H

INTERNATIONAL LABOUR OFFICE

STUDIES AND REPORTS
Series N (Statistics) No. 9

METHODS OF CONDUCTING
FAMILY BUDGET ENQUIRIES

w*

GENEVA
1926

1P,7109

—, i

•

PREFACE

At the Second International Conference of Labour Statisticians, held in Geneva in April 1925, the following resolution
was passed :
It is desirable that in those countries in which no family budget
enquiries have been held since 1920-1921, such enquiries should be
undertaken as soon as economic conditions are
sufficiently favourable,
and if possible not later than the year 1928 1.
This resolution formed part of a series dealing with methods
of calculating index numbers to measure changes in the cost
of living. The information contained in family budget enquiries provides a valuable basis for the calculation of such index
numbers by showing the relative importance in consumption
of the different commodities included in the index. To ensure
uniformity in the basis on which the cost-of-living index numbers are calculated, it is necessary that the same principles
should be adopted in different countries in conducting family
budget enquiries.
As is indicated in the present study, the weights necessary
for the compilation of cost-of-living index numbers may be
obtained by a comparatively simple enquiry in which budgets.
for a small number of families are secured for a short period.
Information obtained in family budget enquiries, however, is
of value for other purposes also. By means of a somewhat
more detailed enquiry, valuable information may be obtained
with regard to the standard of living of the class or classes from
which the budgets are secured. The present study, therefore,
deals with the whole subject and not merely with enquiries where
the sole purpose is to provide weights for the calculation of
cost-of-living index numbers. It forms a continuation of the
series of studies which the International Labour Office is making on the methods adopted in various countries with regard
to different kinds of labour statistics.
The study has been prepared by Mr. J. H. Richardson, under
the direction of Dr. Karl Pribram, Chief of the Statistical
Section.
1

INTERNATIONAL LABOUR O F F I C E :

ference of Labour Statisticians. Studies
tistics), No. 8, p . 70. Geneva, 1925.

The Second

International

Con-

and Reports, Series N (Sta-

CONTENTS
Page
PREFACE
I.
II.

3

INTRODUCTION

7

COLLECTION OF INFORMATION

Selection of the Families
Duration of Enquiry
Methods of Securing the Budgets
III.

SCOPE OF INFORMATION

General Conditions
Income
Expenditure
IV.

COMPILATION OF RESULTS

Classification
Income
Expenditure
Methods of Reducing Data for Families of Different
Size to Terms of a Common Unit

13

13
16
20
26

26
28
30
37

37
40
42
45

APPENDICES
APPENDIX
APPENDIX

I.

Family Budget Enquiries Conducted in Different
Countries
II. Specimen Forms Used in Various Enquiries . . . .

55
79

I
INTRODUCTION

From the beginning of the nineteenth century, and especially
from the middle of that century, there has been an evergrowing
interest in many countries in the conditions under which people
are living. It was evident even to the casual observer that the
standard of living of large numbers of families, whether in the
crowded industrial centres or in rural districts, was very low.
Questions arose which no one could answer. "What was the actual
standard of living of any given section of the community ? On
what incomes were various families living ? Were the incomes
adequate if wisely expended and was a good deal of poverty
the result of improvidence and excess ? What kinds of foods
were being consumed and were the quantities sufficient for the
maintenance of health and efficiency ? What were housing
conditions like ? Were the homes sanitary, or was the unsatisfactory state of the dwellings to a large extent responsible
for the bad health of the community and for the frequent outbreaks of epidemics? Had the different families adequate accommodation, or were their homes so overcrowded as to undermine health and make decency impossible ? What relation
existed between size of family and poverty ?
It was because no satisfactory answers were forthcoming
that the search for facts began. The first investigations were
very tentative in character, being conducted by private individuals, some of whom were merely interested in collecting data
and making comparative studies, while others were desirous
of calling public attention to the evils they believed to exist
and of urging the adoption of remedies. In each case, however,
the object was to provide information as to the actual conditions under which the families covered were living, and to
throw light on the general conditions and standard of living of
the community as a whole or of certain classes of the community.
The early enquiries were of a very limited character as the
individuals who conducted them were unable to do more than

