
the same course as wholesale prices. The following comparison is
made of the Economist index with that of the Board of Trade for
retail prices in London from 1895 to 1910, in each case the year 1900
being taken as the base.
COMPARISON OF W HOLESALE AND R E TA IL PRICES, 1895 TO 1910.
(Base period, 1900=100.)

Year.

1895............................
1896............................
1897............................
1898............................
1899............................
1900............................
1901............................
1902......................: . . .
1903............................
1904............................
1905............................
1906............................
1907............................
1908............................
1909............................
1910............................

Board of
Economist Trade
in­
index
dex
number on ber ofnum­
retail
Jan. 1 of
prices in
each year.
London.
89
93
91
88
89
100
99
91
93
103
99
109
117
108
103
112

93.2
92.0
96.2
100.8
96.4
100.0
101.9
101.6
103.2
104.3
103.7
103.2
105.8
108.4
108.2
109.9

TABLES OF RESULTS.

The following table, covering the period from 1851 to 1910, shows
the total index number for the 22 commodities in the form in which
it was published in earler years.




* The Economist, Aug. 26,1911, pp. 422,423.

268

BULLETIN OF THE BUREAU OF LABOR STATISTICS.
THE ECONOMIST IN D E X (ORIGINAL).*
(Base period, 1845-1850=2200.)

1845-50: Aver1851: January..
1853: July........
1857: July........
1858: January..
1859: January..
1860: January..
1861: January..
1862: January..
1863: January..
1864: January..
1865: January..
1866: January..
1867: January..
1868:
January..,
July..........
January.
July.......
1870:
January.
July.......
1871:
January.
July.......
1872:
January.
July.......
1873:
January.
July.......
1874:
January.
July.......

2947
2914

1875:
January.
July----1876:
January.
July----1877:
January.
July----1878:
January.
July.......
1879:
January.
July.......
1880:
January.
July----1881:
January.
July----1882:
January.
July----1883:
January.
July----1884:
January.
J u ly ....
1885:
January.
J u ly ....

2779

January.
J u ly ....

2200

2310
2463
3059
2667
2556
2713
2751
2878
3492
3787
3575
3564
3024
2582

(2)
2711
2590
2640
2835
3054

2778

1887:
January.
July.......

1899:
January__

2059
2116

1900:
January__

2711
2531

January.
July.......

2715
2625

January.
July.......

2187
2161

2554
2457

January.
July.......
1891:
January.
July.......
1892:
January.
July.......
1893:
January.
July.......
1894:
January.
July.......
1895:
January.
July.......

2236
2259

2225
2577
2479
2376
2302
2435
2442
2343
2220

2221
2098
2048
2023
2023

1 The Economist, Sept. 2,1911, pp. 490 and 491.

2239
2121

2224
2190
2133
2081

1903:
January__
1904:
January__

2120
2105

1905:
January__

2082
1974

1906:
January__

1923
1931

1907:
January__

January.
July.......
1897:
January.
July.......

1999
1947

January.
July.......

1890
1915

2 Figures

1901:
January__
July..........
1902:
January__

1950
1885

1908:
January.. .
July..........
1909:
January__
1910:
January__
July..........

1918
2028
2145
2211
2126
2007
1948
1995
2003
2111
2197
2130
2136
2163
2342
2362
2499
2594
2310
2190
2197
2240
2390
2362

not calculated for July 1,1869..

A comparison of the Economist index number as computed on the
old and new bases is afforded in the following table:
COMPARISON OF ECONOMIST IN DEX NUMBERS COMPUTED ON OLD AND NEW BASES.*

Date.

January 1.
July 1.......
1897:
Januaiy 1.
July I .......
1898;
January 1.
July 1.......
January 1.
July 1.......
1900:
January 1.
July 1.......
1901:
January 1.
July 1.......
1902:
January 1.
July 1.......
1903:
January 1.
July 1.......




Old basis
(1845-1850=
100).

New basis
(1901-1905=
100).

91

90
89
90

86
87

95

87
92

93
98£

97£
100£

1 10

97
91

106
103

91

101

91

99£
104*

111

Date.

Old basis
(1845-1850=
100).

New basis
(1901-1905=*
100).

100
97

102
99

97
98

104
102|

106
107

109
110

114
118

115
121

105
100

111*
106|

100
102

104
110

109
107

113*
113

114

114

1904:
January 1.............
July 1...................
1905:
January 1.............
1906:July 1...................
January 1.............
July 1...................
1907:
January 1.............
July 1...................
1908:
January 1.............
July 1...................
1909:
January 1.............
July 1...................
1910:
January 1.............
July 1...................
1911:
January 1.............

i These data are taken from the Economist of Nov. 28,1911, p. 1035.

269

INDEX NUMBERS---- GREAT BRITAIN.

The following table, reproduced from the Economist of Decem­
ber 5, 1914, shows the manner in which the information relative
to the index number is now presented. The quarterly index figures
represent the average of the figures for the months of the quarter.
It will be noted that the method of grouping here employed is not
the same as is used in the detailed table of prices.
IN D E X FIGURES B Y GROUPS.

Date.

Basis (average 1901-1905).........
1912.

First quarter...............................
Second quarter...........................
Third quarter.............................
Fourth quarter...........................
1913.

End of January..........................
End of February........................
End of March.............................
End of April...............................
End of May................................
End of June................................
End of July................................
End of August...........................
End of September......................
End of October...........................
End of November......................
End of December.......................
1914.

End of January..........................
End of February........................
End of March.............................
End of April...............................
End of May................................
End of June................................
July.............................................
August........................................
September..................................
October.......................................
November...................................

Cereals
and
meats.

Other
Miscel­
food
laneous
products Textiles. Minerals. rubber,
(tea,
timber,
sugar,
oils, etc.
etc.).

Total.

Per­
centage
change.

500

300

500

400

500

2,200

100.0

615
634
624
602

405
379
376
364

571
577
600
619

523
502
531
537

576
603
605
608

2,690
2,695
2,736
2,730

122.3
122.5
124.4
124.1

606
602$
594
603
583
580
584
581$
583
567
569$
563

363
364$
358
352
343
345$
345$
359
359
365
367
355

623
630$
641
638$
630
623$
620
638$
671
667
654
642

534$
519
529
542
542
522$
530
529
523
514
495
491

605$
600$
595
593$
596
597$
609$
585
578
571
575$
572

2,732
2,717
2,717
2,729
2,694
2,669
2,689
2,689
2,714
2,684
2,661
2,623

124.1
123.4
123.4
124.0
122.4
121.3
122.2
122.4
123.3
122.1
121.0
119.2

562$
573$
560
560$
570$
565$
579
641
646
656$
683

356
352
350$
346
349
345
352
369
405
400$
407$

626
630
626|
633$
644$
616
616$
626
611$
560
512

502
491$
493
482$
480
471$
464$
474
472$
458
473

571$
569
567
562$
551
551
553
588
645
652
684$

2,618
2,616
2,597
2,585
2,595
2,549
2,565
2,698
2,780
2,732
2,760

119.0
118.9
118.0
117.5
118.0
115.9
116.6
122.6
126.4
124.2
125.5

INDEX NUMBERS OF AUGUSTUS SAUERBECK.
PUBLICATION.

This index number represents the course of wholesale prices in the
United Kingdom. Prior to 1910 statements were published only once
a year. From January, 1910, to January, 1913, the general result
was published each month for the preceding month, and the yearly
r6sum6 in March, in the Journal of the Royal Statistical Society,
London, but since January, 1913, the information has appeared in
The Statist, London. The general discussion of the results for the
past year is now also published in the April number of the Journal
of the Royal Statistical Society.




270

BULLETIN OF THE BUREAU OF LABOR STATISTICS.
HISTORY.

During 1885, or in the early part of 1886, Mr. Augustus Sauerbeck,
a London wool merchant, prepared a paper upon the gold supply
and its relation to prices, which was published in the September,
1886, Journal of the Royal Statistical Society.1 In much of the
discussion relating to the causes of an “ extraordinary and almost
unprecedented fall of prices” that had continued for 12 years, Mr.
Sauerbeck recognized the lack of statistical information and prepared
this paper in order to supply data upon the subject.
The work thus begun was continued by Mr. Sauerbeck until the
end of 1912, when he relinquished the task and it was taken up by
Sir George Paish, editor of The Statist.
SOURCE OF QUOTATIONS.

The statement is made in the appendix to the first article that
such of the prices from 1846 to 1885 as are not official returns were
received from private firms or collected from the Economist and
other publications. Further than this no information concerning the
source of price quotations is given.
BASE PERIOD.

The 11 years 1867-1877 are taken as the standard period. At the
time the period was chosen the study covered the 40 years 1846-1885
and the base period included the years of the highest prices as well
as a number of years of low prices. The index number for the 11-year
period was found to correspond exactly with the index number of
the 25 years 1853-1877, so that “ a comparison of the aggregate prices
of all commodities in a certain year with the 11 years 1867-1877, is
equivalent to a comparison with the whole 25 years 1853-1877.” 2
PRICES: H O W SHOW N AND COMPUTED.

The prices upon which the index number is based are average
prices for each year. The prices quoted in the report covering the
years 1846-1885 are, with but few exceptions, “ the average prices
in each year, either those officially returned or the averages of the
12 quotations at the end of each month.” 3 Where a range of prices
is given the mean is taken between the highest and the lowest quo­
tations. The prices as given in later reports are the averages of
12 monthly or 52 weekly quotations; in the case of potatoes, of 8
monthly quotations, January to April and September to December.
These annual averages are shown in the tables by articles, as are also
the corresponding relatives. The actual prices from which the
yearly averages are computed are nowhere shown, but relatives based
1 Journal of the Royal Statistical Society, September, 1886, vol. 49, p. 581.
2 Idem, p. 592.
3Idem, p. 632.




INDEX NUMBERS---- GREAT BRITAIN.

271

on the quarterly averages are shown by groups of commodities, cov­
ering the period from 1884 to the present time.
The statement is made in the report of 1893 that all articles have
been calculated at their actual prices and no corrections have been
made for extraordinary fluctuations. The treatment of cotton prices
at the time of the American Civil War is cited as an example.
The prices of all imported articles are quoted “ in bond.”
In the first report the general statement is made that in construct­
ing his table of prices the author has on the whole been guided by the
system adopted in the Economist reports on the course of prices.
NUMBER AND CLASS OF COMMODITIES.

The number of articles used directly in computing the index num­
ber was 43 from 1846 to 1866, 44 from 1867 to 1872, and 45 from
1873 to the present time. All are considered raw materials. In the
original report the statement is made that only commodities were
included which in the United Kingdom at that time represented a
value of about a million pounds or more, counting both domestic pro­
duction and imports. A few important articles, like wine, spirits, and
tobacco, had to be left out, as no reliable data were obtainable.
Certain important commodities are represented by more than one
of the 45 articles; for example, two varieties of wheat are quoted,
and each variety is considered a separate article. The relative prices
of certain others of the 45 articles, as for example coffee, were obtained
by averaging two relatives representing different varieties or grades
of the article. Thus in 1911, when the relative price of Ceylon coffee
was 95 and of good Rio was 91, the relative used for coffee was 93,
the average of these two. This method was followed in cases where
the price of a single variety was not considered sufficiently repre­
sentative of the article. The number of quotations, including these
additional quotations used only indirectly in the computation of the
index number in the report for 1911, was 57. The table of average
actual prices, however, comprised 60 quotations, one each for tea,
copper, and coal being shown in the actual price form without being
represented in the index number. At the time the original report
was published the series of quotations in the table of average prices
comprised a total of 55.
An index number based on the prices of “ the 31 principal com­
modities” from 1818 to 1845 was prepared by Mr. Sauerbeck and
published in his original report. These commodities are not enu­
merated.
DESCRIPTION AND GROUPING OF COMMODITIES.

The 45 articles are divided into six groups. The grouping is shown
. in the following table, which also shows the number of series of price
quotations secured for each commodity, and the number of relative



272

BULLETIN OF TH E BUREAU OF LABOR STATISTICS.

prices for each commodity used directly in the computation of the
index for the year 1911. The table has been compiled from data
appearing in the Journal of the Royal Statistical Society, March,
1911, pages 415 to 420.
NUMBER OF SERIES OF PRICE QUOTATIONS SECURED AND NUMBER OF RELATIVE
PRICES USED IN IN DE X, B Y COMMODITIES.

Commodity.

Number Number
of series of relative
of quota­
prices
tions
used in
index.
secured.

Commodity.

Number ( Number
of series of relative
of quota­ prices
used in
tions
index.
secured.

Iron___
Copper.
T in ;....
L ea d ...
Coal___

W heat...
Flour----Barley...
Oats........
Maize—
Potatoes.
Rice........

Total minerals.

10

Cotton.
F lax...
Hemp..
Jute—
W ool...
S ilk ....
Total textiles.

Total vegetable food___
Beef___
Mutton.
Pork___
Bacon..
Butter..
Total animal food.
Sugar..
Coffee.
Tea___
Total sugar, coffee, and
tea.............................
23

Total food.

H id es....
Leather..
Tallow...
Oil.
Linseed oil and linseed (flax­
seed).....................................
Petroleum................................
Soda crystals...........................
Nitrate of soda........................
Indigo......................................
Total sundry materials

16

Grand total..................

45

A description of the various articles included in the six groups of
commodities follows:1
Vegetable food (8 price series).

1.
2.
3.
4.

5.
6.
7.
8.

Wheat, English Gazette.
Wheat, American.
Flour, town-made white.
Barley, English Gazette.

Oats, English Gazette.
Maize, American, mixed.
Potatoes, good English.
Rice, Rangoon, cargoes to arrive.

Animal food (7 price series).

9.
10.
11.
12.

13. Pork, large and small, average.
14. Bacon, Waterford.
15. Butter, Friesland, fine to finest.

Beef, prime.
Beef, middling.
Mutton, prime.
Mutton, middling.

Sugar, coffee, and tea (8 price series).

16a.
16b.
17.
18a.

Sugar, British West Indian refining.
Sugar, beet, German, 88 p. c. f. o. b.
Sugar, Java, floating cargoes.
Coffee, Ceylon plantation, low mid­
dling.




18b.
19a.
19b.
19c.

Coffee, Rio, good.
Tea, Congou, common.
Tea, average import price.
Tea, Indian, good medium.

i Journal of the Royal Statistical Society, March, 1911, pp. 415-420,

INDEX NUMBERS---- GREAT BRITAIN.

273

Minerals (10 price series).

20a. Iron, Scotch, pig.
20b. Iron, Cleveland (Middlesbrough)
Pig21. Iron, bars, common.
22. Copper, Chile bars.
— Copper, English tough cake.

23. Tin, Straits.
24. Lead, English pig.
25a. Coal, Wallsend, Hetton, in London.
25b. Coal, Newcastle steam.
26. Coal, average export price.

Textiles (11 price

27.
28.
29a.
29b.
30a.
30b.
31.

Cotton, middling American.
Cotton, fair Dholera.
Flax, St. Petersburg.
Flax, Russian average import.
Hemp, Manila fair roping.
Hemp, St. Petersburg clean.
Jute, good medium.

32a. Wool, merino, Port Phillip, average
fleece.
32b. Wool, merino, Adelaide, average
grease.
33. Wool, English, Lincoln half hogs.
34. Silk, Tsatlee.

Sundry materials (16 price

40b.
41.
42.
43.
44.
45a.
45b.

35a. Hides, River Plata, dry.
35b. Hides, River Plata, salted.
35c. Hides, average import.
36a. Leather, dressing hides.
36b. Leather, average import.
37. Tallow, town.
38. Oil, palm.
39. Oil, olive.
40a. Oil, linsead.

Seeds, linseed.
Petroleum, refined.
Soda, crystals.
Nitrate of soda.
Indigo, Bengal good consuming.
Timber, hewn, average import.
Timber, sawn or split, average
import.

SUBSTITUTIONS AND ADDITIONS.

The method of calculating the index adopted when it was deemed
necessary to add or drop quotations for articles is not disclosed. No
mention is made of the necessity of quoting other grades of com­
modities than those formerly quoted, but it is reasonable to believe
that in a period of this length it has been found necessary to do so.
INTERPOLATION.

It may have been impossible to secure complete statistical material
during the full period, but, if so, the author makes no mention of the
fact. Where prices were abnormal, as cotton during the Civil War in
the United States, no corrections were made, quotations being used as
found.
WEIGHTING.

The index number is unweighted. The author has, however, given
to certain important commodities a larger influence in computing the
index number by quoting as separate articles several different varieties
or grades of the same commodity. For example, English wheat and
American wheat constitute two separate articles, as do prime beef
and middling beef; also prime mutton and middling mutton. Simi­
larly, sugar, iron, coal, cotton, wool and oil are each given double
importance in the computation of the index.

14261°—Bull. 173—15------18




274

BULLETIN OF TH E BUREAU OF LABOR STATISTICS.
TESTING

Beginning with the index numbers of the year 1892 1 two tests
were applied, the^one consisting of weighting the relative prices ac­
cording to the “ money-values” of .the commodities in accordance
with their importance in the trade of the United Kingdom during the
3-year period 1889-1891; the second method consisting of weighting
them according to their “ mass-quantities” of other years. In the
latter method the quantities of imports and exports of any one year
are reduced to a nominal money value by multiplying the number
representing the quantity of the article by the number representing
the average prices of said articles during the years 1867-1877.
In his presentation for the year 1895 2 Mr. Sauerbeck used the
geometric average3 of Jevons and calculated by that means a total
index for his 45 articles for the years 1880, 1894, and 1895, and com­
pared it with his own arithmetic averages, both simple and weighted.
After 1907 the single test of weighting according to average “ moneyvalues” of the commodities for the 3-year period 1904-1906 was
employed.
The author makes no direct statement in regard to the process of
finding the nominal values of the several articles, beyond stating that
one factor is the average price of the article during the base period
1867-1877. The other factor, or that quantity which represents the
“ importance in the United Kingdom” of the article, appears to be the
average of the annual production plus imports for the chosen 3-year
period. Due warning is given that this figure, which represents the
total trade in the commodity including reexports, must not be con­
sidered as representing the actual consumption of the commodity in
the United Kingdom.
A second test was in use up to and including the report for 1907.
It is stated that the estimated actual values of the 45 articles con­
sumed in the United Kingdom were obtained by taking the produc­
tion on the basis of Mr. Sauerbeck’s price and the imports at British
Board of Trade values. The ratio of these actual prices to the nom­
inal values on the basis of the average prices from 1867 to 1877 con­
stituted a second weighted index. The explanation of this second
test is not fully stated.
TABLE OF RESULTS.

The principal table in this compilation shows the index number for
the four general groups, and the grand total index. In addition there
are three subindexes of food and an index of all materials contained
1 Journal of the Royal Statistical Society, June, 1893, pp. 215-247,254.
2 Idem, March, 1896, pp. 193,194.
8 To find the geometrical mean, the logarithm is taken of the percentage figure of each article, the total
of all logarithms is divided by 45—the number of articles—and the antilogarithm, the number correspond­
ing to the average logarithm, is the geometrical index number. (Journal of the Royal Statistical Society
March, 1896, p. 194.)




275

INDEX NUMBERS— GREAT BRITAIN.

in the groups of minerals, textiles, and sundry materials. The fol­
lowing data, showing the variations in the group index numbers and
in the general index, have been compiled from various issues of the
Journal of the Royal Statistical Society.1
SUMMARY OF IN D E X NUMBERS, 1846 TO 1913.
(Base period, 1867-1877=100.)

Food.

Date.

Vege­
table
food
(com,
etc.).

1846..........................
1847..........................
1848..........................
1849..........................
1850..........................
1851..........................
1852....... ..................
1853..........................
1854..........................
1855..........................
1856..........................
1857..........................
1858..........................
1859..........................
1860..........................
1861..........................
1862..........................
1863..........................
1864..........................
1865..........................
1866..........................
1867..........................
1868.........../ .............
1869..........................
1870..........................
1871..........................
1872..........................
1873..........................
1874..........................
1875..........................
1876..........................
1877..........................
1878..........................
1879..........................
1880..........................
1881..........................
1882..........................
1883..........................
1884..........................
1885..........................
1886..........................
1887..........................
1888..........................
1889..........................
1890..........................
1891..........................
1892..........................
1893..........................
1894..........................
1895..........................
1896..........................
1897..........................
1898..........................
1899.........................
1900.........................
1901..........................
1902..........................
1903..........................
1904..........................
1905..........................
1906..........................
1907.........................

106
129
92
79
74
73
80
100
120
120
109
105
87
85
99
102
98
87
79
84
95
115
113
91
88
94
101
106
105
93
92
100
95
87
89
84
84
82
71
68
65
64
67
65
65
75
65
59
55
54
53
60
67
60
62
62
63
62
63
63
62
69

Materials.

Ani­
mal
food
(meat,
etc.).

Sugar,
coffee,
and
tea.

81
88
83
71
67
68
69
82
87
87
88
89
83
85
91
91
86
85
89
97
96
89
88
96
98
100
101
109
103
108
108
101
101
94
101
101
104
103
97
88
87
79
82
86
82
81
84
85
80
78
73
79
77
79
85
85
87
84
83
87
89
88

98
87
69
77
87
84
75
87
85
89
97
119
97
102
107
96
98
99
106
97
94
94
96
98
95
100
104
106
105
100
98
103
90
87
88
84
76
77
63
63
60
67
65
75
70
71
69
75
65
62
59
52
51
53
54
46
41
44
50
52
46
48

Total
food.

95
105
84
76
75
74
75
91
101
101
99
102
88
89
98
97
94
89
88
91
95
101
100
94
93
98
102
107
104
100
99
101
96
90
94
91
89
89
79
74
72
70
72
75
73
77
73
72
66
64
62
65
68
65
69
67
67
66
68
69
69
72

Miner­
als.

92
94
78
77
77
75
80
105
115
109
110
108
96
98
97
91
91
93
96
91
91
87
85
89
89
93
127
141
116
101
90
84
74
73
79
77
79
76
68
66
67
69
78
75
80
76
71
68
64
62
63
66
70
92
108
89
82
82
81
87
101
107

Tex­
tiles.

Total
mate­
rials.