— 8 —
obtain information from a small number of families. In making the enquiry the investigator came into direct personal relations with the family or families concerned. Notes were taken
on all points on which information was desired and any special
circumstances bearing on the family life were recorded. Frequently, information was obtained for a considerable period
of time, and the head of the family or the housewife might be
asked to record the details of income or expenditure. These
records were checked by the investigator and doubts cleared
up by supplementary enquiry. The method used by Le Play
is typical of the more detailed researches of the early investigators. In the studies of working class conditions which he
undertook, Le Play made a thorough study of a worker's
family, living in it or near it for some weeks while he examined
the- details of its income, expenditure and general conditions
of living 1 .
Evidently, if the family studied was of exceptional character
it would be impossible to regard the results as at all representative of the condition of other families. Consequently the first
problem before the investigator was the selection of the family
or families to be studied. In making the selection the aim
was to obtain information from families whose conditions where
typical of those of a large number of other families. According to the purpose of the investigator, the family or families
studied might be selected so as to be typical of some comparatively small special group—for example, of workers in a particular industry or locality—or might be chosen as representative
of a larger group—for example, the workers as a whole.
This method of studying a few families in detail, known as
the intensive or monographic method of investigation, had the
advantage that a wealth of information could be obtained by
the investigator as a result of his personal contact with the
families selected for study. Also, if the good will and cooperation of the family were secured, records might be obtained
1
During the period 1829 to 1879, Le Play conducted successive enquiries on a uniform plan into the conditions of life of workers' families
in different European countries. The results were published in Les
ouvriers européens : études sur les travaux, la vie domestique et la condition
morale des populations ouvrières de l'Europe ; Paris, 1879. In this
work, a series of monographs on the budgets of different families is
given. See also a paper by Mr. E. CHEYSSON and Mr. ALFRED TOQUÉ :
" Les Budgets comparés des cents monographies de Familles ", published
in the Bulletin of the International Statistical Institute, Vol. V; Rome,
1890.

— 9 —
for a lengthy period. On the other hand, however careful the
selection, some of the details for any one family were likely
to be peculiar to that family and therefore of little value from
the point of view of general conditions in the community as
a whole or of a given class. Consequently, when generalisations were made on the basis of the information compiled for
a few families they were liable to criticism on the ground that
the data were unrepresentative.
During the period when the earliest enquiries were conducted,
laissez faire doctrines were predominant, and, as has been seen,
investigations were conducted either to satisfy the curiosity
of the student or to prove that an economic system based on
non-interference often resulted in intolerable conditions of
living. As the evils consequent upon individualism became
more recognised, public opinion demanded further investigation. Proposals were made for the removal of various social
evils and the need was felt for more complete information on
living conditions than that which private investigators could
compile. Information was required against which the objection could not be raised that it was biased in accordance with
the views of the investigators. To meet these needs and to
supply a basis for reforms of various kinds, the statistical offices
of central Governments began to conduct family budget
enquiries. Most of these investigations date from the last
decade of the 19th century onwards.
The general object of the enquiries conducted by Government
departments has often been similar to that of investigations
made by private persons or organisations, namely, to secure
accurate information as to the standards of living of the group
studied. The results are of value for a number of practical
purposes. The information may be used to supplement other
data in order to throw light on the distribution of the national
dividend among different classes of the population. Also, if
enquiries covering the same classes are conducted at different
dates, the results may serve to indicate the relation between
the standards of living at the different dates. The conclusions
reached may be considered in relation to changes in the total
production of the community, and deductions be possible
as to whether the class considered is receiving an increased or
decreased share of the national dividend.
In cases where information is required as to the conditions
of living of some given class of the population, the results of

— 10 —
family budget enquiries are of special value. Thus, in a number of countries enquiries have been conducted in order to find
out the conditions of living of workers in sweated industries
with a view to securing improvements in their wages and
other conditions of labour. Mention may be made of the enquiry conducted in France in 1905, and of the investigation
made in York by Mr. B. S. Rowntree in 1899. Many other
similar enquiries have been conducted into the conditions
of living of the lowest classes of the populations of different
countries.
The results of family budget enquiries have in some cases
been used as a basis for the drawing up of standards of subsistence or standards of comfort, and these in turn have been
used for the purpose of fixing minimum rates of wages. The
data collected with regard to consumption may be used as a
means of determining the average consumption of certain
commodities in a country. Thus if a family budget enquiry
shows the quantity of flour, meat, etc., consumed by a typical
family, these figures, in relation to statistics of population,
may be used as a basis of estimates of the consumption of these
commodities in the community as a whole.
On account of the necessity of compiling budgets from which
general conclusions on the lines indicated above can be drawn,
and because the resources at their disposal are generally greater,
the investigations conducted by public authorities have been
on a larger scale than those of private individuals or associations. The choice as to the scale of an enquiry lies between
that of the census, by which data are obtained from all families
in the group covered, and the representative method. In the
case of family budget enquiries, the census method is impracticable. Consequently, public authorities are limited to the
representative method. In resolutions on this method adopted
by the Sixteenth Session of the International Statistical Institute held in Rome in 1925 \ two chief methods, by which a
sample may be taken from the whole, are indicated. According to the first method, that of random selection, units are
taken in such a way that each unit in the whole has an exactly
1

INSTITUT

INTERNATIONAL

D E STATISTIQUE,

Sixteenth

Session,

R o m e : Bulletin quotidien, No. 7, p. 5, 2 October 1925. T h e resolutions were based on reports and memoranda o n t h e representative
method of investigation, its value and limitations, prepared by Mr.
Adolph Jensen, Prof. Bowley, Mr. Lucien March, a n d Prof. Verrijn
Stuart.

— 11 —
equal chance of inclusion in the sample. This method would
be difficult of full application in the case of family budget enquiries, as some of the families selected might be unable or
unwilling to supply the information required. In the second
method, that of " purposive " selection, a number of groups
of units are selected which together yield nearly the same
characteristics as the totality. This method is also impossible
of complete practical application in the case of family budget
. enquiries. To a certain extent it may be applied in order to
secure data from an adequate number of families of different
size, incomes, occupations, etc. In practice it is in this sense
that purposive selection is a feature of family budget enquiries.
The actual selection of the families is, however, often rather
haphazard. This is the case where the forms to be filled in
by the families are distributed widely and somewhat indiscriminately among families of different income or occupational
groups. In such cases the method includes certain of the
characteristics of both purposive and random selection.
Since the enquiries undertaken by public authorities have
been conducted according to the representative method, the
difference between the intensive enquiries of private investigators and the more extensive enquiries conducted by public
authorities is one of degree rather than of kind. The difference
is largely in the number of samples taken. The distinctive
feature of extensive enquiries is that the attempt is made to
secure information from a sufficient number of families to
ensure that the effects of exceptional cases will be largely cancelled and the results be more representative than those obtained from a few families by the intensive method. In making
this attempt, it is, however, found to be relatively difficult to
obtain information for a long period of time from a large number of families. On the other hand, data for short periods, such
as a week or a month, may be unsatisfactory, especially in
the case of relatively durable articles, e.g. clothing and furniture.
In certain cases a combination of the intensive and extensive
plans has been attempted ; thus, in the Bombay working class
family budget enquiry of 1921-1922, the investigators undertook
certain intensive studies partly for the information obtained
and partly as a means of training in the collection of data
on the extensive plan. In a number of enquiries the attempt
has been made to combine the essential characteristics of

— 12 —
both plans by securing budgets from a considerable number
of families for a period of twelve months, agents of the collecting authority coming into direct and close contact with the
different families.
A special use of the results of family budget enquiries has
become important during and since the war. When the calculation of index numbers showing changes in the cost of living
was begun in a large number of countries, the data obtained
in family budget enquiries were widely used to determine the
relative consumption of different commodities in order to provide weights for the calculation of the index numbers. It is
evidently necessary when measuring changes in the cost of living that greater importance should be attached to a change in
'the price of a commodity which enters largely into consumption
than where the change refers to a commodity the consumption of which is small. Information as to the actual quantities
consumed serves as a useful base in establishing the relative
importance in consumption of the different commodities or
groups of commodities.
When the calculation of index numbers was commenced
in most countries, the conditions were unsuited to the
conducting of new enquiries to furnish the weights necessary,
but instead use was made of data already compiled. In certain
cases, however, special enquiries have been conducted largely,
if not entirely, for the purpose of providing weights by means
of which to calculate index numbers of the cost of living. This
is true of the Belgian enquiry of 1921, the Irish enquiry of
1922, and the Bombay Labour Office enquiry into European
consumption in Bombay conducted in 1923.
When enquiries are conducted for the special purpose of
providing weights for use in the calculation of cost-of-living index
numbers, information with regard to expenditure is all that
is required. Also, satisfactory weights may be obtained from
data supplied by a comparatively small number of families for
a short period.
In the above historical survey, some of the problems to be
considered by those who undertake family budget enquiries
have been indicated. In the following chapters these and
other problems are examined under the general headings of the
collection of information, the scope of information, and the
compilation of results.

II
COLLECTION OF INFORMATION

SELECTION OF THE

FAMILIES

Whichever method of investigation is adopted, it is necessary
that the families should be selected in such a way as to be representative of other families for which information is not obtained.
Owing to differences in the incomes and habits of consumption
of different classes of any community, averages based on information supplied by families with widely differing standards of
living are of little value. It is preferable, not to attempt to
include all classes of the community in one general average,
but to consider separate classes. This course has generally
been followed in the budget enquiries conducted in different
countries.
The class or classes selected for study may be defined by
various criteria. One criterion is the level of income. In this
case it is assumed that families whose incomes fall within given
limits have similar standards of living. Therefore the budgets
of a small number of families with incomes falling within the
defined limits may be regarded as representative of other
families in the same income class. It was with the object of
securing data for a class whose standard of living was reasonably
homogeneous, that the households covered by the New Zealand
enquiry of 1910-1911 were to be those whose annual incomes
did not exceed £250.
A second criterion is the nature of the occupation or the type
of labour contract of the head of the family. According to this
criterion the class studied may consist of large groups, e.g. all
wage earners or all salaried employees. Smaller groups are,
however, often distinguished, e.g. industrial wage earners,
agricultural wage earners, skilled or unskilled workers in parti-