86
86
77
75
80
79
84
101
109
109
109
119
102
107
111
109
106
101
98
97
99
100
102
100
99
105
108
106
96
92
95
94
88
85
89
86
85
84
81
76
69
67
67
68
69
69
67
68
64
65
63
62
63
65
71
71
71
69
67
68
74
78
1 September, 1886, p. 648; March, 1891, p. 128; March, 1911, p. 408; April, 1914, p.




77
78
64
67
78
75
78
87
88
84
89
92
84
88
90
92
123
149
162
134
130
110
106
109
106
103
114
103
92
88
85
85
78
74
81
77
73
70
68
65
63
65
64
70
66
59
57
59
53
52
54
51
51
58
66
60
61
66
71
72
80
77

Sundry
mate­
rials.

85
86
73
73
78
76
81
97
104
101
102
107
94
98
100
99
107
115
119
108
107
100
99
100
99
101
115
114
100
93
91
89
81
78
84
80
80
77
73
70
67
67
69
70
71
68
65
65
60
60
60
59
61
70
80
72
71
72
72
75
83
86
556.

Grand
total.

89
95
78
74
77
75
78
95
102
101
101
105
91
94
99
98
101
103
105
101
102
100
99
98
96
100
109
111
102
96
95
94
87
83
88
85
84
82
76
72
69
68
70
72
72
72
68
68
63
62
61
62
64
68
75
70
69
69
70
72
77
80

276

BULLETIN OF TH E BUREAU OP LABOE STATISTICS.
SUMMARY OF INDEX NUMBERS, 1846 TO 1913—Concluded.
Food.

Date.

1908..........................
1909..........................
1910..........................
1911..........................
1912..........................
1913..........................
Average:
1902-1911.
1890-1899.
1878-1887.

Materials.

Ani­
mal
food
(meat,
etc.).

Sugar,
coffee,
and
tea.

70
71
65
70
78
69

89
89
96
90
96
99

48
50
54
61
62
54

72
73
74
75
81
77

89
86
89
93
110
111

62
64
73
76
76
84

73
76
81
81
82
83

74
75
81
83
88
91

73
74
78
80
85
85

66
61
79

88
80
95

49
63
76

71
68
84

90
71
73

70
56
71

74
66
81

77
64
76

74
66
79

Vege­
table
food
(corn,
etc.).

Total
food.

Miner­
als.

Tex­
tiles.

Sundry
mate­
rials.

Total
mate­
rials.

Grand
total.

The general index was 83.5 for January, 1914; 83.8 for February; 82.8 for March; 82.3 for April; and 82.6
for May.

INDIA.
INDEX NUMBERS (RUPEE PRICES) OF FRED. J. ATKINSON.
PUBLICATION AND HISTORY.

Index numbers of rupee prices in India for the years 1861-1895
were first published by Fred. J. Atkinson, Accountant General,
United Provinces, India, in the Journal of the Royal Statistical
Society for March, 1897. Those for the years 1896 and 1897 were
presented in the Journal for June, 1898, and those for the years 1898
to 1901 in the number for March, 1903. In the number for Septem­
ber, 1909, they were brought up to the end of 1908.
In the March, 1907, issue Atkinson makes the following remarks
introductory to his study:
The extraordinary fall in gold prices led to the preparation by
several economists of figures detailing the course that prices had taken
annually for many years back. No attempt has, however, hitherto
been made to deal with the course of silver prices. The consequence
is that various contradictory statements have been made on the sub'ect at different times, and the general impression prevails, probably
>ased on the declared values of exports from India, that silver prices
have been practically stable for the past 25 years, and the inference
drawn is that silver as a measure of value possesses qualities of sta­
bility which gold is declared not to possess. It is with the object of
coming to some definite conclusion that the present figures have been
worked out with considerable difficulty, ana whether the conclusions
arrived at hereafter are agreed to or not, the figures themselves will,
it is hoped, prove useful to economists generally.
This paper deals only with silver prices in India—it would perhaps
be more accurate to say “ rupee prices in India7’ ; but the rupee price
represented the silver price up to the year 1893, and it seems prob­
able that up to that time India practically fixed silver prices through­
out the world. Since 1893, the year in which the mints were closed,
the rupee and silver have diverged, and the prices given are rupee
prices.

i




INDEX NUMBERS---- INDIA.

277

SOURCE OF QUOTATIONS.

In preparing his index numbers, Atkinson, while adopting Sauer­
beck’s principles, did not base his figures on the prices of imported
articles but on those of the native products of India. This was nec­
essary, because the bulk of the trade of India is concerned with the
products of the country, imports representing only some 8 or 9 per
cent of the exported products. Moreover, a considerable portion of
the articles produced are mainly for the purpose of export and are
but little used by the natives of the country. It is evident, therefore,
that prices must necessarily be affected rather by the production in
common use than either by imports or exports.
Sauerbeck takes the majority of the prices he quotes from the Lon­
don market. In India, however, the production and prices vary so
greatly in different parts of the country that to take only one market,
as Calcutta or Bombay, and treat every article as of equal or nearly
equal importance would give a very inaccurate idea of the actual state
of affairs. To avoid this, Atkinson prepared a statement based on
the agricultural returns and financial and commercial statistics pub­
lished by the Government of India, and the administration reports
of the various native States, giving for the year 1893, with a fair
degree of accuracy, the agricultural and manufactured products of
India and the relative importance of each.1
Next, accepting the fact that India in its economic conditions rep­
resents a cluster of different countries, it had to be ascertained in
what particular markets the prices of the various articles should be
taken. Atkinson accordingly prepared a table showing the area of
cultivation in 1893-94 of each product in each Province of India,
and the price of each product was, as far as possible, taken in the
Province or Provinces in which the area of its cultivation is greatest.
The prices taken were obtained from various sources; partly from
the prices current which the chambers of commerce of Calcutta,
Bombay, and Madras issue; partly from the publication “ Prices and
Wages of India,” issued by the Government of India, and partly from
private sources. A few only, when figures were not elsewhere obtain­
able, were taken from the export accounts.
BASE PERIOD.

In the first table of index numbers prepared in 1897 Atkinson took
the year 1871 to represent the number 100. This year was selected
because in it Sauerbeck’s index number of gold prices was 100, and
the price of silver was approximately the same (99.7). This selection
had the advantage of permitting a ready comparison of the course of
rupee prices with that of gold prices, which was useful in connection
with the currency question. It is obvious, however, that prices of a
i See Journal of the Royal Statistical Society, March, 1897, Vol. L X , pp. 124 and 125.




278

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

single year can not be regarded as representing normal prices. As
a matter of fact, rupee prices in 1871 were exceptionally low, and as
the currency question had been settled, Atkinson in 1903 thought it
best to use the average of the years 1868-1876, which may be regarded
as fairly normal years, as the base period for another table of index
numbers. Both tables, that with 1871 and that with 1868-1876 as
the base period, were in 1908 brought up to date.
PRICES: H O W COMPUTED.

Briefly, the procedure adopted in preparing the index numbers was
to ascertain the relative importance of each article as compared with
the total value of all productionin India as shown by the table of pro­
duction for the year 1893. In computing the index numbers for each
month and year for the whole of India, one or more prices at different
important places of production were taken for each article in propor­
tion to its relative importance to the whole. The general index
number is based on the prices of 38 different commodities. Of these
rice, representing three-tenths of the value of all products of India,
was allotted 30 prices out of a total of 100 taken. Similarly wheat
was allotted 5 prices and other grains 5, sugar 4, jowar 3, raggee, gram,
bajra, and meat 2 each, and maize, barley, potatoes, spices, and
ghee 1 each; making a total of 60 prices for food articles. For raw
produce, seeds were allotted 4 prices, cotton 3, hides and skins 3,
jute, indigo, opium, tobacco and timber 2 each, and tea, coffee, salt­
peter, cutch, myrobalans, animal bones, coal, raw silk, and raw wool
1 each, making 29 in all. For manufactures, hides and skins were
allotted 3 prices, cotton goods, jute goods, and oils 2 each, and silk
piece goods and shellac 1 each, making 11 in all. Summarized, the
division was :
Prices.

Articles of food..................................................................................
Raw produce.....................................................................................
Manufactures.....................................................................................

60
29
11

Total.................................................................................................. 100
DESCRIPTION AND GROUPING OF COMMODITIES.

The individual commodities included in the general index numbers
and the markets in which their prices were obtained are the following:
Articles of Food.

1-30. Rice (15prices)}—Monghyr, Calcutta; common, Bengal, eastern division; com­
mon, Calcutta; common, Patna; common, Bengal, Deltaic division; common, Bengal,
Orissa division; common, Madras, southeast coast division; common, Madras, Salem;
common, northwest provinces, eastern division; common, northwest provinces, central
division; common, Burma, Rangoon; common, Burma, Tenasserim division; common,
* As rice is given an importance of 30 and only 15 prices were taken, each price has been doubled in com­
puting the general index number.




INDEX NUMBERS— INDIA.

279

central provinces, Nagpur; common, central provinces, Jubbulpore; and common,
Hyderabad, Bolaram. The quotations used are those published in “ Prices and
Wages” and from prices supplied by dealers.
31-35. Wheat (5 prices).—Common, northwest provinces, Cawnpore; common, Pun­
jab, Delhi; common, central provinces, Nagpur; first quality, Central India, Nussirabad; and flour, Bombay, Poona.
36-38. Jowar 1 (3 prices).—Bombay; Madras, Salem; and Hyderabad, Bolaram.
39-40. Raggee2 (2 prices).—Madras, southeast coast division; and Mysore.
41-42. Gram 3 (2 prices).—Punjab, central division; and, northwest provinces, cen­
tral division.
43-44. Bajra 4 (2 prices).—Bombay, Deccan division; and Madras, Salem.
45. Maize (1 price).—Chota, Nagpur.
46. Barley (1 price).—Delhi.
47-51. Other grains (5 prices).—Arhar, Allahabad; arhar, northwest provinces,
Sub-Montano division; mung, Dal, Lucknow; masur, Lucknow; and chenna, Cawn­
pore. This classification in the agricultural returns includes a large number of differ­
ent varieties of minor grains and pulses grown in different parts of the country, though
the northwest provinces are preeminent in their cultivation. Reliable figures could
be obtained for only a few, and the figures for some of these are not complete. Arhar
(Cajanus indicus), the most important of these minor pulses, for which two prices
have been given, and mung (Phaseolus mungo) are represented by complete figures.
The prices for masur (Erva lens) and chenna (Cicer arietinum), were supplied by the
commissariat department and commence only from 1875 and are averages of the finan­
cial year.
52. Vegetables (1 price).—Potatoes, Bombay.
53-56. Sugar (4 prices).—Gurputty, Calcutta; Dhulloah, Calcutta; Jaggery, cane,
Madras; and Jaggery, Palmyra, Madras. Prices for the two refined sugars have been
taken from the Calcutta prices current, supplemented by prices supplied by Bissouath, Law & Co., and those for raw sugar from the Madras prices current. Consid­
erable difficulty was experienced in the case of refined sugar, as indeed in most of the
quotations taken from the prices current, by the changes in nomenclature, which in
some cases meant a change in actual quality. To continue the same quality through­
out the entire period involved a method of calculation of comparisons. The reault,
however, is said to be approximately accurate.5
57. Spices (1 price).—Ginger, export accounts. Prices taken from the export
accounts.
58. Ghee 6 (1 price).—Bombay.
59-60. Meat (2 prices).—Mutton, Bombay; beef, Bombay.
Raw Produce and Materials.

61. Tea (1 price).—Taken from the export accounts.
62. Coffee (1 price).—Taken from the export accounts.
63-65. Cotton (3 prices).—Dharwar; Broach; and Dholera.

All obtained from the

Bombay prices current.
66-67. Jute (2 prices).—Picked;

and double triangle M.

From Calcutta prices

current.
1A cheap Indian grain used in making a kind of unleavened bread.
2 A cereal grass (eleusina Carocana) largely cultivated for food.
* The chick-pea (Cicer arietinum) of the East Indies, there extensively used as food for men, horses, and
cattle.
4 The spiked or pearl millet (Penicfllaria spicata) one of the commonest food cereals of southeastern Asia.
5Journal of the Royal Statistical Society, March, 1897, Vol. L X , p. 90.
6 Butter clarified by boiling or heating and skimming or straining until it becomes a liquid or semiliquid
oil, capable of being kept for many years. It enters into the composition of nearly everything eaten by
the Brahmans.




280

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

68-69. Indigo (2 prices).—Bengal,

good; and consuming.

From Calcutta prices

current.

70-71. Opium (2 prices).—Behar; and Malwa. From the monthly figures pub­
lished by the Government of India.
72-78. Tobacco (2 prices).—Central India, Nussirabad; ard Bombay.
74-77. Seeds (4 prices).—Linseed, bold; til; rape-yellow, mixed; and castor.
The prices for linseed and rape were taken from the Calcutta prices current, those of
til and castor seeds from the monthly figures published by the Government of India.
78-84. Miscellaneous (7 prices).—Saltpeter, 5 per cent refined; cutch,1 Rangoon;
myrobalans;2 manure, animal bones; coal; raw silk, Surdales; and raw wool. The
prices of saltpeter and raw silk were taken from the Calcutta prices current, those
of cutch, myrobalans, and manure from the export accounts, and those of coal were
furnished by the Bengal Coal Co.
85-87 . Hides and shins (3 prices).—Raw hides, buffalo, Patna, slaughtered, arsenic;
raw hides, cow, Burdwan, slaughtered; and raw skins, goat, Calcutta. Prices taken
from the Calcutta prices current.
88-89. Timber (2 prices).—Bamboos, Calcutta; teak, Rangoon. The prices for
bamboos are taken from the figures published in the Government of India publication
“ Prices and Wages.” The prices on 1st of January of each year beginning from 1871
only are given. Monthly prices for the series of years were not obtainable. The prices
for Rangoon teak were taken from the Calcutta prices current.
Manufacturers.

90-91. Cotton goods (2 prizes).—Yam

1/203, and T cloth, 44 inches. These are
taken from “ Prices and wages” and represent the prices as given to the Government
of India by the Bombay Mill Owners’ Aasociation on 1st of January and July of each
year.
92-98. Jute goods (2 prices).—Bags, No. 2 twill; and bags for California. Prices
taken from the Calcutta prices current.
94-95 . Oils (2 prices).—Castor and coconut. The prices for castor oil were taken
from the Calcutta prices current. Those for coconut oil are from “ Prices and wages ”
the prices on January 1 of each year beginning from 1871 only are given, monthly
prices not being obtainable.
96. Silk piece goods (1 price).—Corah No. 1, from the Calcutta prices current.
97-99. Hides and skins (8 prices).—Tanned hides; cow; tanned skins, goat; and
tanned skins, sheep. Prices are taken from the Madras prices current.
100. Shellac (1 price).—First quality, orange; from the Calcutta prices current.
WEIGHTING.

As has been mentioned above, Atkinson has weighted his general
index numbers by giving to each commodity included as many
quotations as corresponded to its importance in the whole production
value of India in 1893. Since 1893 changes have occurred and some
of the articles included are of more and some of less importance now
than they were then; for example, indigo has declined while coal and
cotton goods have materially increased in importance. But all the
changes combined, according to Atkinson, would not make any
material difference in the percentages as computed. He also remarks
1 An extract from the bark of the mangrove used in tanning and dyeing.
2 Prune-like fruits of several tropical plants of the genus terminalia, used for tanning and calico printing.




INDEX NUMBERS---- INDIA.

281

that during recent years the trade descriptions in some cases and the
quality in others of articles entered in the various chamber of com­
merce prices current, from which several prices are taken, have
undergone a change and that it has been a matter of difficulty to fit
the changes, with accuracy, to the previous order of things.1
TABLE OF RESULTS.

In the table printed in the Journal of the Royal Statistical Society
for September, 1909, pp. 500-502, Atkinson gives, in addition to the
index numbers of 100 articles of Indian production, index numbers for
11 articles of import. These are taken from data published annually
since 1905 by the Commercial Intelligence Department of India.
Under the title “ Variations in Indian Price Levels” this department
has published index numbers prepared on different lines from the sys­
tem adopted by Atkinson— 39 articles only being taken, of which 11
are imported, and all articles being given equal weight. As imported
articles are regulated by their gold price and as the individual articles
are not weighted according to their importance, it is not surprising
that the two sets of index numbers do not agree, though their general
trend is practically the same. It having been suggested, however, that
so far as the dweller in India is concerned the prices of imports now
materially affect his annual expenditure, Atkinson shows in column 6
of his table the index number of 11 articles of import equated to
1868-1876, as given by the commercial intelligence department, and
then adds the index numbers of these 11 articles of import to those of
the 100 articles of Indian production to which his own index numbers
relate, and shows the total index number in column 7 of his table.
In column 8 are shown Sauerbeck's index numbers for gold prices;
in column 9 the gold price of silver; in column 10 the gold price of the
rupee; in column 11 index numbers of articles of export; in column 12
Sauerbeck’s gold prices index number of the 11 articles of import
shown in column 6; and in column 13 Sauerbeck’s gold prices index
number of 11 articles exported by India.
This table is reproduced below with the exception of the data
contained in columns 12 and 13.
1 Journal of the Royal Statistical' Society, September, 1939, Vol. L X X II, p. 497.




282

BULLETIN OF THE BUREAU OF LABOR STATISTICS.
IN D E X NUMBERS OF PRICES IN INDIA, 1870 TO 1908.

[Source: Journal of the Royal Statistical Society, September, 1909, Vol. L X X II, pp. 500-502.]
Index lumbers (percentsiges) of
rupee prices in India (av erage of
1868-1S*76=100).

rear.

1

Index
num­
Index
bers
num­
111 ar­
bers
ticles
of 11
Manu­
All
Raw
articles includ­
Food produce
factures prices
ing 11 ar­
of
(60
(29
import.
ticles of
(100
(n
articles). articles). articles).
articles).
import.
2

1870...........
1871.
1872.
1873.
1874.
1875..
1876.
1877.
1878..
1879.
\.......
1880..
1881.
1882.,
1883.
1884.
1885..
1886..
1887.
i
i.......
1890..
1891.
1892.
1893.
1894.
1895.
1896.
1897..
1898.
1899.i.......
1900.i......
1901.
1902.
1903.
1904.
1905.
1906.
1907.
1908..

4

3

104
85
91
96
107
92
98
142
155
137
108
93
93
96
107
107
103
103
111
116
118
123
138
131
121
113
133
171
131
122
152
148
131
124
118
139
167
178
202

103
99
103
99
103
96
98
102
102
104
108
103
100
100
100
98
99
101
106
111
108
106
115
117
118
125
120
114
90
111
120
117
113
113
115
116
128
134
127

5

106
106
105
100
105
100
88
90
93
95
103
102
97
97
91
85
90
96
100
109
102
100
103
113
116
118
111
103
98
102
104
106
112
108
118
121
140
144
122

7

6

105
91
95
97
105
94
97
125
135
123
106
96
95
96
104
103
100
101
108
114
114
116
128
125
119
116
127
149
122
117
139
135
124
119
117
130
153
162
174

100
93
96
105
104
95
96
93
88
87
93
90
89
83
82
79
84
87
97
96
96
88
88
94
88
91
99
90
84
91
102
101
90
93
98
101
110
122
113

Index
Sauer­
Gold
num­
beck’s
Gold
price
ber of
gold
price
o
f
silver
articles
prices
of
of
(average (60.84d.
rupee
per
export
1867(23.34d.
ounce
(18681877
=100).
1876
= 100). = 100).
=100).
8

105
91
95
98
105
94
97
122
130
119
105
95
94
95
102
101
98
100
107
112
112
113
124
122
116
114
124
143
118
114
135
132
121
116
115
127
149
158
168

9

96
100
109
111
102
96
95
94
97
83
88
85
84
82
76
72
69
68
70
72
72
72
68
68
63
62
61
62
64
68
75
70
69
69
70
72
77
80
73

10

100
100
99
99
96
93
87
90
86
84
86
85
85
83
83
80
75
73
70
70
78
74
65
58
48
49
50
45
44
45
46
45
40
41
43
46
51
50
40

100
100
99
99
96
93
87
90
86
84
86
85
85
83
83
80
75
73
70
70
78
74
65
64
57
57
61
65
68
69
68
68
69
69
69
66
69
69
68

11

103
97
101
102
108
98
96
105
107
109
106
100
96
97
100
94
93
97
101
106
103
105
115
122
116
115
113
115
102
102
112
110
110
108
114
113
128
136
131

ITALY.
INDEX NUMBERS OF THE ANNUARIO STATISTICO ITALLANO.
PUBLICATION.

Index numbers based on the prices of a limited number of com­
modities at wholesale are contained in the annual statistical report
for Italy (.A n n u a r io S ta tis tic o I t a lia n o ), issued from the Office of the
Director General of Statistics and Labor (D ir e z io n e g en er a te d e lla ,
sta tis ca e d e l T avoro).
HISTORY.

Since 1886 the Annuario Statistico Italiano has presented tables
showing fluctuations in the prices of a large number of commodities,
both raw and manufactured, during a series of years. In the earlier
reports these prices were shown for a period extending, in some




INDEX NUMBERS— ITALY.

283

instances, back to 1862. In more recent issues the figures have been
limited as a rule to the last five years preceding the date of publication.
Prior to 1912 no index numbers appear to have been computed,
the data being given only in the form of actual average prices. In
the report for 1912, however, was begun the publication of two
series of index numbers based in the one case on the prices of a few
articles of food furnished to the army, and in the other case on a
larger number of articles of the same class supplied to 43 national
boarding schools (c o n v itti n a z io n a li) of Italy. These index numbers
were continued in the report for 1913.
SOURCE OP QUOTATIONS.

The price quotations on which the index numbers are based were
furnished by the directors of the schools and by the minister of war
( m in is te r o d ella g u e rr a , d ir e z io n e g en era te d e i s e r v iz i lo g is tic i e a m m in is tr a tiv i ).
BASE PERIOD.

The five years 1890-1894 constitute the base period in the series
relating to the boarding schools. In the series for the army the
relatives are based on the period 1900-1904.
PRICES; H O W SHOW N AND COMPUTED.

Only the average annual prices of the different commodities
included in the two indexes are given in the reports. In several
instances data for earlier years are lacking from the figures relating
to the army.
NUMBER AND CLASS OF COMMODITIES.

The table of index numbers for supplies furnished to the army
contains 8 commodities, while that for boarding schools contains 13
commodities. All articles belong to the food group.
DESCRIPTION AND GROUPING OF COMMODITIES.