— 14 —
cular industries 1 (shipbuilding, coalmining, etc.). The assumption is made that the standard of living of the families of persons in the same occupation or with the same type of labour
contract will be sufficiently similar to permit of the budgets of
a small number of families being regarded as representative of
the class as a whole.
Since manual workers and their families constitute a large
proportion of the total population, and because interest frequently centres on their standard of living, a number of enquiries have dealt with the family budgets of manual workers,
and especially of industrial workers in urban centres. This
was the case, for example, in the British enquiries of 1904 and
1918, in the Irish enquiry of 1922, the Osaka enquiry of 19191920, and the Bombay Labour Office enquiry of 1921-1922. In
various Swiss enquiries budgets have been obtained from working class families, from the families of lower-grade officials,
and also from those of higher-grade officials. Separate averages are given for each class. In 1916-1917 the United
States Bureau of Labour Statistics undertook a family budget
enquiry covering workers in the chief shipbuilding centres of
the country. In 1923 the United States Department of Agriculture commenced a series of studies on the living conditions
and cost of living among farm families in various localities. In
Sweden, in 1920, information was collected with regard to the
family budgets of agricultural workers and of artisans in rural
areas. In some of these enquiries locality as well as occupation
was taken as a criterion for determining similarity of living
conditions.
A rough combination of the criteria of income and occupation has sometimes been the basis of selection of the families.
In the United States enquiry of 1918-1919, data were collected
from families of wage earners and salaried workers. No limit
of income was fixed for wage earners, it being apparently considered that the wage contract was a sufficient guarantee of
similarity of standard of living. For salaried workers, however,
a limit of $2,000 per annum was fixed and only families of those
with incomes below this figure were to be included. In the Ger1

The families will be restricted to those of workers in a given industry
only if some purpose definitely associated with that industry is in view.
The results would not be regarded as typical for the families of workers
in other industries, and generalisations applicable to all workers could
not be made from them.

— 15 —
man enquiry of 1907, data were obtained with regard to the
families of manual workers, lower-grade officials and teachers.
Similarly, in the Swedish enquiry of 1913-1914, and in the Belgian enquiry of 1921, data were obtained both from working class
and lower middle class families. The Viennese enquiry of
1912-1914 was restricted to the families of workers and those
in a similar economic and social position. In these cases it
is considered that, largely owing to similarity of income, manual
workers, lower-grade officials and clerks in private industry
may be combined into one class from the point of view of standard of living.
Selection has been made in a few cases on racial grounds.
Thus, the United States enquiry of 1918-1919 was restricted
to white families. The Bombay Labour Office enquiries of
1921-1922 and of 1923 dealt respectively with the habits of
consumption of working class families (natives) and of European
families in Bombay.
Of the two chief criteria indicated above — namely, income
and occupation — the former is in some ways more definite
than the latter. It is, however, difficult of application except
on broad lines. Where the results of an enquiry are required
to provide weights for the calculation of cost-of-living index
numbers of general application, averages based on budgets
collected from families whose incomes and occupations cover
a fairly wide range will give satisfactory results. Where the
information collected is to provide weights for cost-of-living
index numbers for different classes, e.g. unskilled workers,
skilled workers, officials, or where it will be used for the drawing of detailed conclusions with regard to standards of living
of different classes, the best course appears to be to collect
budgets over a wide range of incomes and occupations, and then
to classify the results into clearly defined groups as required.
In addition to selection on the lines considered above, certain special features call for attention. Thus in a number of
enquiries conditions are laid down with regard to the composition of the families to be selected. It is recognised that habits
of consumption differ according to size of family, while the presence of boarders or lodgers may disturb the comparability
of the results. In the United States enquiry of 1918-1919 the
minimum family to be selected was to consist of husband,
wife, and one child. Further, the families selected were not
to have boarders nor more than three lodgers, while at least

— 16 —
75 per cent, of the family income was to come from the chief
breadwinner or others who contributed all their earnings to
the family fund. In the British enquiry of 1918, the families
selected were to be of ordinary size, i.e. of four to nine members.
Households with lodgers, or where members received food
as part of their wages, were to be excluded. In the Bombay
Labour Office enquiry of 1921-1922, the families to be selected
where to consist of husband, wife, and children. Information
was, however, also obtained separately for men living in towns,
singly or in groups, away from their homes.
DURATION

OF

ENQUIRY

Considerable variation exists with regard to the period during
which the records of income and expenditure are kept. In
certain enquiries the families have been asked to keep records
for a period of one year, whereas in others the results have been
based on data for one month or even less.
It is evidently more difficult to secure a given number of
budgets for a long than a short period. Fewer families are able
or willing to undertake the task of recording in detail for a
long period the items of their incomes and expenditure, while
during the course of a lengthy enquiry a number of those who
undertake to keep records fail, for various reasons, to complete
them. To a certain extent, therefore, enquiries covering a
long period may be undertaken more readily in the case of the
intensive method than where the extensive method is adopted.
In fact, records covering a long period appear essential in the
case of the intensive method if the results are to be at all satisfactory. It will be seen in the table on page 19, giving the length
of the period covered by various enquiries and the number
of budgets obtained, that, in certain enquiries in which the extensive method has been used, budgets covering a year have
been obtained from a considerable number of families. Thus, in
Sweden, 1,355 budgets were compiled for 12 months in the
enquiry of 1912-1913, while in the German enquiry of 1907,
which covered 12 months, 852 budgets were obtained. It is
nevertheless evident that there is more difficulty in securing
a large number of budgets for a long than for a short period.
It is partly in order that a large number of budgets may be
obtained, therefore, that in some countries enquiries have been
conducted for short periods only. In such enquiries little diffi-