The following articles are included in the table of index numbers for
the army: Corn, bread (ration), Italian paste, rice, beef (young steer),
coffee (roasted), sugar, and wine. The list of articles supplied to
boarding schools for which index numbers are shown includes bread,
Italian paste, rice, beef, sausage, fish (in oil), eggs, butter, oil, milk,
coffee, sugar, and wine. In the latter series the index for beef is based
on the average of the prices paid for young steer flesh and veal in a
single institution. No further description of the commodities is
furnished.
SUBSTITUTIONS, ADDITIONS, AND INTERPOLATION.

So far as can be determined from the information at hand, no
additions to the list of articles or substitutions of one grade or quality
of an article for another have been made. No prices appear to have
been interpolated.



284

BULLETIN OF THE BUREAU OP LABOR STATISTICS.
WEIGHTING.

All commodities are given equal weight in the computation of the
general index number for each series.
TESTING.

The accuracy of these index numbers is not tested by comparison
with similar data or by other means, so far as the published results
show.
TABLE OF RESULTS.

The table following contains the index numbers for the 13 commodi­
ties furnished to boarding schools during the period 1890 to 1913,
inclusive.
IN D E X NUMBERS BASED ON PRICES PAID BY THE NATIONAL SCHOOLS (CONVITTI
NAZIONALI) FOR COMMODITIES NECESSARY FOR THE NOURISHMENT OF THE
PUPILS, 1890 TO 1913.
(Base period, 1890-1894=100.)

Commodity.

Bread, kilograms___
Italian paste, kilo­
grams ....................
Rice, kilograms........
Beef,1kilograms.......
Sausage, kilograms..
Fish (in oil), kilo­
grams.....................
Eggs, dozen..............
Butter, kilograms.. .
Oil, liters...................
Milk, liters................
Coffee, kilograms___
Sugar, kilograms......
Wine, liters...............

Av­
erage
prices
for 1890
1S901894.
$0,356

1891

1S92

99.7 101.1 107.6

Bread, kilograms___
Italian paste, kilo­
grams ....................
Rice, kilograms........
Beef,1kilograms.......
Sausage, kilograms..
Fish (in oil), kilo­
grams.....................
Eggs, dozen..............
Butter, kilograms__
Oil, liters...................
Milk, liters................
Coffee, kilograms___
Sugar, kilograms......
Wine, liters...............

99.2

.515 102.1 101.2 103.1 99.6
.500 102.1 101.7 103.1 96.5
1.624 99.4 104.7 100.5 98.8
2.681 99.8 98.9 101.1 101.1
2.123
.815
2.792
1.251
.288
4.051
1.506
.357

Av­
erage
prices 1902
for
18901894.
10.356

1894

92.1

1895

95.5

1896

92.4

96.9

1903

95.5

1904

91.3

98.8

1905

91.3

1897

1898

1899

1900

1901

94.1 103.9 103.1 102.8 100.0

94.2 91.8 91.7 97.9 99.8 100.0 99.0 98.1
96.3 97.5 96.1 101.3 99.1 98.5 95.7 93.9
96.4 97.0 95.2 93.8 91.4 90.7 90.6 93.5
98.8 100.3 101.3 101.5 102.2 98.2 104.2 104.8

100.0 101.0 101.7 98.5 98.5 100.0 101.4
101.1 99.3 100.4 99.0 100.3 99.7 100.7
99.5 97.4 101.0 100.2 101.7 101.0 100.2
99.3 99.4 102.1 101.1 98.3 97.6 92.8
103.8 102.1 97.2 100.6 96.5 97.5 95.4
98.8 99.1 96.5 99.5 106.0 105.1 102.7
97.4 96.2 98.6 100.1 107.5 100.9 101.7
115.5 110.0 90.7 90.7 93.2! 94.6 102.8

General index
number........ ......... 101.4 100.9 100.3

Commodity.

1893

|
98.4j 98.3

1906

93.8

1907

93.0

98.0

102.0 102.6 100.8 103.1 100.2
100.9 101.7 100.2 107.8 108.9
102.4 98.8 97.4 98.4 102.0
91.5 105.0 100.7 110.8 111.1
91.6 93.0 93.7 89.2 89.9
98.9 93.9 85.3 83.8 82.4
100.3 99.8 101.0 101.1 100.9
91.3 94.3 95.2 94.9 93.8
97.5

1908 1909

98.9

1910

97.3

1911

98.6

98.4

1912 1913

97.2 103.7 106.8 104.8 108.2 111.1

.515 95.7 91.7 89.7 90.5 91.8 94.4 100.6 102.5 102.1 99.2 103.5
.500 93.7 93.7 93.3 92.1 93.7 94.3 96.0 98.4 100.5 95.5 95.3
1.624 93.8 92.6 96.6 101.4 102.6 102.8 102.6 111.2 112.0 124.7 .132.6
2.681 104.3 107.6 109.1 111.6 109.4 111.7 118.9 118.4 130.6 139.1 138.3

107.3
103.9
124.6
146.5

2.123 103.6 115.0 104.2 102.4 104.8 106.9 117.3 127.5 133.6 140.1 134.6
.815 111.0 111.2 110.9 114.9 115.9 121.4 126.2 131.5 133.4 140.0 140.9
2.792 99.9 101.3 101.7 100.6 103.1 104.9 107.4 111.0 112.4 113.9 118.1
1.251 103.8 102.7 95.2 100.0 94.3 101.0 105.1 145.3 135.8 146.6 138.1
.288 92.7 92.0 89.9 91.6 93.4 98.6 99.6 103.8 109.3 112.5 114.9
4.051 78.5 75.6 74.6 • 74.5 75.7 75.6 77.4 76.4 76.1 84.2 92.9
1.506 96.6 93.8 92.2 97.4 96.0 96.7 96.8 96.4 97.8 100.1 104.8
.357 87.3 89.3 89.6 89.3 92.1 98.3 84.5 71.7 77.5 123.5 133.8

147.0
145.3
120.0
140.1
121.1
96.0
97.6
122.0

General index
number........ .........

96.8

97.1

95.3

96.7

97.4 100.0 102.3 107.5 109.8 117.2 119.7 121.8

1 The prices used represent the average prices paid for beef (steers) and veal in one school. Annuario
Statistico Italiano, 1912, p. 138; 1913, p. 184.




INDEX NUMBERS— ITALY,

285

INDEX NUMBERS OF ACHILLE NECCO.
PUBLICATION AND HISTORY.

A volume entitled “ LaCurva dei Prezzi delleMerci in Italia negli
Anni 1881-1909” (The Price Curve of Commodities in Italy during
1881-1909), which was published in Turin by Achille Necco in 1910,i
contains four series of index numbers based in each case on the import
or export values of certain important articles of commerce. Com­
parative tables showing the price fluctuations in several countries
on a common basis, that of the year 1881, are also contained in the
volume.
A continuation of the two principal series of these index numbers
has been published by Necco in La Riforma Sociale for 1911, pages
68-72, and 1913, pages 621-635; also in a special bulletin entitled
“ Prezzi della Merci in Italia nel 1912” (Prices of commodities in Italy
during 1912), appearing in 1914. In the last-named publication
which also was issued by La Riforma Sociale, the figures for 1910,
1911, and 1912 have been supplied.
SOURCE OF QUOTATIONS.

The data used in computing the index numbers were those formu­
lated by the commissione centrale dei valori per le dogane (central
commission for customs valuation) and published in the reports of
the Ministry of Agriculture, Industry, and Commerce.
BASE PERIOD.

The values of imported and exported articles for the year 1881,
taken as 100, constitute the bases on which the two principal series
were computed. The other two series appearing in the first volume
issued in 1910 are based on the method employed by Pantaleoni and
have for the standard of measurement the values of imports and
exports, respectively, in 1878.
NUMBER AND GROUPING OF COMMODITIES.

In each of the two main series of index numbers the groupings
adopted in the collection of the customs revenue have been followed.
Within the 16 principal groups there are approximately 400 different
articles, each of which in turn may comprise several varieties.
The 16 groups of commodities are as follows:
1.
2.
3.
4.
5.
6.

Spirits, beverages, oils.
Colonial products, spices, tobacco.
Chemical products, medicinal substances, resins, gums, etc.
Coloring and other materials for dyeing and tanning.
Hemp, flax, jute, and other fibrous plants.
Cotton.

1 Societa Tipografia-Editrice Nazionale (gia Roux e Viarenga), Torino. The same information also was
published as a supplement to La Riforma Sociale, Vol. X X I, September-October, 1910.




286
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

Wool, horsehair, and other hair.
Silk.
Wood and straw.
Paper and books.
Hides.
Minerals, metals and their manufactures.
Stone, earthen, glass, and crystal ware.
Cereals, flour, Italian paste, and vegetable products.
Animals and animal products.
Miscellaneous commodities.

The two index numbers of import and export values computed
according to the method adopted by Pantaleoni are likewise based
on commodities selected from the tariff schedules. There are 19
import and 12 export commodities as follows:

Imports:
1. Petroleum, refined.
2. Coffee, raw.
3. Sugar, second grade.
4. Dyes, in dry state.
5. Cotton, raw.
6. Cotton cloth, unbleached.
7. Wool, natural or unwashed.
8. Woolen cloth (combed wool).
9. Woolen cloth (carded wool).
10. Hides of oxen and cows.
11. Leather.
12. Cast iron (in pigs or plates).
13. Bar iron and steel in sheets.
14. Copper, brass, and bronze.
15. Machinery (not specified).
16. Grain

Imports—Concluded.
17. Cheese.
18. Dry goods, ordinary.
19. Dry goods, fine.
Exports:
1. Wine in bottles.
2. Olive oil.
3. Boric acid.
4. Sulphate of quinine.
5. Soap (common).
6. Hemp, raw.
7. Silk, raw.
8. Straw braid for hats.
9. Sulphur, raw and refined.
10. Oranges and lemons.
11. Almonds (shelled).
12. Coral (polished).

WEIGHTING.

In the computation of his two main series of index numbers Necco
adopted the method employed by De Foville in following the changes
in the import and export values of France from 1826 to 1880. It
consists in weighting the prices of the first of any two consecutive
years with the mass quantities of imports or exports of the second
year. The price of the preceding year in any instance is multiplied
by the mass quantity of the current year, giving what is termed the
“ provisional value.” The price of the current year is then multiplied
by the mass quantity to produce the “ actual value.” These pro­
visional and actual values of the imports or exports are then summed
and compared to ascertain the increase or decrease which has taken
place in these values as between the two years under comparison.
For example, it might be found that the import values of 1881 were
2 per cent lower than those of 1880; those of 1882, 4 per cent higher
than those of 1881; those of 1883, 3 per cent lower than those of




INDEX NUMBERS---- ITALY.

287

1882; and so on. Assuming now that the import values of the year
1880 are taken as the base, or 100, the index for 1881 would be 98,
since the import values in 1881 decreased 2 per cent from those of
1880. Again in 1882 the import values increased 4 per cent over
what they were in 1881; that is, 4 per cent of 98, or *3.92. Therefore
the index for 1882 is 98 plus 3.92, or 101.92. In like manner, since
import values in 1883 were 3 per cent lower than in 1882, the index
for 1883 becomes 97 per cent of 101.92, or 98.86.
It is seen that under Necco’s system there is a constantly changing
weight, namely, the quantity of an article imported or exported
each year. No direct relationship exists between the index number
of any one year and that of the basic year, since, as has been said,
the relative importance of a commodity changes from year to year
according to the quantity imported or exported, as the case may be.
Pantaieoni likewise employed a fluctuating weight in determining
the relative importance of the commodities entering into his index
number. Under his original plan there was determined each year
the percentage which the value of each commodity imported or ex­
ported, as the case might be, formed of the total value of all imported
or exported commodities. This figure was then used as a weight
for each commodity included in the final index number. To simplify
this rather laborious process from year to year, Pantaieoni sug­
gested—and Necco followed the suggestion in extending Pantaleoni’s
indexes—that it would be advisable to ascertain the average import
or export value of each commodity concerned over a period of years
and then calculate the ratio between the average value of each com­
modity so ascertained and the total average value of all imports and
exports over the same period and use the result as a weight for each
commodity for each of the years involved in the period under con­
sideration. Necco has done this for each of the three periods, 18901895, 1896-1901, and 1902-1908, in the case of imports and for each
of the two periods, 1890-1898 and 1899-1908, in the case of exports.
The result of this method is that there is employed a constant
weight over a limited number of years, a weight which may be termed
the average importance of the particular commodity as determined
by its proportionate value in the total import or export trade of the
country, as the case may be.
TABLE OF RESULTS.

On page 32 of Necco’s original work1 are given the following index
numbers for imports and exports, respectively. The first series in
each case is Necco’s own number computed according to the method
of Benini. The two remaining series are those of Pantaieoni and
i La Curva dei Prezzi delle Merci in Italia negli Anni 1881-1909,




288

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

Benini reduced for the sake of comparison to a common-base period,
that of the year 1881. Necco’s figures for the years 1910 to 1912
have been supplied from the “ Prezzi della Merci in Italia nel 1912,”
published in 1914.
IN D E X NUMBERS OF ITALIAN IMPORTS AND EXPO R TS.
(Base period, 1881-100.)

Imports.
Necco.
1881...........................................
1882...........................................
1883...........................................
1884...........................................
1885...........................................
1886...........................................
1887...........................................
1888...........................................
1889...........................................
1890...........................................
1891...........................................
1892...........................................
1893...........................................
1894...........................................
1895...........................................
1896...........................................
1897...........................................
1898...........................................
1899...........................................
1900...........................................
1901...........................................
1902...........................................
1903...........................................
1904...........................................
1905...........................................
1906...........................................
1907...........................................
1908...........................................
1909...........................................
1910...........................................
1911...........................................
1912...........................................

100.00
96.86
93.01
87.42
82.68
81.95
79.53
81.19
82.58
83.23
79.25
77.43
76.73
71.81
71.04
70.96
70.42
74.49
79.77
86.47
79.65
76.75
77.73
80.05
79.52
84.29
87.96
84.55
85.45
86.55
87.35
89.85

Pantaleoni.

Exports.
Benini.

100.00
98.98
94.90
86.73
78.57
74.49
71.43
73.47
73.47
73.47
71.43
68.37
66.33
61.22
60.20
60.20
60.20
62.24
66.33
66.33
65.31
63.27
64.30
66.33
67.35
71.43
73.47
73.47
76.53

100.00
98.43
94.06
87.64
82.49
80.81
79.61
79.97
81.58
82.24
78.62
77.04
76.41
71.14
69.47
69.87
69.30
73.23
76.81
83.17
77.37
74.48

Necco.
100.00
96.84
91.96
88.08
84.64
84.11
79.62
76.73
80.49
81.72
76.31
76.37
76.18
71.97
72.83
69.02
67.80
69.09
75.55
75.10
72.73
74.10
76.92
76.07
77.12
79.54
83.72
77.88
79.29
82.12
83.44
S3.54

Pantaleoni.
100.00
92.71
86.46
85.42
79.17
84.37
80.21
69.79
78.12
78.12
67.71
77.08
77.08
68.75
70.83
62.50
62.50
64.57
79.17
72.92
69.79
72.92
78.12
72.92
73.96
78.12
86.46
69.79
72.92

Benini,
100.00
94.44
91.43
85.62
81.95
81.64
77.80
74.40
76.89
79.58
74.25
74.33
74.15
70.59
71.60
67.72
66.40
67.79
74.02
73.58
71.25
72.14

JAPAN.
INDEX NUMBERS OF THE DEPARTMENT OF AGRICULTURE AND COMMERCE.
PUBLICATION.

Wholesale prices, with index numbers for the same, are published
annually in the reports issued by the Department of Agriculture and
Commerce of Japan. These reports, which are printed in English as
well as Japanese, contain various data of a statistical nature relating
to the agricultural, mining, manufacturing, and other industries of
the Empire and its dependencies. No text analysis is given of the
tables included in the reports.
HISTORY.

The quotations of prices, according to N. Hanabusa, director of
the Japanese bureau of statistics, were not matters of record prior to
1886 except for the four standard commodities: Rice, barley, beans,




INDEX NUMBERS---- JAPAN.

289

and sake, fo!r which there were incomplete records for earlier years.1
The latest available report, issued in March, 1914, is the twentyninth of the series.
SOURCE OF QUOTATIONS.

Wholesale prices of the articles for which index numbers are com­
piled are obtained from the cities of six statistical divisions of the
Empire by the Department of Agriculture and Commerce. No
statement is made in the reports as to the methods of securing this
information.
BASE PERIOD.

The base period for each year from 1901 to 1912, inclusive, is the
year 1900 taken as 100. In each of the reports for this series of
years, except the one issued in 1905, index numbers are computed for
the years from 1900 to two years earlier than the date of publication.
For example, the report published in 1908 contains index numbers for
the years 1900 to 1906, inclusive. The latest report, issued in 1914,
comprises index numbers for the years 1904 to 1912 only, 1900 being
still retained as the base or 100.
The report of 1905, which is the earliest one available, uses the
year 1887 as a base or 100 and shows average annual prices and
index numbers for each year for the several commodities from 1887
to 1903 inclusive. In the case of a few commodities for which data
for 1887 were lacking, a subsequent year was used as the base. There
is no general index number in this volume for the groups of com­
modities as in those for succeeding years.
PRICES: H O W SHOW N AND COMPUTED.

Average annual prices are published for each commodity taken
separately for all years subsequent to and including 1900. In the
report of 1905 there is no grouping of the commodities, while in those
for other years the commodities are divided into three main groups—
food, clothing, and material.
Following the average annual prices of the different articles for the
Empire as a whole, there is a table showing for each commodity the
average monthly price and the average price for the year of that
commodity in each of the principal cities of the several statistical
divisions and in the country at large, all data being for the year prior
to the one preceding the publication of the report.
NUMBER AND CLASS OF COMMODITIES.

Sixty-five commodities are quoted in the report for 1912, which is
the latest year for which data are available. Leaf tobacco was not
quoted after 1905, nor cut tobacco after 1907.
1

Bulletin de l’lnstitut International de Statistique, tome X I X , 3®livraison, p. 237.

94261°— Bull. 173— 15--- 19




290

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

The 65 commodities, including both raw and manufactured articles,
are as follows:
Rice, superior.
Rice, medium.
Rice, inferior.
Barley.
Naked barley.
Wheat.
Soy beans.
Small red beans.
Salt.
Soy (sauce).
Miso (sauce, soy beans, rice, water).
White sugar, domestic.
White sugar, foreign.
Brown sugar, domestic.
Brown sugar, foreign.
Sake (rice liquor).
Tea.
Bonito (fish), dried.
Beef.
Eggs.
Milk.
Umeboshi (pickled plum).
Takuwan (pickled radish).
Ginned cotton, domestic.
Ginned cotton, foreign.
Cotton yams, domestic.
Cotton yarns, foreign.
Raw silk, superior.
Raw silk, medium.
Raw silk, inferior.
Hemp.
Bleached cotton fabric, domestic.
Grey shirting, foreign.

Calico.
Silk tissue, for lining.
Kaiki silk.
Petroleum.
Coal.
Firewood.
Charcoal.
Rapeseed oil.
Mino paper.
Hanshi paper.
Indigo (Japanese).
Balk (beam) pine.
Balk (beam) sugi.
Balk (beam) keyaki.
Balk (beam) fir.
Plank, pine, 6 bu (.7155 inch) thick.
Plank, sugi, 6 bu (.7155 inch) thick.
Plank, pine, 4 bu (.477 inch) thick.
Plank, sugi, 4 bu (.477 inch) thick.
Log, pine.
Log, sugi.
Shingles.
Sleeper, chestnut.
Sleeper, Hinold.
Pig iron, domestic.
Pig iron, foreign.
Nails, foreign.
Straw.
Hay.
Dried sardine (for manure).
Herring (for manure).
Rapeseed cake.

DESCRIPTION AND GROUPING OF COMMODITIES.

In the report for 1904 and in subsequent issues the commodities for
which average yearly prices and index numbers are given are divided
into three groups: (1) Food, etc.; (2) clothing; (3) materials. Under
food the following articles are listed: Rice, barley, naked barley,
wheat, soy beans, red beans, salt, sake, soy (soy-bean sauce), miso, tea,
bonito (dried fish), beef, eggs, milk, umeboshi (pickled plum), takuwan (pickled radish), sugar (4 grades), tobacco (2 grades)—a total of
23 articles.
Under clothing are listed ginned cotton (2 grades), cotton yarns (2
grades), raw silk, hemp, bleached cotton fabric, gray shirting, calico,
silk tissue, kaiki silk— 11 articles in all.
Under materials are listed the following: Petroleum, coal, firewood,
charcoal, rapeseed oil, paper (2 grades), indigo, beams (4 grades),




INDEX NUMBERS---- JAPAN.

291

planks (4 grades), logs (2 grades), shingles, sleepers (2 grades), pig
iron (2 grades), nails, straw, hay, manure (fish, 2 grades), rapeseed
pake—altogether 29 articles.
As previously stated, leaf tobacco was discontinued after 1905 and
cut tobacco after 1907.
SUBSTITUTIONS AND ADDITIONS.

So far as the reports show, no substitutions of a particular grade or
quality of an article for another grade or quality of the same article or
for a different article have been made at any time. Additions to the
list of articles have apparently been made from time to time as data
became available. In all such cases the average price for the earliest
year for which data were available was taken as the base, or 100. No
additions have been made since 1901, in which year logs (pine and
sugi1) appear to have been included for the first time.
INTERPOLATION.

No interpolation of prices has been made in any of the reports so far
as can be determined. In cases where prices for a particular month
in a given locality were lacking, the averages for the year and for the
country at large have been based on the data for the remaining months
and localities.
WEIGHTING.

There is no attempt at weighting any of the commodities for which
index numbers have been computed, apart from the inclusion in the
tables of several grades of the more important articles.
TESTING.

No comparison of these index numbers with those for other coun­
tries has been made in the reports, nor have other means of testing
their accuracy been employed, so far as the published results show.
TABLES OF RESULTS.

The index numbers for average wholesale prices of four staple
articles, viz, rice, barley, beans, and sake,2 from 1881 to 1909, inclu­
sive, are shown in the following table published in the bulletin of the
international statistical institute.3 This table was compiled by the
director of the Japanese imperial statistical bureau, Mr. N. Hanabusa,
and is based on data collected annually during the months of March,
June, September, and December from six principal cities of the Empire
by the minister of agriculture and commerce. Only medium grades
of the articles for which quotations were secured have been included
in the compilation.
1A genus of evergreen trees of the pine family.
2 A native beer made from rice.
3 Bulletin de l’lnstitut International de Statistique, tome X IX , 3« livraison, p. 239.




292

BULLETIN OF TH E BUBEAU OP LABOR STATISTICS.
INDEX NUMBERS FOR FOUR PRINCIPAL COMMODITIES, 1881 TO 1909.

Year.

lice.

1881.

1882.
1883.
1884.
1885.
1886.
1887.
1888.
1889.
1890.
1891.
1892.
1893.
1894.
1895.
1896.
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909

Barley.

Beans
Sake (rice
(Japanese). wine, clear).

100

100

100

100

84
60
52
64
56
52
48
61
90
75
77
78
91
90
103
129
145
108
125
127
132
153
143
137
155
173
167
139

88
63
59
64
55
45
41
47
72
70
64
64
72
73
71
94
117
85
84
72
81
113
137
123
85
106
115
103

102
77
61
64
57
58
57
71
78
74
73
79
81
84
93
113
119
123
115
108
104
116
147
146
139
149
134
117

89
78
73
94
80
83
76
80
85
84
84
83
91
102
118
146
170
165
192
201
201
205
208
223
239
240
252
251

General
index.
100

90
72
64
77

66
66
61
69
83
78
78
79
87
92
103
129
147
133
147
149
150
164
172
175
180
189
192
180

The following table, compiled from the 28th and 29th reports of the
Japanese Department of Agriculture and Commerce, shows the
index numbers for each of the three groups—food, clothing, and
materials, and for the three groups combined, by years, from 1900 to
1912. The indexes for the groups as a whole apparently were obtained
by taking the simple average of the index numbers for all articles
included in the three groups reported.
IN DEX NUMBERS FOR THE TH REE PRINCIPAL GROUPS OF COMMODITIES, 1900 TO 1912.

Years.

Food.

Clothing.

Materials.

All groups
combined.

1900 ...........................................................................................

100

100

100

98
102
111
124
135
129
135
136
133
132
139
154

100

1901.....................................................................................
1902.....................................................................................
1903.....................................................................................
1904....................................................................................
1905....................................................................................
1906....................................................................................
1907....................................................................................
1908....................................................................................
1909....................................................................................
1910....................................................................................
1911....................................................................................
1912....................................................................................

100
98
102
109
120
122
129
120
119
124
133
130

95
92
94
95
100
101
110
113
107
110
114
119

97
97
101
108
116
114
122
122
118
120
126
133

1 For index numbers of the separate commodities of each group see the statistical reports of Department
of Agriculture and Commerce for 1911 and 1912.




INDEX NUMBERS— NETHERLANDS.

293

NETHERLANDS.
INDEX NUMBERS OF THE NETHERLANDS STATISTICAL OFFICE.
HISTORY AND PUBLICATION.

This index of wholesale prices has been prepared by the Nether­
lands Statistical Office and appeared for the first time in the monthly
journal of that office in June, 1914.1 It grew out of a compilation of
tables on wholesale prices presented by the Statistical Office in its
yearbook of 1913, and covered the years 1885 to 1913; it is being
continued in the monthly journal of this same office.
SOURCE OF QUOTATIONS.

Price quotations are reported from the different wholesale markets,
the number and place of these markets not being specifically men­
tioned. Monthly price quotations are averaged annually.
BASE PERIOD.

The base period is 1893, the average price of that year represent­
ing 100.
NUMBER AND DESCRIPTION OF COMMODITIES.

There are twelve commodities for which separate index numbers
are given; no general index for all commodities combined has been
presented. A graphical presentation is shown for each of the com­
modities in four groups: (1) Wheat, maize, and rye; (2) crude grain
alcohol, rapeseed, and flaxseed oil; (3) oleomargarine (best quality),
coffee (Java), and beet sugar; (4) petroleum, tin (Banca), and refined
grain alcohol.
WEIGHTING.

A simple arithmetical average has probably been employed, no
mention being made of any kind of weighting.
TABLE OF RESULTS.

Index numbers for each of the twelve commodities are shown in
the table which follows.
>Maandschrift van het Centraal Bureau voor de Statistiek. The Hague, 1914, vol. 9, No. 6 (June) pp.
461-466.




294

BULLETIN OF THE BUREAU OP LABOR STATISTICS.

RELATIVE PRICES (IN DEX NUMBERS) FOR TW ELVE COMMODITIES IN THE WHOLE
SALE MARKETS OF THE NETHERLANDS, 1885 TO 1913.
[Source: Maandschrift van het Centraal Bureau voor de Statistiek.
p. 462.]

The Hague, 1914, vol. 9, No. 6 (June),

(Base period, 1893=100.)

Year.

Rye,
Wheat, Maize,
Amer­
Odessa. ican. Petrograd.

Alco­
hol, Rapegrain,
crude.

OleoOil, margaCoffee, Sugar,
flax- rine,
first Java. beet.
lualT y.

106.88 102.96 113.42 110,94 108.09
100.79 90.81 80.93 94.60 100.68
101.42 81.94 90.0' 94.89 102. i4
112.49 81.16 88.91 108.39 93.28
96.84 105.16 102.68 129.69 98.72

1885.,
1886..
1887..

124.19
124.58
123.88
122.03
119.01

1890..
1891..
1892..
1893..
1894..

129.52 98.97 106.77 112.62 121.18 112.43
147.22 137.63 150.51 135.53 119.50 105.00
107.45 107.29 125.50 106.02 105.9' 88.49

1895..
1896..
1897..

84.99
96.88
118.59
124.77
110.16

1900..
1901..
1902.
1903..
1904.

109.45
110.23
106.01
109.84
121.64

1905.
1906.
1907.
1908.
1909.

130.93
123.52
137.95
137.01
158.37

191 0
191 1
191 2
191 3
1914 (Jan. to
July) i .........
1914 (Aug. to
D ec.)1........
1915 (Jan.) i ...
1915 (Feb.) i . .

137.85
142.45
150.14
135.09

95.73

86.21

92.41
81.69
83.55

Kero­
Spir­
sene, Tin,
its,
70° Banca. grain,
Abel
re­
test.
fined.

48.85 93.48 162.67
57.05 79.11 143.82
94.33 80.0' 139.40
80.22 103.26 147 72
98.06 107.04 154.28

81.55 106,83
99.53 110.20
91.35 103.47

97 95 116.93
110.13 78.91
125 32 90.76
132.33 101.56
105.43 120.70

82.29 146.1' 124.36 123.24
89.24 134.81 102.52 131.12
90.39 119.57 104.93 104.69

100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00
78.96

98.52

97.84

81.23

97.46

71.68 102.81
53.35 96.78
58.54 88.01
57.22 66.22
73.70 54.63

64.41
68.77
57.68
61.18
65.21

160.32 71.72 82.16
144.91 67.54 81.38
153.92 68.78 81.05
163.48 79.84 114.00
197.93 136.77 115.43

69.81
70.84
96.31
69.38
69.59

67.74
63.16
65.91
52.62
57.40

66.58
55.82
44.44
53.82
65.37

218.67
177.73
176.96
188.14
192.65

85.18

58.93

75.

189.31 160.62 117.42

117.88 146.99 130.89 121.22 112.28 84.79
137.27 140.
130.53 129.43 103.67 104.31
131.85 125.78 120.30 116.84 116.03 108.85

65.14
68.30
75.50

63.04 215.06 192.18 138.35
220.02 150.38 120.02
68.
71.79 220.02 151.78 128.97

95.60
105.39
127.40
104.85
104.97

81.19

66.06

86.86
68.12

75.71
93.85

103.32 96.95 111.64 146.08
95.04 100.44 118.74 141.49
102.99 94.89 98.99 140.56
102.69 106.82 79.43 102.05
100.06 132.34 74.52 78.52

110.05 110.01 121.50

86.19

80.03

98.50

75.42 104.60

85.91

93.90 76.03 76.72 84.61
71.
73.50 74.18 90.84
106.46
71.66 81.49 71
84.61 101.82 110.30 99.26
85.12 108.32 109.57 94.64

79.72

149.32
131.33
135.09
141.91
141.88

68.23

115.62
104.75
103.87
110.35
128.09

ioi. o; 119.81 109.46 105.43 96.87 81.85 63.10 57.40 199.05 202.25 126.79

115.02
121.08
144.92
123.78

110.92 101.81 90.88 185.65 100.02 81.27
119.44 111.
100.33 204.33 85.39 91.60
133.68 161.93 121.05 170.08 112.30 100.51
121.12 148.19 126.13 120.47 96.98 88.90

143.78 130.24 118.60 128.83 131.04 121.60
208.94
200.00
202.58

127.63
160.99
174.62

83.33
82.99
80.93
62.88

217.49
205.77
235.57
243.46

172.92 128.61

212.20 111.85

234.53 181.38
227.23 150.88

83.14

78.07

62.46 243.46 183.40 123.40

149.04 113.40
191.25 124.7'
214.12 132.62

83.83
88.50
88.50

74.17 243.46 222.55 208.27
81.44 243.46 223.17 201.56
86.23 243.46 176.68 230.21

1 Maandschrift van het Centraal Bureau voor de Statistiek. The Hague, 1915, vol. 10 (March), p. 262.




INDEX NUMBERS— N E W ZEALAND.

295

NEW ZEALAND.
INDEX NUMBERS OF JAMES W. McILRAITH.
PUBLICATION.

This index, which is based on the wholesale prices of certain
important articles in New Zealand from 1861 to 1910, was published
by the Government of that Dominion in 1911 in “ The Course of
Prices in New Zealand,” by James W. Mcllraith. It is stated in the
introduction to the volume that the author intends to continue the
tables from year to year, the results to appear annually in the “ New
Zealand Official Yearbook.” 1
HISTORY.

The report is the result of a post-graduate research in economics at
Canterbury College. It has two main objects: (1) To measure the
changes in the general level of prices, year by year, since 1860; and
(2) to attempt to ascertain the causes of the changes in the local
price level. In the absence of any official index numbers for New
Zealand the study was undertaken in the hope that it would “ help
all engaged in the solution of those practical problems of social life
in which the changes in the purchasing power of money are an
important factor.” 2
SOURCE OF QUOTATIONS.

The author states that prices for all imported goods and for a few
colonial products have been obtained at Wellington, the geographical
and political center of New Zealand, while for all the cereals (including
flour) and for pastoral products he has used prices at Christchurch, a
city in the heart of the agricultural district of New Zealand. He
further states, however, that while it seemed inadvisable to use
Wellington prices for certain articles for certain years and then to use
the prices from some other city for the same articles for other years,
the nature of things demanded that the principle of continuity must
be subordinated to that of accessibility. The data contained in the
report were secured mainly from Wellington and Christchurch news­
papers: “ The New Zealand Trade Review and Price Current”
(Wellington), “ The Press” (Christchurch), and “ The Lyttleton
Times” (Christchurch) being the principal sources of information.
BASE PERIOD.

The base or standard period is the decade 1890-1899. The author
at first felt inclined to use 1867-1877, because this would have enabled
i An examination of the yearbooks for 1911, 1912, and 1913, however, fails to disclose any data bearing
on this subject.
^
« The Course of Prices in New Zealand, James W . Mcllraith, p. 3.




296

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

him to compare his figures more directly with those of Sauerbeck,
but he decided that the period was too early in the development of
New Zealand and that the data were insufficient. He found the
decade chosen a period of comparatively stable prices, the average
being almost identical with that for the 20 years 1886-1905.
PRICES: HOW SHOWN AND COMPUTED.

Wholesale prices for the articles upon which the index number is
based were taken quarterly, during the first week of January, April,
July, and October, or as near as possible to those dates. The simple
average of the four prices was taken as the average price for the year.
From these averages the simple average for 1890-1899 was com­
puted and was taken as the base, or 100, the prices for each of the
years included in the report being reduced to percentages of this
figure. For each year the percentages representing the prices of the
several articles were then added together and the result divided by
the number of articles. The resulting figure is the general index
number for the year.
NUMBER AND CLASS OF COMMODITIES.

* The index number from 1887 to 1910 is based on the prices of 45
commodities, both raw and manufactured. Previous to 1887, accord­
ing to the tables, the number of articles included was smaller and
varied from time to time, being lowest in 1861 and 1862, when only
33 articles were represented. Since 1875 the index number has in
every year been based on 41 or more commodities. The author state
that his aim has been to make the list include, as far as data were
available, the articles of the greatest importance in the trade of the
country. He considers his selection of commodities superior to the
selections of Sauerbeck and the Economist, because in his judg­
ment these indexes contain undue proportions of raw materials.
DESCRIPTION AND GROUPING OF COMMODITIES.

Concerning nearly all of the 45 articles the statement is made that
only the best grade is quoted. Further description of the articles, as
far as given, appears in the following list which classifies the 45
commodities under 8 heads:
Agricultural products (5 articles).

Wheat, best on Christchurch market.
Flour, best brands of New Zealand roller
flour.
Barley, best on Christchurch market.




Oats, best on Christchurch market.
Oatmeal, first-class New Zealand manu­
facture.

INDEX NUMBERS— N E W ZEALAND.

297

Pastoral products (8 articles).

Wool, best merino, greasy.1
Wool, best half-bred, greasy.1
Beef, best quality, Addington (Christ­
church) sales yard.
Mutton, carcasses exported.2

Lamb, carcasses exported.2
Bacon, best New Zealand product.
Cheese, best New Zealand product.
Butter, best New Zealand product.

Liquors (5 articles).

Beer (ale), Bass’s “ Dog’s Head.’
Whisky, Teacher’s (in bond).
Port, Superior (in bond).

Claret (in bond).
Brandy, Hennessy’s (in bond).

Beverages (3 articles).

Tea, Congou, fine.
Coffee, ground.

|Cocoa, Yan Houten’s.
Oils (3 articles).

Kerosene.
Linseed oil (boiled).

} Castor oil.
Minerals (6 articles).

Iron, galvanized,
gauge).
Iron, bar.
Lead, sheet.

4‘ Orb ’ ’

brand

(26

Materials

Cement, Portland.
Soap, New Zealand.
Matches, plaids.
Candles, Priced London Sperm.

Zinc.
Wire, black fencing, No. 8.
Coal, Newcastle (New South Wales), on
ship.
(7 articles).
Hops, Nelson (New Zealand).3
Soda, carbonate.
Soda crystals.

Other foods (8 articles).

Sugar, Auckland (New Zealand), refined
No. I.4
Currants.
Sultanas (raisins).
Rice.

Sago.
Salt, Liverpool, fine.
Salmon, in tins.
Pepper, white.

SUBSTITUTIONS AND ADDITIONS.

The substitution of one description of an article for another
description can not, except in a few cases, be traced in the report,
owing to the vagueness of the descriptive terms employed. It
appears quite probable, however, from the extensive period of time
covered by the tables that a considerable number of such substitu­
1 The quotations are prices of New Zealand wool at the London wool sales, minus freight from New Zea­
land, as reported by the London agents of H. Watson & Co., one of the principal wool-broking firms in
New Zealand.
2 Prices are deduced from estimated values of carcasses exported, as published in the “ Statistics of New
Zealand.”
s This kind is quoied in later decades; Kentish is quoted in earlier periods.
* Best imported sugar was quoted prior to the opening of the Auckland refinery.




298

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

tions must have been made. In a number of instances additions to
certain groups of commodities have been made since the initial year
of the series. In such cases the index numbers for the added articles
have been carried into the total of index numbers for the year and
this total divided by the number of articles to obtain the level of
prices for that year.
INTERPOLATION.

Much difficulty was experienced in securing data for the earlier
years covered by the report, and for years prior to 1875 there was an
occasional quotation lacking which the author deemed it necessary
to supply by an interpolated figure. He thus describes his method
of interpolation.
“ I examined the price of the particular commodity in preceding
and succeeding years. If those prices showed a continuous rise or
fall, that was strong evidence of the probability of the missing price
being one of an uninterrupted series. To test this probability, I
referred to similar products which would most likely vary in price
in the same manner as the commodity whose price was missing; and
if the price movement in these commodities was in the same direction
as the movement in the latter one, I presumed that the missing price
would most probably vary in the same manner as the price of the
similar articles in the same year. This method was applied chiefly
where there was a causal connection between the fluctuations in the
prices of the two articles, e. g . , where both were produced from the
same raw material (as galvanized iron and bar iron), or where one
was raw material and the other the finished product (as wheat and
flour), or where the production of both would be affected by the
same causes, as by drought (in the case of wheat and oats, mutton
and beef).” 1
All index numbers based on interpolated prices are inclosed in
brackets.
WEIGHTING.

The author does not attempt to assign definite weights to all
commodities. He has, however, quoted prices on more than one
variety of certain articles considered as of great importance. He
illustrates as follows: “ Thus, I have taken three cereals, and to
emphasize the importance of wheat I have taken flour as well. I
have sought to give wool adequate representation by including two
kinds, greasy merino and greasy half-bred. Iron is represented by
bar iron and galvanized iron, while meat is represented in a similar
manner by beef, mutton, lamb, and bacon.” 2
1The Course of Prices in New Zealand, James W. Mcllraith, p. 29.




2Idem, p. 28.

INDEX NUMBERS— N E W ZEALAND.

m

TESTING.

The tables of the report compare the index numbers derived by
the author with—
(1) A series of index numbers based on the median instead of the
simple arithmetic mean.
(2) The figures of the Economist and Sauerbeck reduced to the
New Zealand basis: Annual average prices 1890-1899 = 100.
(3) Index numbers indicating the local movements in certain
related phenomena, as, for example, the marriage rate, the bank­
ruptcy rate, world gold production, and New Zealand gold production.
The third set of comparisons is made by single years; the first and
second are made both by single years and by decades.
TABLE OF RESULTS.

The first eight tables of the report show for the several groups of
commodities and for each year the index numbers by articles, and
also the prices from which the index numbers were computed.
The next table summarizes the data contained in the preceding
tables by showing the index number for the total of all articles.
Other tables compare the New Zealand results with the figures
of Sauerbeck and the Economist (1) in their original form,
and (2) reduced to the New Zealand basis: Average annual prices
1890-1899 = 100.
Tables similar to these show index numbers for decades (e. g.,
1861-1870, 1862-1871, 1863-1872, etc.) instead of for single years.
In the two succeeding tables are shown the New Zealand index
numbers for farm products and for nonfarm products (1) by single
years, and (2) by decades.
The remaining tables show for New Zealand (1) an annual index
number of prices based on the volume of foreign trade and the tonnage
of vessels carrying such trade, (2) an annual index number of prices
based on the volume of foreign trade per head of population, (3) a
comparison of index numbers of prices (a ) of farm products and (6)
of all commodities with the index numbers of the marriage rate and
the bankruptcy rate, and (4) the index numbers (a ) of the gold pro­
duction of the world, (6) the gold production of New Zealand, and (e)
the price of wheat. The last table also gives the New Zealand wheat
yield per acre, in bushels.
The table showing the comparison of the index numbers for New
Zealand derived by the author with those of Sauerbeck and the
Economist reduced to the same base period, and also with index
numbers based on the median, instead of the arithmetic mean, is
reproduced herewith.




300

BULLETIN OP THE BUREAU OP LABOR STATISTICS.

COMPARISONS OF INDEX NUMBERS COMPUTED UPON THE SAME BASE PERIOD.

Year.

1Econo­
mist/

184
186
193
195
189

133
141
171
185
175
174
149
131
130
131
127
139
144
140
135
132
132
124
109
124
116
119
114
109

=

1861.
1862.
1863.
1864.
1865.
1866.
1867.
1868.
1869.
1870 .

1871.
1872.
1873.
1874.
1875.
1876.
1877.
1878.
1879.
1880.
1881.
1882.
1883.
1884.
1885.

Sauer­ New Zea­
beck,
land
1890-1899
(by
= 100.0.
median).
= 100.0.

New Zea­
land,
1890-1899
100.0.

200

187
184
164
154
150
154
164
161
148
140
144
135
127
130
125
123
118
115

111

102

149
153
156
159
153
155
152
150
149
146
152
165
168
155
146
144
142
132
126
133
129
127
124
115
109

172
172
176
185
183
187
192
189
163
148
137
154
152
160
148
134
148
133
125
129
125
122

118
113
110

Year.

1887,
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909,
1910

New Zea­ “ Econo­ Sauer­ New Zea­
mist,”
land
beck,
land
1890-1899 1890-1899
(by
= 100.0.
= 100.0.
= 100.0. median).
108
103
103

101

111

109
106

100

104
103

107
108
104
98
93
96
97
97
98

110
110

102

101

94
98
95
93
94
104
104
96
98
106
108
116

107
104

109

101
98
100
100

95
98

101

103

121

110

105
103
106
109
109
109
103
96
96
94
92
94
94
103
114
106
105
105
106
109
116
121

111

112

118

107
102

104

111

108
108
102
100
100

96
98
100
100
95
99
100
101
101
104
101

The author has this to say concerning the foregoing table: “ A
general review of this table shows a marked similarity in the move­
ments of columns 1 and 3. The fluctuations in both columns are
parallel, i. e., a rise or fall in both is synchronous. The index num­
bers are fairly even in both columns, and particularly so since 1872,
the numbers in column 4—those calculated by the median—being on
the whole slightly lower than those in column 1, thus indicating that
exceptional variations have been of the nature of a rise more often
than of a fall.”
NORWAY.
INDEX NUMBERS OF EINAR RUUD.
PUBLICATION AND HISTORY.

This index of wholesale prices, based on the prices of imports into
Norway, was first published in 1911 in the official journal of the Nor­
wegian Labor Office,1 No. 9-10 of that year. It covers a period from
1880 to 1910, but whether or not it has since been brought down to
date is not known, although the prices on which it is based continue
to be published in the annual summary of commercial statistics
issued by the Norwegian office of general statistics.2
1 Sociale meddelelser (fortsaettelse av maanedsskrift for socialstatistik) utgit av Socialavdelingen under
Departementet for sociale saker, handel, industri og fiskeri. Christiania, 1911-1915. lste aargang, Nr. 9-10,
1911, pp. 136-149.
2 Norges handel, 1908- (Statistique du commerce de la Norvege pendant l’annee 1908-) Utgit av det Statistiske centralbureau. (Norges officielle statistik, V , 87,116.)