— 17 —
culty arises in the case of income and items of expenditure
which occur at regular and frequent intervals. Thus, an enquiry covering a period of one month will generally give adequate information with regard to wages paid on a weekly
basis or income from boarders or lodgers who make their payments every week, fortnight, or month. The results will also
be satisfactory in the case of expenditure on articles which are
purchased at frequent intervals. Thus for bread, milk, meat,
and most other items of food, expenditure is daily in character
and a short enquiry will furnish good results 1 . Similarly for
rent, where payments are made each week or other short period
a short enquiry may be adequate.
Difficulty arises with commodities of a durable character
which are purchased at irregular and infrequent intervals. For
example, articles of clothing generally wear for several months
or in some cases for a year or more. For such a group of commodities an enquiry of a month's duration would appear to
be inadequate if the budget records include only the items of
income and expenditure which occur during the month. Thus,
certain families may have incurred no expenditure for suits
of clothing, overcoats, etc. Other families, however, may
have purchased such articles during the month of the enquiry,
and consequently if budgets are obtained from a large number
of families the results may be satisfactory. It should be noted
that some articles are purchased at certain seasons only. The
budgets will show an expenditure on such articles relatively
too high or too low according as the enquiry is conducted at a
time when these commodities are or are not being purchased.
The difficulty can be avoided either by holding the enquiry at
a season when expenditure on the article under consideration
is intermediate, or by making enquiries at different seasons
and averaging the results.
The difficulty of durable and seasonal articles has been
partly avoided, in certain countries which have conducted
short-period enquiries, by the method of asking for estimates
as to annual expenditure on such articles. This was done,
for example, in the Irish enquiry, which covered one week only
in June 1922, and in the Bombay working class budgets
1
On account of seasonal variations in expenditure on food, the enquiry
should be conducted at a season when average conditions prevail.
Alternatively, several enquiries at different seasons might be conducted
and the results averaged.

2

— 18 —
enquiry in which records were obtained for one month in 19211922. In the Irish enquiry, the families supplying the information were asked to record the expenditure on all commodities
purchased during the week covered. In the case of clothing,
fuel, and light, however, they were also asked to give as accurately as possible the cost of items in these groups purchased
during the twelve months from July 1921 to June 1922. For
articles which usually last more than twelve months, the proportionate cost for the twelve months was to be stated.
Thus, in the case of an overcoat which usually lasts three years,
only one-third of the cost was to be entered on the form. Where
the exact amount of money spent on certain items, e.g. stockings, could not be remembered, but where it was known that
the article lasts about four months, then three times the price
of the last article purchased was to be recorded. Similarly
for fuel and light, householders were asked to estimate their
expenditure on different items during the preceding twelve
months, together with the quantities of each consumed.
In the Bombay Labour Office enquiry a separate part of
the form distributed was provided for the recording of expenditure on clothing, shoes, and other necessaries not bought
every month. Those furnishing information were asked to
state the number of such articles in use and the cost of each
article when bought, together with the total cost. They were
further asked to estimate the number of months that each
article would last and its average cost per month.
Evidently estimates on the lines indicated above are less
reliable than information obtained from records kept throughout the year. The most accurate results are likely to be
obtained if an adequate number of budgets can be secured
for a period of twelve months K Satisfactory results may,
however, be obtained by means of shorter enquiries on the lines
indicated. Alternatively, information with regard to income
and food, rent, fuel, light, and other items of regular and fre1
An objection which has been raised against records covering a long
period is that the working class families able to keep budgets for twelve
months are not fully representative. The keeping of records for such
a period implies a certain degree of intelligence and a systematic method
of recording income and expenditure. Families in the lowest class of
the community would be unable to keep such records. Also the keeping of accounts regularly is more likely to be undertaken by people
whose economical habits are above the average than by those who are
extravagant and careless.

— 19 —
quent consumption might be obtained from a large number of
families for a period of one month, supplemented by complete
budgets kept for twelve months by a smaller number of families.
The difficulty with regard to seasonal variations in consumption
could be avoided by conducting two or preferably four shortperiod enquiries distributed suitably throughout the year.
LENGTH OF PERIOD COVERED BY VARIOUS FAMILY BUDGET ENQUIRIES,
TOGETHER WITH THE NUMBER OF SATISFACTORY BUDGETS OBTAINED

Enquiry in

Australia

Date

1919-1920
1917
1918-1919

1 month
12 months

43
82

1910-1911

12 months

69

1918-1919
1913-1914
1912
1921
1904
1918 5

12 months
12 months
12 months
12 months
Food and rent : 1 week
All groups except clothing : 1 week
Estimates for 12 months

Austria (Vienna)
Belgium
Denmark
Germany
India (Bombay)