INDEX NUMBERS---- NORWAY.

301

SOURCE OF QUOTATIONS.

The summary of commercial and customs statistics alluded to pre­
sents the average annual prices of some 135 different commodities
imported into the Kingdom each year, and from this list the compiler
of the index has selected 39 articles for inclusion in his series. The
prices of these articles are obtained from a number of manufacturers
and wholesalers in different parts of the Kingdom.
BASE PERIOD.

The base period chosen extends from 1891 to 1900. This period
was selected, it is explained, because it contained both a rising and
a falling tendency of prices, not only in Norway but also in foreign
countries, and therefore, when taken as a whole, may be regarded as
representing average market conditions.
NUMBER AND CLASS OP COMMODITIES.

The 39 articles chosen from the larger list of 135 imports are divided
into five groups: (1) Food commodities; (2) grains and flours; (3)
imports from the East; (4) manufactured products; (5) miscellaneous.
Group I includes beef, pork, cheese, oleomargarine, eggs, and pota­
toes; Group II, barley, oats, wheat, rye, also hulled and prepared bar­
ley and oats, and flour or meal ground from the above grains; Group
III includes coffee, tea, sugar (two kinds, loaf and granulated),
tobacco (smoking and chewing), and rice; Group IV is made up of
wool yarn (plain and dyed), cotton yarn (single and several twist),
cotton cloth or goods (three kinds, printed, dyed and bleached, and
unbleached); and Group Y includes petroleum, coal and coke, dressed
skins or leather, pig iron, steel, bar and hoop iron, zinc, lead, and tin.
W EIGHTING.

In the construction of the index a simple arithmetic average is
employed, there being no weighting. The total index is the mean of
the indexes for each of the five groups.
TABLE OF RESULTS.

The following table, which appears in the journal of the Norwegian
Labor Office for September and October, 1911 (No. 9-10)1 shows in
detail the main results of this series of index numbers, as published.
l Maanedsskrift for socialstatistik. Utgit ay det Statistiske Centralbyraa, lste aargang, Nr. 9-10,1911,
pp. 141-148.




302

BULLETIN OF TH E BUREAU OF LABOB STATISTICS.
IN D E X NUMBEKS OF W HOLESALE PRICES.

(Based on 7 food commodities imported into Norway,
1891-1900 *=100.)

Year.

Beef.

141
138
151
154
136
121
103
92
90
90
90
95
92
100
95
95
92
95
105
108
113
113
118
110
103
115
128
133
123
123
131

1880...........................................
1881...........................................
1882...........................................
1883...........................................
1884...........................................
1885...........................................
1886...........................................
1887...........................................
1888...........................................
1889...........................................
1890...........................................

1895...........................................
1897...........................................
1899...........................................
1901...........................................
1902...........................................
1903...........................................

1909...........................................

Pork.

Cheese.

103
113
98
94
96
87
87
96
96
100
104
109
104
104
96
91
96
96
100
100
100
100
104
104
104
104
113
122
113
113
113

124
129
149
134
112
102
98
110
122
107
97
98
112
136
115
95
75
76
93
86
115
131
153
132
114
119
136
136
132
166
180

1880-1910.

Oleo­
Butter. marga­
rine.

98
98
98
105
101
105
98
97
91
91
105
105
105
105
105
105
105
108
105
108
115
115
129

Base period,

Eggs.

Pota­
toes.

71
86
97
97
97
97
86
86
91
91
91
91
97
97
97
97
102
102
102
102
108
108
108
108
108
108
108
108
108
108
108

97
97
150
140
133
117
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
83
83
83
92
88
72
88
111
97
83

96
101
101
112
112
112
107
96
90
84
92
96
96
96
96
96
96
96
101
101
112
112
112

(Based on 10 grains and flours imported into Norway, 1880-1910.
1891-1900=100.)

Year.

1881.......................................
1882.......................................
1883.......................................
1884.......................................
1885.......................................
1887.......................................
1888.......................................
1889.......................................
1890.......................................
1891.......................................
1892.......................................
1893.......................................
1895.......................................
1896.......................................
1897.......................................
1898.......................................
1899.......................................
1900.......................................
1901.......................................
1902.......................................
1903.......................................
1904.......................................
1905.......................................
1906.......................................
1907.......................................
1908.......................................
1909.......................................
1910.......................................




All 7
com­
modi­
ties.
107
115
129
124
115
105
95
97
99
98
97
101
103
108
101
96
92
92
100
100
105
105
110
105
103
105
109
114
116
119
122

Base period,

Pot
All
or Hulled Bar­ R ye Oat­ Wheat 10
Bar­ Oats. Wheat. Rye. pearl
ley flour. meal. flour. com­
oats.
ley.
bar­
flour.
mod­
ley.
ities.
153
157
137
134
132
119
106
97
102
109
114
134
109
94
78
86
87
87
101
109
115
108
108
98
97
105
109
131
134
126
111

126
136
126
110
105
105
99
84
89
110
99
115
115
105
89
76
89
103
105
102
102
107
107
97
105
105
112
127
126
112
105

156
163
140
156
117
111
107
1C6
109
113
113
132
115
97
78
82
89
105
105
100
96
96
97
97
109
104
102
118
125
128
111

164
164
130
124
115
104
92
84
84
94
102
148
126
96
79
78
86
86
102
102
97
93
94
93
93
103
100
128
128
121
100

166
139
133
146
133
126
123
113
119
119
119
136
123
96
80
83
90
88
96
105
105
97
96
94
86
87
93
108
103
101
91

124
129
112
112
97
135
107
79
84
90
96
112
132
120
88
87
87
87
96
98
93
93
112
105
105
104
110
121
122
121
106

146
155
146
138
130
130
114
98
106
114
114
142
118
87
83
89
85
85
103
107
99
98
102
98
98
100
106
138
130
118
106

141
165
153
143
129
114
127
103
90
101
109
146
135
97
81
82
71
71
103
111
103
96
101
96
95
101
101
132
125
116
92

108
114
108
102
95
95
92
83
86
98
105
111
120
108
95
89
95
95
95
95
* 95
95
108
102
102
210
108
117
117
109
102

156
167
144
139
121
110
110
107
111
108
113
129
104
95
81
85
90
110
107
103
96
92
95
97
107
100
96
112
115
127
117

144
149
133
130
117
115
108
95
98
106
108
131
120
100
83
84
87
92
101
103
100
98
102
98
100
101
104
123
123
118
104

303

INDEX NUMBERS— NORWAY.
IN D E X NUMBERS OF W HOLESALE PRICES—Continued.
(Based on 6 commodities imported into Norway from the East, 1880-1910.
period, 1891-1900 = 100.)
Sugar.
Year.

Coffee.

Tea.

96
82
70
70
72
64
74
111
100
117
129
117
117
125
121
122
112
82
70
61
69
63
54
51
56
63
61
52
58
61
70

1889..........................................
1881..........................................
1882..........................................
1883..........................................
1884..........................................
1885..........................................
1886..........................................
1887..........................................
1888..........................................
1889..........................................
1890..........................................
1891..........................................
1892..........................................
1894..........................................

1898..........................................
1900..........................................
1902..........................................

1910..........................................

138
146
131
151
131
126
121
116
106
106
106
106
106
106
106
101
98
95
95
95
94
93
95
95
95
95
95
101
103
101
103

Tobacco.

Other.

Loaf.
187
194
187
171
139
123
113
113
119
126
116
113
116
119
103
94
94
84
84
87
90
84
74
77
90
97
81
81
87
90
103

177
192
173
162
123
119
108
108
116
127
119
115
123
123
108
92
88
81
85
89
92
81
69
73
89
100
81
81
89
92
108

80
80
80
76
107
107
107
107
107
107
103
103
103
99
99
99
99
99
99
99
99
99
99
99
96
96
92
92
88
92
96

Base

All 6 com­
modities.

Rice.
148
119
105
95
95
90
90
90
90
95
100
105
100
138
90
86
90
95
100
100
100
100
95
100
100
100
100
110
100
105
105

138
136
124
121
111
105
102
108
106
113
112
110
111
118
105
99
97
89
89
89
91
87
81
83
88
92
85
86
88
90
98

(Based on wool and cotton yarns and cotton goods (7 articles) imported into Norway,
1880-1910. Base period, 1891-1900= 100.)
t Cotton yam, un­
bleached.

Wool yam.
Year.

1880.....................................
1881
..............................
1882 .................................
1883 ...................................
1884
............................
1885 .................................
1886.....................................
1887.....................................
1888.....................................
1889.....................................
1890.....................................
1891.....................................
1892.....................................
1893.....................................
1894.....................................
1895.....................................
1896.....................................
1897.....................................
1898....................................
1899....................................
1900.....................................
1901.....................................
1902.....................................
1903.....................................
1904.....................................
1905.....................................
1906.....................................
1907....................................
1908.....................................
1909....................................
1910...................................




Undyed
Dyed,
and un­ mixed,
bleached.
etc.
144
144
144
132
128
121
122
122
122
115
115
109
106
106
103
103
101
86
86
92
106
101
101
93
95
102
115
115
101
109
118

148
148
148
140
132
119
120
120
119
114
112
106
103
106
101
101
98
93
93
95
106
103
103
111
111
116
132
138
116
122
127

Single
thread.

97
93
93
117
109
109
121
121
113
125
137
133
113
144

Multi­
ple
twist.

100
93
93
113
120
117
127
133
120
133
143
133
133
147

Cotton goods.

All 7
com­
modi­
Print­ Bleached
Un­
or dyed. bleached. ties.
ed.
142
142
120
120
120
114
114
108
108
108
107
101
98
97
91
99
99
103
99
103
108
103
103
108
114
117
121
125
125
124
125

134
134
118
118
118
118
110
104
106
108
108
102
100
100
94
94
94
98
98
102
112
102
102
110
114
122
124
128
128
132
134

128
128
117
117
114
110
103
107
110
114
114
107
103
103
99
99
99
96
92
92
110
99
99
114
124
135
142
149
142
142
145

139
139
129
125
122
116
114
112
113
112
111
105
102
102
98
99
98
96
93
96
110
105
105
112
116
118
127
134
125
125
134

304

BULLETIN OF THE BUKEAU OF LABOR STATISTICS.
IN D E X NUMBERS OF W HOLESALE PRICES—Concluded.

(Based on 6 metals, and leather, fuels, and petroleum imported into Norway, 1880-1910.
Base period, 1891-1900=100 .)

Year.

1880...........................................
1881...........................................
1882...........................................
1883...........................................
1884...........................................
1885...........................................
1886...........................................
1887...........................................
1888...........................................
1889...........................................
1890...........................................
1891...........................................
1892...........................................
1893...........................................
1894...........................................
1895...........................................
1896...........................................
1897...........................................
1898...........................................
1899...........................................
1900...........................................
1901...........................................
1902...........................................
1903...........................................
1904...........................................
1905...........................................
1906...........................................
1907...........................................
1908...........................................
1909...........................................
1910...........................................

Leath­ Petro­ Pig
leum. iron.
er.
157
141
140
134
122
116
113
107
101
98
101
95
92
92
89
107
101
101
104
110
110
113
110
110
110
110
116
116
104
116
128

188
150
131
131
131
131
122
113
127
127
122
113
103
94
94
122
113
89
85
94
94
85
75
75
75
69
77
77
86
81
75

Steel.

Bar
and
hoop
iron.

Zinc.

Lead.

148
157
157
144
135
135
126
117
112
121
126
117
99
94
90
90
94
94
99
103
121
103
94
90
85
81
82
82
72
68
72

112
110
110
102
99
99
93
90
89
101
106
99
92
91
86
86
91
95
99
118
142
105
101
96
95
95
103
107
89
86
94

103
77
72
66
72
72
80
86
100
114
114
100
86
86
74
86
86
114
129
114
100
103
100
114
129
143
129
109
114
114

144
108
104
104
80
80
88
96
104
104
108
100
100
100
96
92
88
92
96
104
128
100
92
92
92
100
132
144
108
104
104

109
100
95
87
78
75
71
75
77
96
103
98
89
86
87
87
84
90
99
125
154
104
104
104
96
107
112
120
107
104
104

All 9
Coal, com­
coke, modi­
etc.
ties.

Tin.

101
101
101
91
91
101
126
120
107
116
110
110
107
88
82
79
77
91
113
145
126
142
142
142
160
195
189
157
157
173

86
78
78
85
87
91
93
95
90
90
103
98
94
98
94
86
81
87
96
111
158
150
107
96
90
83
87
106
94
88
86

135
116
110
107
99
99
98
100
101
105
111
105
98
94
90
92
91
90
98
112
130
110
103
101
100
104
116
119
103
102
106

(Based on 39 articles imported into Norway, 1880-1910, fry groups o f commodities .
tfase period, 1891-1900 = 100.)

Year.

1880...........................................
1881...........................................
1882...........................................
1883...........................................
1884...........................................
1885...........................................
1886...........................................
1887...........................................
1888...........................................
1889...........................................
1890...........................................
1891...........................................
1892...........................................
1893...........................................
1894...........................................
1895...........................................
1897...........................................
1898...........................................
1899...........................................
1900...........................................
1901...........................................
1903...........................................
1904...........................................
1905....................... ...................
1908...........................................
1909...........................................
1910.......... ...............................




I.
Food
products.

107
115
129
124
115
105
95
97
99
98
97
101
103
108
101
96
92
92
100
100
105
105
110
105
103
105
109
114
116
119
122

II.
Flour
and grain.

144
149
133
130
117
115
108
95
98
106
108
131
120
100
83
84
87
92
101
103
100
98
102
98
100
101
104
123
123
118
104

III.
Articles
imported
from the
East.
138
136
124
121
111
105
102
108
106
113
112
110
111
118
105
99
97
89
89
89
91
87
81
83
88
92
85
86
88
90
98

IV.
Manufac­
V.
tured arti­ Miscellane­
cles (cot­ ous (metals All com­
modities.
ton and
princi­
woolen
pally).
yam and
goods).
139
139
129
125
122
116
114
112
113
112
ill
105
102
102
98
99
98
96
93
96
110
105
105
112
116
118
127
134
125
125
134

135
116
110
107
99*
99
98
100
101
105
111
105
98
94
90
92
91
90
98
112
130
110
103
101
100
104
116
119
103
102
106

133
131
125
121
113
108
103
102
103
107
108
110
107
104
95
94
93
92
96
100
107
101
100
100
101
104
108
115
111
111
113

INDEX NUMBERS---- RUSSIA.

305

RUSSIA.
INDEX NUMBERS OF MINISTRY OF COMMERCE AND INDUSTRY,
PETROGRAD.
PUBLICATION.

The Ministry of Commerce and Industry of Russia pubhshes annu­
ally a “ Summary of Prices for Commodities in Representative Rus­
sian and Foreign Markets,” in which is included an index number
based on the average annual prices of the various articles under con­
sideration. Wholesale prices are used in this summary.
HISTORY.

The history of the index number shown in connection with the
prices can not be learned from the translations of the reports at hand.
SOURCE OF QUOTATIONS.

Quotations of prices on Russian and foreign exchanges as printed
in the bulletins of these exchanges are used as the original material
for compilation. In the absence of exchange quotations for some
of the articles, information has been secured from special periodicals
such as: “ Commerce and Industry,” “ Baltische Wochenschrift,”
“ Iron and Coal Trades Review,” etc. Prices for cereals in some
markets, as well as data on freight charges and insurance premiums
on cereal freight, are taken from the reports of the Bureau of Com­
merce in Cereals of the Department of Commerce which compiled
them from telegrams of its special agents. They have been supple­
mented by quotations from the bulletins of local exchanges, and are
to be found on the last pages of the 1908 report. Prices of dutiable
foreign articles on foreign markets are given without the inclusion
of the Russian customs duties; those articles the prices of which in­
clude customs duties are marked with a star. Prices on Russian
markets are always given including custom duties.
BASE PERIOD.

The average price for the ten-year period, 1890-1899, taken as
100, is used as a base.
PRICES: HOW SHOWN AND COMPUTED.

The following is a translation of the introduction to the summary
of 1908:
The present bulletin (1908) of wholesale prices contains data relat­
ing to the principal commodities in Russian and foreign markets
and is compiled in the same way as the preceding issues. In addition
to detailed prices for each month of the present year, the average
prices for each of the preceding years— 1907,1906,1905—aro given,
as well as the general average prices for the 5 years, 1900-1904, and
the 10 years, 1890-1899.
j)4261°—Bull. 173—15----- 20



306

B U L L E T IN

OF T H E B U R E A U OF LABOR STATISTICS.

In the summary table (see pp. ii-vii) average annual prices of
the principal commodities in Russian markets for each year of a period
of 19 years, 1890-1908, are shown. For each article included m this
table the prices in one or two of the representative markets furnishing
sufficiently accurate data are given. In order to minimize the influ­
ence of local conditions on the prices of the principal cereals, general
average prices are quoted for several markets combined in the follow­
ing three groups:
1. Markets in northern ports (Petrograd, Reval, Riga, and Libau).
2. Markets in southern ports (Odessa, Nicolaie, Taganrog, Rostov
on Don, and Novorossysk).
3. Markets in central Russia (Moscow, Yelts, Samara, and Saratov).
For purposes of comparison of the average prices for the years
1890-1895 with those of the following years, the former have been
converted from paper into gold values (1 ruble equals 1-15 im­
perial), taking as a base the average quotations for the correspond­
ing years as follows:
Rubles in gold.

100 rubles in paper money equals...............

1890

1891

108.8G

100.14

1892
94.58

1893
97.91

1894
100.53

1895
101.22

NUMBER AND CLASS OF COMMODITIES.

Prices and index numbers were reported in 1912 for 66 commodities.
Wholesale prices are used in the computation of the tables for all
years. Both raw and manufactured articles are included.
DESCRIPTION AND GROUPING OF COMMODITIES.

The 66 articles for which prices were secured in 1912 are combined
in the following 7 groups or classes:
1. Cereals and by-products.
2. Cattle and cattle products.
3. Oils.
4. Spinning materials.
5. Minerals (including petroleum, etc.).
6. Drugs and chemicals.
7. Groceries.
The commodities quoted are as follows:
1. Cereals and by-products.
Articles.

Markets of—

Bye...... .......................Northern ports and southern ports.
Wheat.......................... Riga (Russian), southern ports, and central Russia.
Oats..............................Northern ports, southern ports, and central Russia,
Barley.......................... Southern ports.
Com............................ Odessa,




307

INDEX NUMBERS---- RUSSIA.
Articles.

Markets of—

Peas............................. Saratov and Rybinsk.
Buckwheat...................Y elts.
Buckwheat groats........ Moscow, large groats.
Millet........................... Moscow, from Orenburg and Ural.
Wheat flour.................. Moscow (highest quality in Saratov).
Rye flour......................Moscow (sifted).
Bran............................. Libau, of wheat, medium; Rybinsk, of wheat.
Malt............................. Moscow, imported.
2. Cattle and animal products.

Large homed cattle___ Petrograd, highest quality.
Small cattle................. Petrograd, calves, hogs.
Meat............................. Petrograd, best quality, beef; Moscow, pork.
Hides, steer.................. Moscow, gray.
Lard.............................Petrograd, for gruel.
Butter...........................Moscow, fresh made of warmed cream.
Herring........................ Riga, Scotch.
S. Oils.

Flaxseed...................... Petrograd, 95 per cent.
Hempseed.................... Orel.
Sunflower seed............. Saratov, for oil.
Linseed oil................... Moscow.
Hempseed oil...............Petrograd.
Sunflower-seed oil........ Moscow.
Rapeseed oil.................Warsaw, raw.
Olive oil....................... Petrograd.
Oil cake, flax............... Riga.
Oil cake, hemp............ Riga.
4 . Spinning materials.

Flax............................. Riga, Livonian; Petrograd, from Vologda.
Hemp........................... Riga; Orel, pure.
Cotton..................... .. .Moscow, from Fergan; of American seeds.
Cotton yam.................. Petrograd, spool, No. 14-20.
Coarse calico.................Petrograd, 16 vershok 1 wide by 585 arshin 2 long, 80 funts
in weight.
Wool............................ Moscow, medium quality.
S. Minerals.

Coal..............................Petrograd, from Newcastle; Rostov on Don, anthracite.
Naphtha....................... Baker, near wells; Moscow.
Crude naphtha............. Baker, near wells; Moscow.
Petroleum.................... Baker, on ships and in kettles; Moscow.
Cast iron.......................Petrograd, southern.
Iron..............................Petrograd, sheet iron, Russian bar.
Copper..........................Petrograd, foreign, in bars.
Tin............................... Petrograd, in bars.
Zinc............................. Petrograd, Silesian.
Lead.............................Petrograd, ordinary ingots.
i Xvershok equals 1.75 inches,




21

arshin equals 0.77 yard.

3 1 funt equals 0.9 pound.

308

BULLETIN OP THE BUBEAU OF LABOR STATISTICS.
6. Drugs and chemicals.

Articles.
Markets of—
White lead................... Petrograd.
Indigo.......................... Moscow, from Bengal.
Dry paints....................Moscow, blue.
White resin.................. Petrograd.
Tragacanth...................Moscow, highest quality.
Borax........................... Petrograd, in crystals.
Vitriol.......................... Petrograd.
Potash.......................... Petrograd.
Soda (caustic).............. Petrograd.
Saltpeter...................... Petrograd, from Chile.
Sulphur........................ Petrograd, in lumps.
7. Groceries.

Salt...............................Rybinsk.
Granulated sugar..........Kiev, in territory of the Southwestern R. R.
Lump sugar, refined .. .Kiev, lump.
Coffee........................... Petrograd, round, various qualities.
Tea............................... Moscow, from Kyakhta.
Rice............................. Odessa, Patua, highest quality.
Pepper..........................Petrograd, black, from Singapore.
Currants........................Riga.
Almonds.......................Odessa, from Messina.
Hops.............................Riga, from Bavaria.
SUBSTITUTIONS AND ADDITIONS.