1922
1907
1921-1922

Irish Free State

1922

United Kingdom
United States

Number of
budgets
obtained

212
12 months
1 month
392
12 months
119 !
Food : 14 days
)
Other groups: estimates
848
for 12 months
12 months
352 a
12 months
852
1 month ; for certain
groups estimates were
made
2,473 3
Income, food, etc. : 1 Ì
week
308
Clothing, fuel a n d light: >
estimates for 12 months
12 months
99

1910-1911
1913
1912-1914
1921

J a p a n (Osaka)
Netherlands (Amsterdam)
Manual workers 4
Non-manual workers
New Zealand (4 chief centres)
Norway (Christiania and
Bergen)
Sweden
Switzerland

Length of period covered

1918-1919

82
1,355
791
323
1,944
1,306
12,096

' Seventy of these families continued their budgetary records for a second year.
Altogether 62i budgets were collected, but a number were incomplete without supplementary enquiries.
a
Also 603 budgets of men living away from their families.
* One of the objects of this and subsequent enquiries was to indicate changes in the
cost of living.
5
Data obtained for Great Britain only.
1

It may be added that in certain cases — for example, Switzerland and the Netherlands (Amsterdam)— enquiries have been
conducted at more or less regular intervals and attempts
made to ensure the comparability of the information collected

— 20 —
during the different periods. Thus, in Switzerland a succession
of enquiries has been conducted from the year 1919 onwards,
each enquiry covering a period of twelve months. In Amsterdam, from the year 1917 to 1922, a series of enquiries was
conducted generally at intervals of one quarter, the budgets
covering a period of four weeks. The results were used in this
case as a means of indicating changes in the cost of living.
The table on page 19 gives for a number of important enquiries
the length of period covered by the budget records, together
with the number of budgets obtained. In drawing conclusions
as to the adequacy of the number of budgets secured in different
enquiries, account should be taken of differences in the scope
of the enquiries. Some cover only a single town, while for
others the results are regarded as representative for countries
of different size.
METHODS OF SECURING THE BUDGETS

As family budget records are generally made voluntarily,
the first necessity is to secure as far as possible the confidence
and interest of the householders.
In securing the co-operation of the different families, it is
usually indicated that the information supplied by each family
will be treated as confidential and that only average results,
based on data supplied by a number of families, will be published. Also in some enquiries it is stated specifically that the
information supplied will not be used for taxation purposes.
In some enquiries of an extensive character it is not known
beforehand how many families will be likely to supply information. In such cases forms or account books are distributed
widely with the request that records should be kept and the
results forwarded to the authority undertaking the investigation.
Interest in the enquiry may be stimulated by means
of notices published in the newspapers indicating the objects
of the investigation, by securing the co-operation of the trade
unions and employers' federations, and by propaganda undertaken by agents of the department conducting the enquiry.
Thus, in the Australian enquiry of November 1913, notices
were inserted in the daily press throughout the Commonwealth
directing attention to the enquiry and indicating its purpose
and scope. Also the Bureau of Census and Statistics communicated with trade and labour councils, trade unions, employers'

— 21 —
federations and associations and with newspaper proprietors in
the different States inviting their co-operation in the enquiry
and their assistance in distributing the account books. In
the enquiry conducted in Western Australia in 1917-1918, an
attempt was made to stimulate interest by the appointment
of an organiser to give lectures and addresses in all the principal
centres of the State and to bring the importance of the enquiry
before all classes. In the British enquiry of 1904, forms were
sent to a large number of members of workmen's organisations,
co-operative societies, and individuals who were asked to furnish the desired information or to obtain particulars from fellowworkmen. In the British enquiry of 1918, the distribution
of the forms was entrusted to the voluntary services of trade
unions, co-operative societies, medical officers of health, etc.
The number of budgets secured by the method of distribution
through the intermediary of voluntary organisations has often
been disappointing. Thus, in the first budget enquiry conducted
in Australia, covering the twelve months ending 30 June
1911, the number of books distributed was 1,500. Of these
only 212, or 14 per cent, of the number distributed, were available for the compilation of the results. It was thought that
the smallness of the number was due to the labour involved
in keeping records for twelve months. Consequently, in the
second enquiry the period for which the information was asked
was limited to four weeks in the hope of securing a much larger
number of budgets. Nevertheless, although the enquiry of
November 1913 was given wide publicity and 7,000 books
were sent out for distribution, only 417, or less than 6 per cent.
of the total despatched from the Bureau of Census and Statistics, were returned, and of these 25 were unsatisfactory for
various reasons. Again, in the enquiry conducted in Western
Australia, notwithstanding the efforts of the agent who spent
a month in touring the State to stimulate interest, the response
was " utterly disappointing, the householders sending in
returns being too few in number to admit of any safe conclusions
being drawn ". The total number of families which sent in
returns that could be tabulated was only 66, and the information
supplied by some of these was for periods shorter than the
thirteen weeks for which data had been asked. In the case
of the British enquiry of 1904, data adequate for tabulation
were obtained from 1,944 families.
In comparing these
numbers with those obtained in the Australian enquiries, it