No substitutions or additions have been made so far as the reports
disclose, but certain commodities reported in 1908 were dropped in
subsequent years. This changes the index number for the group
affected and also the general index number.
Alcohol and woolen yarn were dropped in 1909. Tea in 1909-1912,
inclusive, is reported in a different market and with a different unit
of measure from that used in 1908.
In the group “ Cat tie and animal products” lard is given an average
price of 6.90 for the year 1909 in the report for that year, while in all
subsequent reports it is quoted at 6.60 for 1909. In the 1912 report
new index numbers are published for the group embracing “ Cereals
and by-products” and for the “ Groceries” group, also for all groups
combined, owing to the substitution in some cases of other markets
for those carried in reports for preceding years.
INTERPOLATION.

If any interpolations of prices have been made, they are not called
to the attention of the reader.
WEIGHTING.

Whatever weighting there is consists in the use of a number of
different descriptions of the same commodity.




309

INDEX NUMBERS— :SPAIN.
TABLE OF RESULTS.

Index numbers for each of the 7 groups into which the total num­
ber of commodities is divided, together with a general index for the
66 commodities taken as a whole, are shown in the following table
compiled from the 1912 report:
IN DEX NUMBERS FOR 7 GROUPS OF COMMODITIES (66 ARTICLES) AND GENERAL
IN DEX FOR A L L COMMODITIES.

Year.

Cattle
Cereals
and
and by­
cattle
products. products.

1890-1899....................
1890.............................
1891.............................
1892............................
1893.............................
1894.............................
1895.............................
1896.............................
1897............................
1898............................
1899.............................
1900-19C9....................
1900.............................
1901............................
1902............................
1903............................
1904.............................
1905.............................
1906............................
1907............................
1908............................
1909............................
1910............................
1911.............................
1912.............................

100.0
108.2
128.8
126.4
103.1
78.6
76.2
77.3
87.2
106.3
107.3
119.9
98.7
106.8
112.8
102.0
106.3
117.0
122.7
146.7
147.4
138.5
117.8
128.0
145.7

100.0
99.3
93.9
96.7
104.6
102.5
98.7
94.6
94.9
104.0
112.0
131.4
117.8
115.3
121.8
121.9
121.7
127.8
137.3
148.4
148.4
152.5
154.6
143.9
153.7

Oils.

100.0
102.6
108.3
104.5
109.3
103.8
86.1
80.1
92.7
110.4
102.2
111.1
108.0
132.1
113.3
94.3
96.7
102.0
115.4
109.5
106.1
130.9
140.7
134.1
126.8

Spinning
materials. Minerals.

100.0
100.9
93.7
94.1
108.0
105.1
99.4
101.4
98.9
96.1
100.3
133.0
124.5
131.0
124.8
130.0
134.1
127.0
139.8
146.4
134.7
138.6
154.3
151.9
152.2

100.0
109.8
99.3
89.6
96.1
91.0
93.9
96.1
98.7
102.7
119.0
118.7
131.1
109.1
96.5
98.2
106.9
119.0
137.9
141.8
124.4
121.7
116.7
129.9
155.6

Drugs
ana
chem­
icals.
100.0
103:6
100.8
102.7
106.5
100.6
97.9
96.6
96.4
96.0
99.4
107.2
103.2
101.5
99.4
98.9
102.7
105.3
112.7
121.0
115.8
111.6
113.6
117.9
122.4

General
for
Groceries. index
all com­
modities.
100.0
113.3
105.5
97.5
104.7
97.1
91.8
92.4
95.2
99.7
103.3
105.4
103.8
108.0
102.6
104.5
108.7
108.2
108.4
106.7
102.4
101.1
104.8
110.4
117.1

100.0
105.4
104.2
101.6
104.6
97.0
92.0
91.2
94.9
102.2
106.2
118.1
112.4
114.8
110.2
107.1
111.0
115.2
124.9
131.5
125.6
127.8
128.9
130.9
139.1

SPAIN.
INDEX NUMBERS OF FRANCISCO BERNIS.
PUBLICATION.

A series of index numbers based on food prices in Spain is found in
El Problema de las Subsistencias (The Problem of the Food Supply),
by Francisco Bernis, professor of political economy in the University
of Salamanca, which was published in pamphlet form in 1911. The
volume is divided into two parts, the first of which refers to prices of
food commodities in general uso, and the second to laws and regula­
tions in regard to food commodities, their distribution and prices.
A general study of the variations of prices is included in the first
part, but the information does not show whether wholesale or retail
prices are considered.
HISTORY.

The author of El Problema de las Subsistencias made a survey of
the various attempts of the State to deal with the subject of the
increased cost of living as it affected the working classes. The results




310

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

of this survey were published at the request of the Asociaci6n de
Patronos Mineros de Vizcaya, in order that this association might
make a study of the strike in 1910 of the miners of that locality. The
tables of prices and of index numbers contained material that was
considered sufficiently comprehensive to be used as a basis for the
work undertaken by the association.
SOURCE OF QUOTATIONS.

The data on which Bernis based his tables were taken from un­
published records of the chief statistician of Barcelona, and, in the
absence of information of an official character, from the works of
college directors and others who have made a study of the subject.
Some specific sources of information are as follows:
The Asociaci6n de Patronos Mineros de Vizcaya published in 1907
a pamphlet in which appears a table of comparative prices of food­
stuffs* for 1903-1907. The document is entitled “ El trabajo en las
minas de Vizcaya.”
El Instituto de Reformas Sociales issued a bulletin in 1904 rela­
tive to labor conditions in the mines of Vizcaya, which gives a table
of comparative prices for 1893-1903.
The Camara Oficiai de Comerco of Madrid, in its report entitled
“ Informacion publica sobre el problema de las subsistencias,” 1905,
presents a study of variations of prices for several years.
“ El presupuesto de reconstrucci6n,” by Garcia Alix, gives under
the title of “ Subsistencias” average prices for Spain.
BASE PERIOD.

The year 1901, taken as 100, is used as the base period.
PRICES: H O W SHOW N AND COMPUTED.

Prices are shown for yearly periods, and cover the years 1891 to
1908, inclusive. The compiler states that “ the prices for forming my
indexes are wholly those which I possess, though these had to be
corrected and completed.” 1 At times market prices were used when
there were no. others available.
NUMBER AND CLASS OF COMMODITIES.

The table of prices includes 13 articles of food, as follows: Wheat,
potatoes, bread, rice, pulse (chick-peas), kidney beans, beef, mutton,
pork, codfish, sugar, salt, and coffee.
DESCRIPTION AND GROUPING OF COMMODITIES.

There is no separation of the commodities into groups except in the
case of meats, of which three kinds are listed. No description of the
articles is given.




i El Problema de las Subsistencias, Francisco Bernis, p. 57.

311

INDEX NUMBERS---- SPAIN*
WEIGHTING.

Two index numbers are computed for all articles combined, one
unweighted and the other weighted.
The weighting is accomplished by dividing the commodities into
four groups, each of which represents a certain percentage of the
whole number as regards importance in consumption. The four
groups are as follows:1
Per cent.

I.
II.
III.
IV.

Wheat, potatoes, bread.................................................................30
Rice, pulse (cliick-peas), kidney beans.......................................30
Beef, mutton, pork, codfish..........................................................20
Sugar, coffee, salt...................................................................... ...10

The importance assigned to each of these four groups is based on
an estimate of the relative values of the articles of food as consumed
in the family of a laborer.
TESTING.

The only test made of the index numbers is their comparison with
index numbers for the United Kingdom, Germany, and France.
Diagrams are added to show the variations of these index numbers
during the same period of time.
TABLES OF RESULTS.

The following table shows the variations in the weighted relative
prices of the different food articles in Spain for the years from 1891
to 1908, inclusive. Weighted and unweighted price relatives for all
commodities combined are also shown in the last two columns:
VARIATIONS IN W EIGHTED RELATIVE PRICES OF FOOD COMMODITIES IN SPAIN,
1891 TO 1908, B Y YEARS.2
(Base period, 1901=100.)

Year.

1891.............................
1893.............................
1894.............................
1895.............................
1896.............................
1897.............................
1898.............................
1899.............................
1900.............................
1901.............................
1902.............................
1903.............................
1904.............................
1905.............................
1906.............................
1907.............................
1908.............................

Potatoes. Wheat.

85.2
77.8
90.7
89.5
85.0
92.9
94.5
102.0
102.0
99.3
100.0
99.8
122.5
111.8
131.7
111.6
115.0
119.8

Bread.

82.2
100.0
96.3
120.0
1892.............................
92.2
101.0
79.2
104.6
97.0
73.1
94.8
94.8
97.6
106.8
116.4
103.3
108.4
104.4
101.8
101.9
100.0
100.0
99.6
91.9
97.2
97.0
106.4
109.4
112.3
96.8
97.2
98.3
98.1
96.3
108.5
100.6

Rice.

101.5
102.3
102.1
99.7
99.9
102.5
108.6
102.9
102.1
100.0
100.0
107.6
110.1
112.2
111.9
109.3
105.8
105.5

Pulse
(chick­
peas).

Kidney
beans.

89.4
76.5
84.6
73.9
85.7
95.3
98.2
96.0
96.0
91.8
100.0
98.6
96.5
103.5
109.1
118.7
115.2
104.3

85.8
85.3
89.3
90.1
85.9
88.7
91.9
95.0
94.6
94.3
100.0
104.0
99.7
105.7
105.5
114.6
108.9
83.1

Beef.

86.1
90.9
86.4
82.7
93.9
94.8
91.9
92.4
97.2
100.0
100.0
103.7
95.1
102.4
100.7
100.7
100.7
100.7

Mutton.

92,3
96.1
93.2
94.4
76.4
89.9
93.4
89.1
95.6
97.6
100.0
110.8
93.1
114.5
106.2
106.2
106.2
103.9

1 It will be noted that the figures used to represent the groups show a total of only 90. Evidently the re­
maining 10 per cent includes commodities not in general use, and for this reason not included in the study,
2El Problema de las Subsistencias, Francisco Bemis, p. 58.




312

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

VARIATIONS IN W EIGHTED R E LATIVE PRICES OF FOOD COMMODITIES IN SPAIN,
1S91 TO 1908, B Y YEARS—Concluded.

All com modities.
Year.

1891............................
1892............................
1893............................
1894............................
189."*............................
189G............................
1897............................
1898............................
1899............................
1900............................
1901............................
1902............................
1903............................
1904............................
1905............................
1900............................
1907............................
1908............................

Pork.

i

90.9
96.3
89.5
89.5
91.0
83.5
91.3
100.2
101.9
99.5
100.0
100.4
91.4
104.5
101.4
101.4
96.9
96.9

Codfish.

83.5
80.5
72.2
74.4
74. 1
78.9
79.7
70.4
91.0
100.0
100.0
105.3
103.8
113.5
111. 5
103.0
87.9
101.5

Sugar.

83.3
83.3
83.3
83.3
78.9
87.7
83.3
87.7
95.6
99.1
100.0
96.5
100.0
104.4
107.0
92.1
96.5
117.5

Salt.

75.0
75.0
75.0
62.5
62.5
62.5
62.5
62.5
87.5
87.5
100.0
100.0
100.0
100.0
87.5
87.5
87.5
75.0

Coffee.

99.0
108.9
96.0
98.6
98.6
98.6
101.0
105.0
102.0
102.4
100.0
108.9
108.0
112.3
112.3
110.5
112.3
112.3

Un­
Weighted. weighted.
89.6
92.1
90.7
87.7
85.6
91.7
95.2
97.3
99.5
97.6
100.0
101.5
102.3
108.0
108.8
106.0
101.5
102.6

84.1
91.4
89.2
84.8
84.9
89.9
92.5
94.7
98.5
98.0
100.0
101.8
101.7
107.2
107.4
104.2
101.9
102.4

A second series of index numbers is contained in a pamphlet
published by Bemis in 1914 1 as a contribution to the current litera­
ture on the economic condition of the working classes in Spain.
Under the subject of cost of living (e l co ste d e la v id a ) the author
traces the movement of food prices through a period of years by
means of index numbers based on commodities purchased by con­
tract in certain charitable institutions of Salamanca.
The information was gathered directly from the records of the
institutions and covered the years from 1892 to 1913, inclusive. The
base period selected was the year 1901.
Average yearly prices, both absolute and relative, are shown for
17 articles, viz, flour, bacon, pulse (chick-peas), beef, potatoes, wine,
codfish, vermicelli, rice, sugar, milk, vinegar, lard, oil, pepper, salt,
and charcoal. There is no grouping of commodities and no descrip­
tion of kinds or grades is furnished.
Three general index numbers are published. The first two are un­
weighted and were obtained by computing the arithmetic mean and
the geometric mean respectively of the several price ratios based on
the prices in 1901. In calculating the weighted index the relative
price for each article on the 1901 base was multiplied by a figure de­
noting its importance in family expenditure, as determined by a study
of 13 family budgets collected by the author, and the sum of the
products thus obtained was divided by the sum of the weights. The
3 index numbers are as follows:2
1 Estudios estadisticos. Contribucion a la investigacion de la situacion economica de los trabajadores en .
Espana. Barcelona, Tipografia “ La Academica” de Serra Hnos y Russell, Ronda Universidad, 6.
2 Estudios Estadisticos, pp. 10-13.




INDEX NUMBERS-----SPAItf.

313

(Base period, 1901=100.)

Year.

1892.....................
1893.....................
1894.....................
1895.....................
1896....................
1897.....................
1898....................
1899................. .
1900.....................
1901....................
1902.....................
1903....................
1904....................
1905.....................
1906.....................
1907....................
1908....................
1909.....................
1910.....................
1911....................
1912.....................
1*13..............

Arithmetic Geometric
mean.
mean.
88.6
92.5
97.9
91.8
87.9
92.7
96.6
97.3
96.8
100
104.5
102.9
107.6
104.6
106.4
109
106.9
102.3
104.8
102.9
97
98.8

88.07
91.75
95.59
91.23
87.32
92.09
96.15
96.88
96.55
100
102.60
101
106.40
103. 50
104.66
108
104.56
100.79
103.32
101.88
96.05
97.71

Weighted
index.
84.8
90.2
90.9
91.6
86.8
91.3
98.5
95.8
96. 7
100
112
107
112.7
107.2
111.9
114. 4
107. 8
104.2
104
100.5
96.5
99.9

CONFERENCE ON INDEX NUMBERS OF THE INTERNATIONAL
INSTITUTE OF STATISTICS, SEPTEMBER, 1911.
At the thirteenth session of the International Institute of Statistics
held at The Hague, September 4 to 8, 1911, the section of economic
statistics devoted a part of its time to a discussion of index num­
bers.1 In order to secure a basis for a possible international study
of the variations of prices and the increased cost of living, a circular
had been sent to the members of the section in the various countries
early in January of that year, requesting each to prepare a table of
index numbers. A free translation of the circular reads as follows:
At the present time the high cost of living is disturbing many coun­
tries. Economists are endeavoring to ascertain how far the phenom­
enon may be attributed to unfavorable seasons, to the effect of labor
and social legislation, to the supply of gold in the world, etc. But it
is for statisticians primarily to measure and correlate the fluctua­
tions in prices as far as possible. And since the thirteenth session of
the International Institute of Statistics this year will bring together at
The Hague specialists from many countries, it seems desirable to
solicit from them answers to this question of universal interest.
We wish to request each of our colleagues to prepare a complete
series of index numbers for the common period 1881-1910 (namely
30 years), and beg of them to send in with the results obtained a
statement of what prices were used for the computations and what
methods were employed. Of course, it is understood that whatever
supplementary explanations may accompany the tables will be most
welcome.
1Bulletin de l’lnstitut International de S^atistiquo, tome X IX , 1® livraison, p. 6.3*




314

BULLETIN OF THE BUREAU OF LABOR STATISTICS.
F orm

op

Table.

The average of the prices for the year 1881 being taken as 100, the general
course of prices in the following years is as follows:
Y
Iear*

Index
number.

1881............
1882............
1883...........

100

* *..........

* *

1910............
Signature:
Address:

When the section of economic statistics met on September 5, Mr.
Waxweiler’s report for Belgium was presented printed in completed
form. The substance of all the papers received had been summarized
by Mr. de Foville for the use of the section, and he also read the fol­
lowing list of index numbers submitted:
Germany.—Mr. Zahn (for Bavaria), Mr. Zimmermann (for Brunswick), Mr. Hartwig
(for Ltibeck).
Austria-Hungary.—Mr. Mario Alberti (for Trieste), Dr. von Jankovich.
Belgium.—Mr. Hector Denis, Mr. Armand Julin, Mr. Nicolai, Mr. Sauveur, Mr.
Waxweiler (for Brussels).
Denmark.—Mr. Michael Koefoed.
France.—Mr. de FoviUe, Mr. Levasseur (for Lyc6es frangais), Mr. L. March.
Italy.—Mr. Achille Necco.
Netherlands.—Mr. Falkenburg (for Amsterdam), Mr. Methorst.
Australia.—Mr. Coghlan (for New South Wales).
Canada.—Mr. Godfrey.
Japan.— Mr. Hanabusa.

Since no papers were sent from England Mr. de FoviUe included in
his summary the English index numbers of Sauerbeck and a few
others.
When it was proposed to have these tables of index numbers
printed in the Bulletin of the International Institute of Statistics and
to take up their discussion at the next meeting of the institute after
the members had had an opportunity to study the fluctuations of
prices shown in the tables and possibly the causes that produced
them, some discussion arose as to the period of years selected, the
disparity appearing between similar tables by different authors, and
the choice of wholesale or retail prices.
Upon the first point Mr. Denis (Belgium) objected to the year 1881
being taken as the beginning of the series of index numbers, particu­
larly if the various tables constructed were to be used in attempting
to get an insight into the possible causes of the fluctuation of prices.
He summed up the periods of rise and fall of prices as evidenced in all
tables of indexes and pointed out that 1881 was a year in the midst



INDEX

NUMBERS-----INTERNATIONAL INSTITUTE OF STATISTICS.

315

of a period of depression of prices going back as far as 1874, and sug­
gested that therefore the authors of the tables submitted be asked
to lengthen their series by going back to an earlier date.
Mr. de Foville (France) justified the use of 1881 as the initial year
upon two grounds: The period of 30 years thus included from
1881 to 1910 represents approximately the lifetime of a generation,
and in going back even for so short a period it had been found difficult
to secure the earlier data in many countries. Furthermore, 1881 is
placed halfway between the extreme high level of prices in 1873 and
the lowest depression following it in 1896.
No action was taken and Mr. Waxweiler (Belgium) raised the next
question as to the disparity between similar tables. He noted that
while the curves for the price of wheat in Belgium as presented by
Mr. Sauveur and by himself are almost identical, those of meat are
quite different1 and that it might be desirable to have authors adjust
such discrepancies before publishing the tables. But Mr. Sauveur
explained the disparity in the indexes for meat by the fact that Mr.
Waxweiler chose his prices from the purchases of charitable institu­
tions in Brussels, whose consumption is exceptional in that it is con­
fined to meat of first quality and does not include pork. Both
of these meats are exactly the ones that presented the greatest fall
in prices after 1881, while his own prices covered consumption of
meat in general, which includes a large proportion of second quality
meat as well as a large quantity of pork that is used particularly in
country districts. He concluded by saying that the explanations
added by the authors to their tables seemed sufficiently complete to
clear up such points, but also called attention to the fact that some
general rule ought to be followed by the authors of index numbers
to insure the comparability of the statistics so gathered. It was
finally voted to print the tables with such comments as their respec­
tive authors thought desirable to add in view of this discussion so
that the minutes of the meeting of this section as printed in the
bulletin would clear up the matter of the comparability of the tables
even to the least experienced of readers.
The question of including retail as well as wholesale prices was
brought up by Mr. Bowley, a member of the section, upon the ground
that the fluctuations in retail prices for both long and short periods
were less abrupt than those of wholesale prices. Mr. Lucien March
(France) called to the attention of the member the extreme difficulty
of identifying the class of articles sold at retail—a difficulty which
existed even in wholesale trade as exemplified by the disparity pointed
out by Mr. Waxweiler previously. He concluded with the statement
that as long as the level of retail prices followed that of wholesale
prices in general, it justified the exclusive use of the more easily
i See Bulletin de l’lnstitut International de Statistique, tome X IX , 3®livraison, pp. 209,216, et 217.




316

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

ascertained wholesale prices. He added that in his opinion the task
before the Institute was to compare the tables already prepared, to
note the discrepancies that appeared, and to ascertain the reasons
for such discrepancies so that the Institute would be in a position to
prescribe the precautions that ought to be taken in comparing the
prices of different countries or of different periods of time. With
this the discussion of index numbers practically closed.
Among the tables of index numbers submitted to the meeting of
the Institute in response to the request quoted, the following already
existed in printed form and were not reprinted in the Bulletin of the
International Institute of Statistics:
H. Denis.—Index numbers of moral phenomena.
A. Julin.—Indexes of the economic progress of Belgium from 1890
to 1908, which appeared in the Revue des Questions Scientifiques.
E. Levasseur.—Cost of living, which appeared in the Revue ficonomique Internationale.
M. Alberti.—The cost of living, salaries and wages in Trieste.
Irving Fisher.—The purchasing power of money.
A. Necco.—The course of prices of commodities in Italy.
A. Sauerbeck.—Prices of commodities in 1910, which appeared in
the Journal of the Royal Statistical Society.
Rene Thery.—Variation of prices in India, which appeared in the
Rconomiste Europeen.
A brief description of the tables printed in Bulletin X IX of the
Institute follows:
GERMANY.

Bavaria.

The course of prices in Bavaria, 1881-1910. By Dr. Fr. Zahn, director of the
Royal Bavarian State Statistical Office, pp. 126-131.
Dr. Zahn utilizes the results of an investigation by the German Imperial Statistical
Office concerning household budgets. Of the 19 articles that were thus found to
represent a typical family budget, he expands flour into 2 independent varieties,
meat into 3, and groups the rest into 7 articles, thus obtaining a list of 12 food articles
nstead of 19, but he retains the relative proportion of consumption for each as ascer­
tained by the imperial office.
Three tables are presented by him:
I. Prices of food articles (12 as explained above) per pound (one-half kilogram).
These actual average annual prices were obtained from the reports of the Bavarian
Statistical Office.
2. The course of the cost of living in Bavaria, 1881-1910. The prices of Table 1
are here multiplied by the per cent of consumption in the typical budget for the
respective food articles.
3. The general index number for each year is computed from the totals of Table 2
and for the base year, 1881.
Brunswick (city).

The course of prices in the city of Brunswick, 1881-1910. By F. W. R. Zimmermann, city treasurer, pp. 132-133.




IN D E X NUMBERS-----INTERNATIONAL INSTITUTE OF STATISTICS.

317

The single table presented gives a general index number computed for the base year
1881 for each year from 1881 to 1910. The average annual prices used for this computa­
tion were obtained from the monthly average prices of the city for 14 food articles
and straw and hay.
Llibeck.

The course of prices in Llibeck, 1886-1910. By Dr. Hartwig, director of the statistical
office of the free Hanse town of Llibeck. pp. 134, 135.
The table begins with the year 1886 because data previous to that were not available.
Therefore the general index number had to be computed on the base 1886 instead of
1881 as requested by the Institute. Average annual prices are given for 27 food
articles, 10 wholesale and 17 retail. The sources of quotations were various.
AUSTRIA-HUNGARY.

Index numbers of 45 commodities in the Austro-Hungarian Monarchy, 1867-1909
(according to the system of Sauerbeck, partially revised by the author). By Dr. Bela
von Jankovich, vice president of the Hungarian Chamber of Deputies, pp. 136-156.
These tables are discussed in detail on pages 166 to 168 of this bulletin.
BELGIUM.

Tables of index numbers. By Hector Denis, professor in the University of Brussels,
pp. 157-195.

Besides the comparative tables and the tables of exports already described on pages
172 to 175 of this bulletin, more than 20 tables are presented to illustrate various
places of economic history. These tables make no attempt to confine themselves to
the period 1881-1910 or to the base period 1881, as requested by the circular of the
Institute.
Paper on index numbers for Belgium. By Edmond Nicolai, pp. 195-200.
The paper presents a table and chart covering 40 articles placed in eight groups,
with the various sources of prices designated in a separate column. Index numbers
are given yearly from 1881 to 1910, both for individual articles and for groups; 1881
is used as the base.
Variation in prices in Belgium from 1881 to 1909. By Maurice Sauveur. pp. 201-209
The two tables of index numbers and chart submitted by Mr. Sauveur are based on
the wholesale prices published in the Annuaire Statistique de la Belgique—official
figures ascertained by the Department of Agriculture. The indexes are computed
on the base period 1881. Table I presents yearly unweighted indexes of 18 articles—
16 foods, and hay and straw. Table II groups these into cereals, vegetables, both
groups combined, and meats; and weights the indexes for each group in proportion
to the amount of home consumption.
Course of prices of 10 articles of current consumption in Brussels, 1881-1910. By
Prof. E. Waxweiler, director of the Solvay Institute of Sociology, pp. 210-218.
Prof. Waxweiler prepared his two tables and six charts in strict conformity with
the purpose of the International Institute in requesting such information, namely a
possible international study of the causes of the increase in the cost of living. He used
as a base the year 1881, computed indexes for the period 1881-1910, and selected such
commodities as he considered most important of those available influencing the cost
of living. His prices were taken from the accounts of a large charitable institution
of Brussels and are the yearly average wholesale prices paid by that institution through
public bids. The 10 commodities selected by him are meats, eggs, wheat, potatoes,
butter, milk, linen cloth, cotton cloth, coal, and wood. Table I shows the average
yearly prices and Table II the yearly index number of each commodity and also a
general index number for each year. The charts present these index numbers
graphically.




318

BULLETIN OF THE BUREAU OF LABOR STATISTICS.
DENMARK.

Variation of prices in Denmark from 1881 to 1910. By M. Koefoed, director of the
Danish Bureau of Statistics, pp. 219-220*
The single table submitted shows only the yearly general index number from
1881 to 1910, 1881=100, of 38 commodities, unevenly weighted. For computing this
index number the 38 commodities were arranged in three groups, according to their
importance in commerce, and the prices of those in group 2 were multiplied by 2 and
those of group 3 by 3. No prices are given, but they were taken from the Statistique
de l’Echange exterieur.
FRANCE.

Variation of prices in France from 1881 to 1910. By A. de Foville, member of the
Institute of France and chief counsellor of the Government auditing department
(Cour des comptes). pp. 220-222.
A double table of index numbers shows indexes for imports and for exports sepa­
rately. These were computed from values calculated by the administration of cus­
toms (.Direction Generate des Douanes). To ascertain the variation in the value of
exports and imports the administration of customs first multiplied the total imports
and exports, respectively, in any given year by the price for the previous year, then
also by the price for the given year, and from the two computed the per cent rise or
fell of the price of exports and of imports. These proportions form the basis for the
table that appears, but no details of commodities, prices, etc., are given.
Wholesale prices in France (index numbers computed for the base period 1881).
By Lucien March, director of the statistical office of France, pp. 222-223.
The table is computed for three index numbers—one for food articles, one for mis­
cellaneous articles, and one for the two together, or a general index number. Fortythree articles were used, almost the same as those of Sauerbeck, but they are not
specified. Sauerbeck’s method of computation also was used and the prices are those
fixed every year by the permanent commission on customs values.
NETHERLANDS.

Variation of prices in Amsterdam from 1881 to 1911. By Ph. Falkenburg, director
of the bureau of statistics, pp. 224-229.
A table of index numbers is presented for 23 food articles separately and an index
number for each of three groups into which these are classified, as well as a general
index number for all of them. The base period used is 1881. A second table gives
the actual prices of these 23 food articles besides 10 other commodities. Both tables
are based on the contract prices paid by the Amsterdam municipal hospitals.
Variation of prices in the Netherlands. By H. W. Methorst, director of the Central
Bureau of Statistics of Netherlands, pp. 230-234.
Table I presents index numbers for eight food articles based on contract prices paid
by two prisons, three workhouses, and an asylum for the insane. 1881 is used as the
base, and the table covers the period from 1881 to 1911. Besides the separate index,
numbers for each article a general index is also given.
For Table II data previous to 1903 were not available. It covers the period from
1903 to 1911 and gives relative prices for four kinds of bread and a general index
number for all. The base period is 1903 and the prices were obtained from bakeries
in cities of 10,000 population or over.
Table III presents index numbers for 29 commodities from 1893 to 1911, based on
retail prices of six cooperative societies. 1893 is used as the base.
CANADA.

Variation of prices in Canada from 1890 to 1910. pp. 235-236.
Reliable data previous to 1890 could not be secured in Canada and the table of
general index numbers presented is computed on the base period 1890-1899 and was




IN DEX NUMBERS-----INTERNATIONAL INSTITUTE OF STATISTICS.

319

taken from the report of R. H. Coats of the Department of Labour as presented in the
volumes entitled ' ‘Wholesale Prices in Canada, 1890-1909, ” and “ Wholesale Prices
in Canada, 1910. ”
JAPAN.

Statement of prices in Japan. By N. Hanabusa, director of the bureau of general
statistics, pp. 237-243.
Table I gives average wholesale prices, index numbers, and a general index for
four food articles—rice, barley, Japanese beans, and sake (rice wine)—for the years
1881-1909. 1881 is used as the base. The prices were taken from the figures obtained
by the Ministry of Agriculture and Commerce in its quarterly investigations into
prices in six large cities.
Table II gives average prices, index numbers, and a general index of 26 commodities
selected from among 60 which the Ministry of Agriculture and Commerce reports
regularly since 1886. The table covers the years 1886-1909. 1886 is taken as the base.
A more complete account of the index numbers compiled by Mr. Hanabusa appears
on pages 288 to 292 of this bulletin.
NEW SOUTH WALES.

Variation of prices in New South Wales. By T. A. Coghlan, general agent for
New South Wales, p. 244.
The single table presents a general index number from 1860 to 1910 with 1881 as the
base. No explanations accompany the table.

SELECT BIBLIOGRAPHY OF ADDITIONAL INDEX NUMBERS.
[The publications that are starred were not available for reference either in the Library of Congress or
in the library of the Bureau of Labor Statistics, and statements concerning them are based upon informa­
tion found in the works of other authors.]

Avenel, Vicomte Georges d\ Histoire economique de la propriety, des salaires, des
denrees, et tous les prix en general depuis Tan 1200 jusqu’en Fan 1800. Paris,
1894-1912. 6 vol.
“ Apparently draws, very roughly, an average of the mass-quantities of goods pur­
chasable with given amounts of silver at different epochs from 1200 to 1890.” 1
Barker, Wharton. The course of prices. (Published in The American. Philadel­
phia. 1896-1900.)
Beginning in January, 1896, quarterly index numbers and a general index are given
for 13 groups of 100 articles, going back as far as the year 1891 and continued until
October, 1900, shortly after which the publication of Tne American was discontinued.
The prices of January 1, 1891, were used as base.
Bourne, Stephen, On some phases of the silver question. (Published in the Journal
of the Royal Statistical Society of London, 1879.)
The chapter “ On the fall of prices” presents a general index number for the 22 arti­
cles used by The Economist, but combines the four kinds of cotton into a single index
and adds coal. The indexes are computed yearly from 1847 to 1879 for the base
period 1845-1850.
-------- On index numbers. (Published in the Reports of the British association
for the advancement of science for 1885 and 1888 and concluded in the Journal
of the Royal Statistical Society of London in 1889.)
Yearly index numbers and a general index are given for the total imports and
exports of Great Britain from 1876 to 1888, computed for the base year 1883. This
base year was taken as equal to 1,000. The component parts of this 1,000 and of the
index numbers for 1888 are shown in two tables, one for imports and one for exports.
i C. M, Walsh, The Measurement of General Exchange-Value.




320

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

Bulgaria. Index nombres des marchandises les plus importantes et des salaires des
ouvriers, d ’apres leur prix dans les villes principales de la principaute. (Pub­
lished by the Direction Generale de Statistique in the “ Statistique des prix
moyens des animaux domestiques, des principaux articles alimentaires et des
salaires des ouvriers en Bulgarie pendant la periode decennale 1893-1902. Sofia.
1906.)
Index numbers are given for 99 articles computed yearly from 1893 to 1902 for the
base period 1888-1892.
-------- Annuaire statistique du royaume de Bulgarie. (Published by the Direction
Generale de Statistique. Sofia. l e annee 1910. 2e annee 1911.)
A yearly general index is given for 86 articles from 1899 to 1910, computed for the
base 1894-1898.
Burchard, H. C. Tables are published in the Finance Reports of the Secretary of
the Treasury, Washington, 1881, 1882, 1883, and in the Report of the Director of
the Mint on the production of the precious metals in the United States in 1884.
In these reports a series of tables appears comparing the yearly average prices of
over 80 articles in each of the years 1881,1882,1883, and 1884, with the average prices
of each of the preceding years, respectively, and also with the average prices of the
56 years preceding 1881.
Calwer, Richard. Das Wirtschaftsjahr. Annual numbers. Jena. (These are pub­
lished in two parts and several years late. The last year to appear is 1909, of
which Part I was published in 1912 and Part II in 1914.)
Beginning with the year 1907, a permanent chapter is published on “ Einkommen
und Konsum. Waarenpreise.” Three tables of index numbers are given for 17 articles
of consumption in Germany. One shows the monthly index numbers for these
articles for the year, another the general index computed as far back as 1895, and a
third the yearly index for each article computed back to 1903.
*Carli, G. R. Del valore e della proporzione de’ metalli monetati con i generi in
Italia prima delle scoperte dell’ Indie col confronto del valore e della proporzione
de’ tempi nostri. ----- , 1764.
Prices of grain, wine, and oil in the year 1750 are reduced to proportions of those in
the year 1500.
Commons, Prof. John R. Comparative prices, freight rates, stock quotations for the
years 1876 to 1900, shown by percentages or index numbers. (Published in No. I
of the Quarterly Bulletin of the Bureau of Economic Research. New York.
1900.)
Yearly index numbers are given for 66 articles individually and also as arranged in
5 groups and 10 subgroups from 1878 to 1900, computed for the base period 1878-1889.
-------- Wholesale prices by monthly and quarterly averages, 1896 to 1900 and 1878
to 1882. Shown by index numbers and diagrams. (Published in No. 2 of the
Quarterly Bulletin of the Bureau of Economic Research. New York. 1900.)
Monthly index numbers are given for the same 66 articles from 1896 to 1900 and also
by groups and subgroups, as above, and for the same base. Quarterly index numbers
are given for the groups and subgroups only and for the years from 1878 to 1882 only.
Daggett, Ellsworth. A quarter of a century of prices. Salt Lake City. 1896.
By the use of a ‘ ‘ commodity unit” he computes yearly index numbers and a general
index for 21 articles in the United States from 1870 to 1894, for the base period 18701872.
Drobisch, Moritz Wilhelm. Ueber die Berechnung der Veranderung der Waaren­
preise und des Geldwerthes. (Published in the Jahrbucher fur Nationaldkonomie
und Statistik. 1871.)
He illustrates his own method of computing index numbers by applying it to
Hamburg prices of 26 articles for the years 1854 and 1867.



INDEX NUMBERS— BIBLIOGRAPHY.

321

*Dutot,-------- . Reflexions politiques sur les finances et le commerce. The Hague.
1738.
Prices at the time of Louis X II and of Louis XIY are compared.
*Ellis, A. The money value of food and raw materials. (Published in the London
Statist. June 8, 1878.)
Index numbers are given for 25 articles for the years 1859,1869,1873,1876, and the
first quarter of 1878, computed for the base year 1869.
Eulenberg, Dr. Franz. Die Presisteigerung des letzten Jahrzehnts. (Published in
the Vortrage der Gehe-Stiftung zu Dresden. 4. Band, 1912.)
Yearly index numbers and a general index are computed for 9 groups of 45 articles
from the prices of the “ Vierteljahrsheft zur Statistik des Deutschen Reiches, 1912,”
from 1899 to 1911 and for the base period 1889-1898.
Evelyn, Sir George Shuckburgh. An account of some endeavors to ascertain a
standard of weight and measure. (Published in the Philosophical transactions of
the Royal Society of London, 1789, part I; reprinted in the Bulletin de l’lnstitut
International de Statistique, 1887.)
Index numbers are given for wheat, butcher’s meat, day labor, and 12 agricultural
products at irregular periods from 1050 to 1800, computed for the year 1550 as base.
Flux, A. W. Some old trade records reexamined: A study in price movements during
the present century. (Published in the Transactions of the Manchester Statis­
tical Society. London. 1898-99.
General index numbers are given for the total export values and total import values
of British products, for seven-year periods from 1798 to 1869, computed for the base
year 1694. The same is done for France from 1873 to 1897 ana for Germany from 1891
to 1897.
*Forbes, Francis B. The causes of depression in the cotton industry of the United
Kingdom. London. 1886. (Occasional paper of the Bimetallic League, No. 3.)
The period 1884-85 is compared with 1875-76 for 12 classes of exports and 7 of imports.
Foville, Alfred de. La mouvement desprix dans le commerce exterieuer de la France.
(Published in a series of articles in the Economiste Fran9ais, July 5, 19, Nov. 1,
1879, and Apr. 29, 1882.)
Index numbers are given for imports and exports of France from 1847 to 1880,
computed for the base year 1862.
Giffen, Sir Robert. Report to the Secretary of the Board of Trade on recent changes
in the amount of the foreign trade of the United Kingdom and the prices of
imports and exports. (Parliamentary Document. Session 1885, c. 4456.)
Index numbers are given at irregular intervals of one to four years for 67 exports
from 1840 to 1883 and ior over 100 imports from 1854 to 1883, computed for the base
1861. These exports and imports include several varieties of many articles. Earlier
and less complete forms of these tables appeared, also as parliamentary documents, in
1879 (c. 2247), 1880 (c. 2484), and 1881 (c. 3079).
*Hanauer, A. Etudes economiques sur l’Alsace ancienne et modeme. Vol. II.
Denies et salaires. Paris. 1878.
The average of 10 articles purchasable with one franc in 1351-1375 is compared
with that of 1851-1875 at 25-year periods from 1351 to 1875.
Hansard, Luke. On the prices of some commodities during the decade 1874-1883.
(Published in the Journal of the Institute of Bankers. London. 1885.)
Yearly index numbers are given for 25 articles in Great Britain from 1874 to 1885,
computed for the base year 1874.
94261°— Bull. 173— 15--- 21




322

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

Hooker, R. H. The course of prices at home and abroad 1890-1910. (Published in
the Journal of the Royal Statistical Society. London. 1912.)
Various index numbers of Great Britain, United States, Germany, France, Belgium,
and Italy are all reduced to the same base period, 1890-99, and a comparison is made
between the resulting indexes for several groups of articles and also for 16 separate
articles.
Hungary. Preisstatistik. (Published by the Kon. ungarische statistische Zentralamt in “ Ungarische statistische Mitteilungen. n. s. bd. 44. Budapest. 1913.)
In the neighborhood of 40 tables of index numbers of wholesale prices are included
in this exhaustive study of prices. Many tables are computed for three different
bases: 1867-1877,1890-1899, and 1899-1903, and for five-year as well as for yearly pe­
riods, while in a few cases even monthly indexes are shown. Index numbers of the
most important grains are given for the leading countries of Europe and the United
States. For Hungary individual articles as well as a list of articles are given for the
country as a whole and for different markets. General index numbers are also given
for various countries for the several bases, usually based on existing price studies in
those countries.
Inama-Sternegg, Dr. Karl T. M. von. Der Riickgang der Waarenpreise und die
oesterreichisch-ungarische Handelsbilanz 1875-1888. (Published by the k. k.
Statistische Zentral-Kommission in the “ Statistische Monatschrift.” XVI
Jahrgang. 1890.)
Index numbers are given for 30 imports and 26 exports of Austria-Hungary from
1880 to 1888, computed yearly for the base period 1875-1879.
The same author also published a pamphlet “ Beitrage zur Geschichte der Praise,”
in which he describes the various tables of index numbers and prices that were sub­
mitted for exhibition in the exposition in Vienna in 1873.
India. Prices and Wages in India. Compiled in the office of the Director General
of Commercial Intelligence. Calcutta. 1913.
Half-yearly and yearly index numbers are given for staple articles of import, 11 at
Calcutta and 3 at Bombay, from^1873 to 1913, for the base 1873. The same is done
for exports for prices selected mainly at Calcutta.
Another table compares the index of prices of 9 articles at Calcutta and London,
yearly from 1888 to 1912, base 1873.
Index numbers are also given for 8 staple articles in Calcutta in January of each
year from 1888 to 1913, base 1873.
Yearly index numbers are also computed for the prices of articles of consumption
of troops at 16 stations in India from 1887 to 1912, base 1882-83, for each station and
for each article.
*James, Henry. The state of the nation. Causes and effects of the rise in value of
property and commodities from the year 1790 to the present time. London.
1835.
Average prices of British produce from 1798 to 1823 as compared with 1694 are
given.
Jevons, Prof. William Stanley. Investigations in currency and finance. London.
1909. (Reprints of various articles published earlier, including: A serious fall in
the value of gold ascertained and its social effects set forth, 1863; The variation
of prices and the value of currency since 1782, 1865; The depreciation of gold,
1869.)
Three sets of index numbers appear. One series gives yearly index numbers for
39 articles from 1845 to 1865, computed for the base period 1845-1850. The second
gives yearly index numbers and a general index for 12 groups of 40 articles from 1762
to 1865, computed for the base 1782. The general index for these 40 articles is also
given in 10-year periods from 1789 to 1869, but for the base year 1849. The third
series gives a general index for 50 leading articles of commerce from 1847 to 1869,
computed yearly for the base 1849.




INDEX NUMBERS— BIBLIOGRAPHY.

323

Juergens, Carl H. Movement of wholesale prices in New York City, 1825-1863.
(Published in the quarterly publications of the American Statistical Association.
1910-11.)
Yearly index numbers are given for each of 74 articles whose prices were taken
from the report of the Secretary of the Treasury for June 30, 1863. A general index
is also given for each year for the whole period covered, 1825-1863. The year 1860
is used as base.
, Julin, Armand. The economic progress of Belgium from 1880 to 1908. (Published
in the Journal of the Royal Statistical Society. 1912.)
In a presentation of a large number of index numbers that do not concern prices
there appear also yearly index numbers for exports, imports, agricultural products,
and several other articles from 1880 to 1908, computed for the base year 1884.
Levasseur, Emile. La question de Tor. Paris. 1858.
Yearly index numbers are given for 2 groups (8 articles) of exports and 3 groups
(28 articles) of imports from 1847 to 1856, computed for the base 1826.
March, Lucien. Le mouvement des prix et l’activite productrice. (Published in
the Bulletin of the Bureau de la Statistique Generale. Paris. 1911.)
In making a comparative table of general index numbers for France, Germany,
England, and the United States he computes a yearly index for France of 43 articles
out of Sauerbeck's 45 from 1840 to 1910 and for the base period 1891-1900.
Mulhall, Michael George. History of prices since the year 1850. London. 1885.
Several tables of index numbers are given. Yearly, 5-yearly, and 10-yearly index
numbers are given for total imports, total exports, and both together from 1854 to
1884, base 1841-1850. Likewise 10-yearly indexes are given for each of 50 imports
and 50 exports from 1850 to 1884, base 1854-1860. Another table shows 10-yearly
index numbers for each of 7 agricultural products and 7 manufactured products
from 1782 to 1884, base 1782-1790.
Netherlands. Prijzen van levensmiddelen te Amsterdam. Amsterdam. 1911.
(Published by the Bureau van statistiek in the “ Statistische medeelingen
uitgegeven door het Bureau van statistiek der gemeente Amsterdam” No. 35.)
The contract prices paid by the municipal hospitals are used for computing the
yearly index numbers of 26 articles from 1881 to 1911 for the base year 1881. A gen­
eral index is also computed.1
-------- . Maandschrift van het centraal bureau voor de statistiek. 1913; 1914.
The series of index numbers published in the 1913 volume is based^ on the contract
prices paid by 5 large institutions. Yearly indexes and a general index are given
for 9 articles, computed for the base year 1881, from 1881 to 1912.
New South Wales. Index numbers of exports at Sydney, principal articles of
domestic produce. Published in the Official Yearbook 1913.
Index numbers are given for two groups and also for the total of exports, computed
yearly from 1901 to 1913, for the base year 1901. The monthly index numbers for this
table are also given for the year 1913, and this latter table likewise appears in the
Monthly Statistical Bulletin of New South Wales.
Palgrave, Sir R. H. Inglis. Currency and standard of value in England, France, and
India, and the rates of exchange between these countries. (Published in the
Third Report of the Royal commission appointed to inquire into the depression
of trade and industry. London. 1886.
The index numbers of the Economist are presented as reduced to the base 1865-1869
and computed^ for each article yearly and five-yearly from 1870 to 1886. The same is
done for 22 articles in France, corresponding as nearly as possible to those of the Econo­
mist and computed from 1865 to 1885. Each of these series is then weighted to form a
new table. The series for India gives yearly, five-yearly, and general indexes for 10 ar­
ticles (also as arranged in three groups) from 1870 to 1884 for the same base 1865-1869.