— 22 —
should be remembered that particulars for one week only were
asked for, and the enquiry was further limited to income and
expenditure on food and rent. In the British enquiry of 1918,
in which information as to expenditure during one week only
was requested the number of forms distributed was 10,000, of
which 1,306 were satisfactorily filled in. In the New Zealand
enquiry covering twelve months in 1910-11, nearly 1,800 account
books were distributed, but only 69 satisfactory budgets were
obtained. In the further enquiry of 1919, although the period
of the enquiry was reduced to six months, the results were
disappointing ; for, although several thousand books were
printed, only 109 returns of value were returned to the Census
and Statistics Office.
A method of securing information which has in some respects
been more successful than the indiscriminate distribution of
forms or books through voluntary organisations is that of
appointing agents who secure the collaboration of the families
and personally distribute the forms or books to the householders. In such cases, the agent secures the promise of cooperation of various families before the enquiry commences and,
except for a few records which may be insufficiently accurate for
use, the number of budgets which will be available for the final
compilations is known before the enquiry is completed. Agents
also keep in touch with the householders during the period
covered by the enquiry and assist them in the filling in of
the data.
Agents have also been employed in the case of
enquiries not based on the keeping of daily records throughout
a given period but on estimates supplied by the householders.
By checking the information supplied and by asking for explanations of exceptional or doubtful figures, the value of the
information supplied is increased. The employment of agents
is more costly, however, than the method of distribution by
means of voluntary organisations.
An extensive investigation conducted by means of agents
was the United States cost-of-living enquiry of 1918-1919.
By personal visits to the homes, the agents obtained data in
interviews with the wife or other members of the family, 12,096
budgets being obtained. In the Bombay enquiry of 19211922, the information was collected by investigators of the
Labour Office, who visited each home a number of times.
Indian men, and also women, of good education and with
other suitable qualifications, such as experience in indus-

— 23 —
trial welfare work, were mainly selected as agents. As stated
above, before the enquiry proper was undertaken, investigators
were requested to secure one or more budgets as a means of
training. The Viennese enquiry of 1912-1914 was conducted
with the help of agents, mostly women, who were instructed to
assist the families in the filling in of the forms, to check the
information supplied and to make supplementary enquiries
where necessary. For this purpose the agents kept in touch with
the families at frequent intervals during the course of the
enquiry. In the Amsterdam enquiry of February-March 1917
the names of various families willing to collaborate were obtained from the offices of different Socialist trade unions *.
Before making the investigation the heads of the families and
their wives were asked to attend a meeting at which the method
of conducting the enquiry was discussed. During the course
of the enquiry, three officials of the department conducting
it remained in contact with the collaborators during the whole
period in order to assist in the keeping of the records. In
Belgium and other countries agents of the central Government
have assisted in the enquiries by selecting the families and
helping them with the records. In the Irish enquiry of 1922,
forms were sent to National School teachers in every school in
the country, who acted as agents and assisted in the filling in
of the forms and in checking the reliability of the details
recorded.
As regards the actual recording of the information, this may
be done either by agents or by the householders themselvesWhere agents enter the data on the form, the statistics are
based on what the householder and his wife remember as to
the amounts of income and the quantities and prices of articles
purchased or on receipts and accounts in the possession of the
family. Where the householder records the information, the
data may be based either on what is remembered as to the
income and expenditure and on receipts and other accounts
or on detailed records.
The method of collecting information based on what the
householder remembers as to expenditure on different items
has certain advantages, especially in the case of durable articles
the purchase of which is relatively infrequent. Data covering
1
This was mainly to secure budgets from the same families, as far aspossible, as those which had supplied information in 1911 in connection
with the investigation conducted by the Socialistische Siudieclub.

— 24 —
a considerable period of time can be obtained without the
necessity of waiting during a period in which the householder
keeps detailed records *. Thus information of reasonable
accuracy may be obtained as to expenditure on clothing simply
by asking the head of the family to state what articles have been
purchased during the last twelve months, together with their
cost. If the agent records the data himself, a considerable
degree of uniformity as to the nature of the information supplied by the different families may be introduced. The forms
on which the information is recorded are of a summary character
and the results for a number of families are more readily computed thanwhere more detailed records are kept. A disadvantage,
already indicated, is that the information is likely to be less
reliable if based on what the householder remembers than if
detailed records are kept.
To avoid this disadvantage, in some enquiries householders
have been supplied with account books with a page for each
day covered by the investigation. On these they enter all the
items of family income and expenditure as these occur. Summaries are then made either by the householders or by agents
who, as stated above, are often appointed to keep in touch with
the families during the enquiry. In the Swiss, Viennese, and
other enquiries, forms for each day during the period covered
were distributed to the families who undertook to supply
information, while in the Swedish enquiry of 1913-1914 the
account book included, for expenditure, pages for each week
with columns for each day and also for a weekly summary.
In Australian and other enquiries similar books were provided 2.
In some enquiries a combination of the method of securing
estimates and that of keeping detailed records has been adopted,
detailed records being kept for food, etc., and estimates given
for items of durable character.
It may be added that in a few cases the keeping of budgets
has been accompanied by some small pecuniary acknowledgment. In the Amsterdam enquiry of March 1922, the wives
1
As has been already noted, for food and other items the purchase
of which recurs frequently, records for a short period give satisfactory
results. In such cases actual records should always be kept instead
of relying on the memory of the householder.
a
Specimens of the forms used in different enquiries are given in
Appendix II.