1Compare pp. 293 and 294 of this bulletin.

324

BULLETIN OF THE BUREAU OF LABOR STATISTICS.

Porter, George Richardson. The progress of the nation, in its various social and
economic relations, from the beginning of the nineteenth century. London. 2d
edition. 1847.
A monthly general index number is given for 50 articles in London, also for wheat
separately, from January, 1833, to December, 1837. This is computed upon the prices
of the first week in January, 1833, as base.
Powers, Le Grand. Modern variations in the purchasing power of gold; an investi­
gation into the extent and causes of recent price variations. (Published in the
fifth biennial report of the Bureau of Labor of the State of Minnesota. St. Paul.
1896.)
Index numbers are given for 16 articles of farm production for ^the years from 1862
to 1895, computed yearly for the base year 1872, and combined in a great variety of
ways, i. e., by States, years, groups of articles, etc.
Rogers, J. E. Thorold. A history of agriculture and prices in England from the year
after the Oxford parliament (1259) to the commencement of the continental war
(1793); compiled entirely from original and contemporaneous records. Oxford.
1887. Vol. Y.
The prices of 39 articles are compared for the periods 1541-1582,1583-1702,1583-1642,
1643-1702, the average prices of each period in turn being compared with those of the
period immediately preceding.
Rumania. Bulletin statistique de la Roumanie. La Statistique Generale. Bucha­
rest. 1911.
The rise and fall per cent of prices as compared with the period immediately pre­
ceding is shown semimonthly, monthly, quarterly, semiannually, and annually from
1908 to 1911 for five grains.
Tyszka, Carl von. Die Bewegung der Preise einiger wichtiger Lebensmittel, insonderheit der Fleischpreise in Deutschland und im Auslande, unter besonderer
Berucksiehtigung Englands. (Published in the Jahrbiicher fur Nationalokonomie und Statistik. 3. F. 42. 1911.)
Index numbers are computed for 10-year periods for 10 articles (8 meats, wheat,
rye) from 1881 to 1910, for the base 1891-1900. Other tables compare similar indexes
for Berlin with those of London, New York, Paris, from 1895 to 1910.
United States. Index numbers of production per capita and price of important farm
products. Published by the Department of Agriculture in The Crop Reporter,
Washington. April, 1912.
These index numbers are not given in figures, but are represented by graphs.
Yearly indexes for the total of 10 important crops are computed from 1866 to 1911 for
the base period 1866-1908.
Walras, Leon. Etudes d’economie politique appliquee. Lausanne and Paris. 1898.
Yearly index numbers and a general index are given for 20 articles at Berne, Switzer­
land, from 1871 to 1884, computed for the base period 1871-1878.
Wasserab, Karl. Preise und Krisen. Gekronte Preisschrift “ Ueber die Veranderungen der Peise auf dem allgemeinen Markt seit 1875 und deren Ursachen.”
Stuttgart. 1889.
Index numbers are computed for 31 articles in Germany by comparing the average
price of the period 1882-1885 with that of 1861-1870 as base.
Whitehead, Thomas Henderson. The critical position of British trade with Oriental
countries. (Paper read before the Royal Colonial Institute, February 12, 1895,
and reprinted from the proceedings of the Institute.)
On page 35 of this volume the author presents a table of yearly index numbers
compiled by W. S. Wetmore, of Shanghai, from official returns, covering 20 staple
commodities of China from 1874 to 1893, computed for the base year 1873.




INDEX.
K'
Page.
Aim and scope of present bulletin............................................................................................................
5
Alberti, Mario, index numbers of, Austria-Hungary...........................................................................168-172
American index numbers, comparison of the leading, 1890 to 1913.......................................................93-112
Annalist (New York Times), index numbers of................................................................................... 138-140
Arithmetic mean, and other forms of averages.............................................................................. 21,22,80-93
Atkinson, Fred J., index numbers of, India........................................................................................ 276-282
Australia, index numbers of Commonwealth Bureau of Census and Statistics..................................157-166
Austria-Hungary:
Alberti, Mario, index numbers of................................................................................................... 168-172
Jankovich, Dr. B61a von, index numbers o f..................................................................................166-168
317
Tables submitted to International Institute of Statistics, description of......................................
Averages and aggregates, in the making of index numbers.................................................................... 80-93

B.
Base periods, in the making of index numbers........................................................................................36-44
Base periods used in index numbers in different countries:
Australia..............................................................................................................................................
158
Austria-Hungary.............................................................................................................................. 167,169
Belgium...............................................................................................................................................
174
Canada.................................................................................................................................................
177
Denmark..............................................................................................................................................
187
France................................................................................................................................. 189,193,196,204
Germany.......................................................................................................... 209,222,225,228,232,241,251
Great Britain...................................................................................................................... 256,257,264,270
India................................................................................................................................................. 277,278
Italy.................................................................................................................................................. 283,285
Japan...................................................................................................................................................
289
Netherlands.........................................................................................................................................
293
New Zealand........................................................................................................... ........................ 295,296
Norway................................................................................................................................................
310
Russia..................................................................................................................................................
305
Spain.................................................................................................................................................... 310
United States........................................................................................................ 116,129,138,143,149,153
Bavaria, Germany, tables submitted to International Institute of Statistics, description of................
316
Belgium, index numbers of Hector Denis................................................................................................ 172
Belgium, tables submitted to International Institute of Statistics, description of................................ 317
Bernis, Francisco, index numbers of, Spain........................................................................................ 309-313
Bibliography, select, of additional index numbers..............................................................................319-324
Bradstreet’s index numbers................................................................................................................ 141-148
British Association for the Advancement of Science, table of weights of..............................................
184
Brunswick (city), Germany, tables submitted to International Institute of Statistics, description of. 316,317
“ Business barometer,” use of index numbers as a ...............................................................................66, 111
C.
Canada, index numbers of Department of Labor................................................................................ 176-186
Canada, tables submitted to International Institute of Statistics, description of.............................. 318,319
Carli, G. R ., Italian inventor of index numbers.................................................................................... 5,6,34
“ Chain” and fixed-base index numbers................................................................................... 36,38,39,41,42
Commodities, number and kinds of, included in the making of index numbers...................................44-71
Commodities, number, class, description, and grouping of:
Australia...........................................................................................................................................158,159
Austria-Hungary............................................................................................................................. 167,169
Belgium...................................................................... '........................................................................
174
Canada..............................................................................................................................................178-182
Denmark..............................................................................................................................................
187
France.................................................................................................................... 190,193,194,196-198,205
Germany........................................................................ 209-217,223-226,229-231,233,234,242-248,252,253
Great Britain............................................. .......................................................... 257,258,264-266,271-273
India................................................................................................................................................. 278-280
Italy........................................................................................................................................... 283,285,286
Japan................................................................................................................................................ 289-291
Netherlands.........................................................................................................................................
293
New Zealand.................................................................................................................................... 296,297
Norway..............................................................................................................................................
301
Russia............................................................................................................................................... 306-308
Spain....................................................................................................................................................
310
United States, Annalist (New York Times).....................................................................................
139
United States, Bradstreet’s ...................................................... ...................................................... 144-146
United States, Bureau of Labor Statistics..................................................................................... 117-124
United States, Committee on Finance, United States Senate...................................................... 129-135
United States, Dun’s Review.............................................................................................................
150
United States, Gibson’s......................................................................................................................
154
Contract and market prices, and import-export values, in the making of index numbers...................30-33
“ Cost of living,” use of index numbers to measure changes in...............................................................70,89




325

326

INDEX.
D.

Page.
Decils and medians....................................................................................................................................14-16
Denis, Hector, index numbers of, Belgium.......................................................................................... 172-175
Denmark, index numbers of the State Statistical Bureau................................................................... 186-188
Denmark, tables submitted to International Institute of Statistics, description of..............................
318
Dun’s Review, index numbers of.......................................................................................................... 148-152
E.
Economist (London), index numbers of...............................................................................................261-2C9
F.
Formulae:
Computing geometric mean with or without use of relative prices.................................................
83
Commuting relative prices from aggregates of actual prices..............................................................
93
Shifting relative prices derived from aggregates of weighted money prices....................................
91
France:
Annuaire Statistique de la France, index numbers of.................................................................. 188-192
La Reforme, Economique, index numbers of................................................................................. 195-203
Levasseur, Emile, index numbers of.............................................................................................. 204-207
Statistique Generale de la France, index numbers o f................................................................... 192-195
Tables submitted to International Institute of Statistics, description of........................................
318
G.
Geometric mean, and other forms of averages.......................................................................... 81-83,86,88,90
Germany:
Imperial Statistical Office, index numbers of................................................................................ 208-218
Jahrbucher fur Nationaldkonomie und Statistik, index numbers of........................................... 219-239
Schmitz, Otto, index numbers of................................................................................................... 240-250
Soetbeer, Adolf, index numbers of................................................................................................, 250-255
Tables submitted to International Institute of Statistics., description o f .....................................316,317
Gibson’s weekly market report, index numbers of.................................................................... . ........ 153-156
Great Britain:
Board of Trade, index numbers of................................................................................................. 255-261
Economist (London), index numbers of................................................................................. .... 261-269
Sauerbeck, Augustus, index numbers of........................................................................................ 269-276
H.
Harmonic mean, and other forms of averages.........................................................................................
81
History of index numbers,.......................................................................................................................
5-8
History of index numbers, different countries:
Australia..............................................................................................................................................
157
Austria-Hungary............................................................................................................................. 166-168
Belgium...............................................................................................................................................
172
Canada.................................................................................................................................................
176
Denmark.......................................................................................................................................... 186,187
France................................................................................................................................. 189,192,195,204
Germany...................................................................................................................... 208,219,240,250,251
Great Britain................................................................................................................ 255,256,262-264,270
India....................................................................................................................................................
270
Italy........................................................................................................................................... 282,283,285
Japan................................................................................................................................................ 288,289
Netherlands.........................................................................................................................................
293
New Zealand......... .............................................................................................................................
295
Norway.......................................................................................................................................*........
300
Russia............................................................................ - ...................................................................
305
Spain................................................................................................................................................ 309,310
United States.................................................................................................. 115,128,138,141-143,149,153

I.
Import-export values, and contract and market prices, in the making of index numbers.................. 30-33
Index numbers, conclusions of present study...................................................................................... 112-114
Index numbers, different countries. (See Tables of results.)
India, index numbers of Fred. J. Atkinson......................................................................................... 276-282
International Institute of Statistics, conference of, on index numbers............................................. 313-319
Interpolation of prices:
Australia..............................................................................................................................................
160
Canada.................................................................................................................................................
183
France............................................................................................................................................... 191,199
Great Britain............................................................................................................................. 259,266,273
Italy............................................................................................. - .....................................................
283
Japan...................................................................................................................................................
291
New Zealand.......................................................................................................................................
298
Russia.................................................................................................................................................
308
United States......................................................................................................... 125,135,140,147,150,154
Italy, index numbers of Achille Necco.................................................................................................. 285-288
Italy, index numbers of Annuario Statistico Italiano................................................................... .
282-284
J.
Jankovich, Dr. B&a von, index numbets of, Austria-Hungary.......................................................... 166-168
Japan, index numbers of Department of Agriculture and Commerce................................................ 288-292
Japan, tables submitted to International Institute of Statistics, description of...................................
319
K.
Koefoed, Michael, index numbers of, Denmark................................................................................... 186-188




IN D E X .

327

L.

Tage.
Levasseur, £mile, index numbers of, France...................................................................................... 201-207
Liibeck, Germany, tables submitted to International Institute of Statistics, description of...............
317
M.
Mcllraith, James W ., index numbers of, New Zealand....................................................................... 295-300
Market and contract prices, and import-export values, in the making of index numbers................... 30-33
Median, and other forms of averages............................................................................... 14-16, 81,84,85,87-89
Methods used in making index numbers, varieties of.............................................................................25-93
Actual versus relative prices...............................................................................................................34,35
Averages and aggregates..................................................................................................................... 80-93
Base periods........................................................................................................................................ 3B-44
Commodities included, number and kinds of................................................................................... 44-71
Market prices, contract prices, and import-export values................................................................30-33
Quotations, collecting and publishing............................................................................................... 27-30
TJses and methods, relations between................................................................................................ 25-27
Weighting, problems of...................................................................................................................... 71-80
Mode, and other forms of averages............................................................................................................21,81
N.
Necco, Achille, index numbers of, Italy............................................................................................... 285-288
Netherlands, index numbers of Netherlands Statistical Office........................................................... 293,294
Netherlands, tables submitted to International Institute of Statistics, description of.........................
318
New South Wales, tables submitted to International Institute of Statistics, description of................ 319
New Zealand, index numbers of James W . Mcllraith......................................................................... 295-300
“ Normal law of error ” ............*................................................................................................................
18
Norway, index numbers of Einar Ruud............................................................................................... 300-304
P.
Price quotations, additions and substitutions:
Australia........................................................................................................................................... 159,160
Austria-Hungarv.................................................................................................................................
170
Canada.............................................................................................................................................. 182,183
France................................................................................................................................. 190,191,198,199
Germany................................................................................................................ 217,218,234,248,253,254
Great Britain............................................................................................................................. 258,266,273
Italy.....................................................................................................................................................
283
Japan...................................................................................................................................................
291
New Zealand.................................................................................................................................... 297,298
Russia..................................................................................................................................................
308
United States, Annalist (New York T im es)..................................................................................... 139
United States, Bradstreet’s................................................................................................................
147
United States, Bureau of Labor Statistics.................................................................................... 124,125
United States, Committee on Finance, United States Senate.........................................................
135
United States, Dun’s Review............................................................................................................
150
United States, Gibson’s ...................................................................................................................... 154
Price quotations, source of:
Australia........................................................................................................................................... 157,158
Austria-Hungary............................................................................... ..............................................167,169
Belgium................................................................................ ; ......................................................... 173,174
Canada.................................................................................................................................................
176
Denmark.............................................................................................................................................. 187
France........................................................................................................................... 189.192.193,196,204
Germany................................................................................................... 203,209,220-222,232.240,241,251
Great Britain.................................................... , ....................................................................... 256,264,270
India....................................................................................................................................................
277
Italy.................................................................................................................................................. 283,285
Japan...................................................................................................................................................
289
Netherlands.........................................................................................................................................
293
New Zealand........................................................................................................................................ 295
Norway................................................................................................................................................. 301
Russia.................................................................................................................................................. 305
Spain...................... .............................................................................................................................
310
United States, Annalist (New York Tim es)....................................................................................
138
United States, Bradstreet’s................................................................................................................
143
United States, Bureau of Labor Statistics......................................................................................
116
United States, Committee on Finance, United States Senato......................................................128,129
United States, Dun’s Review............................................v..............................................................
149
153
United States, Gibson’s ................................................ ....................................................................
Prices:
Actual versus relative......................................................................................................................... 34,35
Fluctuations in, characteristics o f......................................................................................................10-24
Level of, difficulties of measuring changes in the.............................................................................. 8-10
Market and contract prices, and import-export values, in the making of index numbers.............30-33
Quotations of, the collecting and publishing of original...................................................................27-30
Prices, methods of showing and computing:
Australia..............................................................................................................................................
158
Austria-Hungary.................................................................................................................................
169
Canada.................................................................................................................................................
177
France........................................................................................................................................ 189,193,196
Germany............................................................................................. 209,222,223,225,228,233,241,251,252
Great Britain............................................................................................................................. 257,264,270
India....................................................................................................................................................
278
Italy.....................................................................................................................................................
283
Japan.................................................................................................................................................... 289
New Zealand........................................................................................................................................ 296
Russia............................................................................................................................................... 305,306




328

INDEX.

Prices, methods of showing and computing—Concluded.
Page.
Spain....... ............................................................................................................................................ 310
United States, Annalist (New York T im es)................................................................................. 138,139
United States, Bradstreet’s............................................................................................................. 143,144
United States, Bureau of Labor Statistics........................................................................................
117
United States, Committee on Finance, United States Senate.........................................................
129
United States, Dun’s Review............................................................................................................
149
United States, Gibson’s ......................................................................................................................
153
Publications, containing index numbers:
Australia..............................................................................................................................................
157
Austria-Hungary.............................................................................................................................166-168
Belgium...............................................................................................................................................
173
Canada.................................................................................................................................................
176
Denmark.......................................................................................................................................... 186,187
France........................................................................................................................... 188,189,192,195,204
Germany....................................................................................................................... 208,219,240,250,251
Great Britain...................................................................................................................... 255,261,262,269
India....................................................................................................................................................
276
Italy.................................................................................................................................................. 282,285
288
Japan...................................................................................................................................................
Netherlands.........................................................................................................................................
293
New Zealand.......................................................................................................................................
295
Norway................................................................................................................................................
300
Russia..................................................................................................................................................
305
Spain.................................................................................................................................................... 309
United States......................................................................................................... 115,128,138,141,148,153
Q.
Quotations of prices, the collecting and publishing of original............................................................... 27-3C
R.
Russia, index numbers of Ministry of Commerce and Industry......................................................... 305-309
Ruud, Einar, index numbers of, Norway............................................................................................. 300-304
S.
Sauerbeck, Augustus, index numbers of, Great Britain...................................................................... 269-276
Schmitz, Otto, index numbers of, Germany........................................................................................ 240-250
Schuckburg-Evelyn, Sir George, English inventor of index numbers...................................................
6
Scope and aim of present bulletin............................................................................................................
5
Shifting index numbers from one base to another, methods o f........................................................... 39-44
Soetbeer, Adolf, index numbers of, Germany....................................................................................... 250-255
Spain, index numbers of Francisco Bemis........................................................................................... 309-313
T.
Tables of results:
Australia, Commonwealth Bureau of Census and Statistics....................................................... 162-166
Austria-Hungary, Dr. von Jankovich........................................................................................... 167,163
Austria-Hungary, Mario Alberti..................................................................................................... 170-172
Belgium, Hector Denis.................................................................................................................... 174,175
Canada, Department of Labor...........................................................................................................
186
Denmark, State Statistical Bureau...................................................................................................
188
France, Annuaire Statistique de la France.................................................................................... 191,192
France, Statistique G£n6rale de la France.....................................................................................194,195
France, La R^forme Economique.................................................................................................. 199-203
France, Emile Levasseur................................................................................................................. 206,207
Germany, Imperial Statistical OUice.................................................................................................
218
Germany, Jahrbucher fur Nationalokonomi3 un i Statistic......................................................... 235-239
Germany, Otto Schmitz.................................................................................................................. 249,250
Germany, Adolf Soetbeer...................................................................................................................
255
Great Britain, Board of Trade........................................................................................................ 260,261
Great Britain, Economist (London)............................................................................................... 267,269
Great Britain, Augustus Sauerbeck................................................................................................ 274,276
India, Fred. J. Atkinson................................................................................................................. 281,282
Italy, Annuario Statistieo Itaiiano....................................................................................................
284
Italy, Achille Necco................................. ....................................................................................... 287,288
Japan, Department of Agriculture and Commerce........ ............................................................. 291,292
Netherlands, Statistical Office....................................................................................................... 293,294
New Zealand, James W . McIlraKIi................................................................................................ 299,300
Norway, Einar Ruud...................................................................................................................... 301-304
Russia, Ministry of Commerce and Industry...................................................................................
309
Spain, Francisco Bernis...................................................................................................................311-313
United States, Annalist. (New York Times).....................................................................................
140
United States, Bradstreet’s............................................................................................................. 147,148
United States, Bureau of Labor Statistics.....................................................................................126,127
United States, Committee on Finance, United States Senate...................................................... 137,138
United States, Dun’s Review......................................................................................................... 151,152
United States, Gibson’s ......................................................................................................................
156
Testing of index numbers:
Australia..............................................................................................................................................
102
Austria-Hungarv................................................................................................................................
170
Canada............................................................................................................................................. 183-185
France................................................................................................................................ 191,199,205,206
Germany..................................................................................................................... 228,232,234,249,254
Great Britain..................................................................................................................... 260,266,267,274
Ita ly ....................................................................................................................................................
284
Japan...................................................................................................................................................
291
New Zealand.......................................................................................................................................
299
Spain..................................................................................................................................................
311
United States................................................................................................. 126,137,140,147,151,155,156




IN D E X .

329

u.
United States, index numbers published in.........................................................................................
Annalist (New York Times)............................................................................................................
Bradstreet’s......................................................................................................................................
Bureau of Labor Statistics..............................................................................................................
Committee on Finance, United States Senate...............................................................................
Dun's Review..................................................................................................................................
Gibson’s ...........................................................................................................................................
Uses of and methods of making index numbers, relations between..................................................

115-156
138-140
141-148
115-127
128-138
148-152
153-156
25-27

W.
Weighting, methods used in different countries:
Australia.......................................................................................................................................... 160,161
Austria-Hungary................................................................................................................................
170
Belgium..............................................................................................................................................
174
Canada................................................................................................................................................
183
Denmark.............................................................................................................................................
183
France.............................................................................................................................................. 191,199
Germany............................................................................................................................. 226-228,248,254
Great Britain............................................................................................................................ 25^,266,273
India................................................................................................................................................. 280,281
Italy........................................................................................................................................... 284,286,287
Japan...................................................................................................................................................
291
Netherlands.........................................................................................................................................
293
New Zealand.......................................................................................................................................
298
Norway................................................................................................................................................
301
Russia.................................................................................................................................................. 308
Spain...................................................................................................................................................
311
United States, Annalist (New York T im es)....................................................................................
140
United States, Bradstreet’s...............................................................................................................
147
United States, Bureau of Labor Statistics..................................................................................... 125,126
United States, Committee on Finance, United States Senate..................................................... 135-137
United States, Dun’s Review......................................................................................................... 150,151
United States, Gibson’s.................................................................................................................. 154,155
Weighting, problems of, in the making of index numbers..................................................................... 71-80
Weights, table of, for construction of an index number, of British Association for the Advancement of
Science.............................................................................................................. ......................................
184