— 25 —
who collaborated received a small payment for the work involved. In the Osaka enquiry of 1919-1920, each family
keeping records was at several dates given admission tickets
to a variety show or a moving picture theatre and, at the end
of the enquiry, which extended over a period of twelve months,
a piece of cotton cloth. In the New Zealand enquiry of 1919,
prizes were offered for the most accurately compiled returns.

Ill
SCOPE OF INFORMATION

The information which the householder should be asked to
supply will vary according to the object of the enquiry. Where
the object is to obtain statistics as to general conditions and
standard of living, detailed questions of a general character,
including some regarding the composition of the family, together
with questions as to income and expenditure are necessary.
If the enquiry is being conducted to provide weights for the
calculation of cost-of-living index numbers, questions in some
detail as regards expenditure together with a few general
questions may be adequate. In the sections which follow,
indications are given of the more important questions asked in
enquiries which have been conducted in different countries,
special attention being devoted to the more complete investigations. For additional details, reference may be made to
Appendix II which reproduces the forms distributed in a number
of enquiries.
GENERAL

CONDITIONS

Many enquiries not confined to some special district ask
for the name of the town, county, or other district in which
the family is living. This information enables the families
to be classified according to locality so that comparisons may
be made of the conditions in different districts. Similarly,
if the head of the household is asked to state his industry and
occupation, the results of the enquiry may be classified into
industrial or occupational groups.
With regard to the composition of the household, details
for each member of the household are requested including their
relationship to the head of the household, their sex and age.
In some enquiries, instead of asking for the age of each person,
information as to age is requested only for persons under a

— 27 —

given age, e.g. 18 years of age. In others, even less details
are requested, for example, the number of men, the number
of women, and the number of children under a given age.
Members of the household include, in addition to members
of the family, any persons who have meals or rooms in the household, e.g. boarders, lodgers, servants, and resident visitors 1 .
For such persons, the extent to which they participate in the
consumption of the household is to be indicated. Persons
living in the household for part only of the period covered by
the enquiry are generally to be included and the length of
period during which they formed part of the household stated.
Similarly any absences of members of the family are usually
to be recorded.
For detailed enquiries information as to the composition of
each family is necessary as the family, being a variable " u n i t " ,
does not provide a suitable basis for comparisons. The consumption of two families of different size can be compared only
if the membership of each is known. Even if the average
consumption of a large number of families is calculated, it
is necessary, from the point of view of the standard of living,
to know the average size of the families covered. The most
satisfactory course is to reduce families of different size to
terms of a common unit, such as an adult man, or a family of
given size, for example, husband, wife and three children aged
15 years, 9 years, and 4 years respectively. The method adopted for this purpose is outlined in a later section. In applying
it, however, details as to the composition of each family, and
the age and sex of each member are essential.
In most enquiries, questions with regard to housing are
asked. The object of such questions is partly to obtain information as to housing conditions, and partly to complete the
information on income and expenditure. It is very desirable
that data for the latter purpose should be obtained, for example,
to show whether the family lives in a house for which rent is
paid, or is living rent free in its own house or otherwise. This
question is considered later. Among the questions for securing
information on housing conditions are those with regard to
size of dwelling, e.g. the number of rooms and the amount of
1
Some enquiries ask whether domestic animals are kept, for example,
dog, cat, rabbits, hens. This information is invited partly on account
of the cost of keeping them, but partly because, in some cases, they are
also a source of revenue.

— 28 —
supplementary accommodation such as pantry, cellar, bathroom, and washhouse. One of the most detailed investigations
from the point of view of housing conditions was the Austrian
enquiry of 1912-1914 which asked for the dimensions of each
room, whether they were heated or not, the number and height
of the windows, whether the dwelling had its own water supply,
etc. In the Swedish enquiry of 1913-1914, in addition to questions as to the number of rooms and annexes, the householders
were asked to state whether the dwelling was provided with
central heating, electric light, gas for cooking, and water supply.
The United States enquiry of 1918-1919 also entered into considerable detail with regard to the quality of housing accommodation.
Certain enquiries have included also questions with regard
to family history, the physical condition and education of its
members, the condition of the furniture, clothing, etc. The
Austrian enquiry should be specially mentioned for the details
on these lines regarding which information was requested. It
even went so far as to ask for an inventory of furniture and
clothing. Such details may be regarded rather as supplementary to, than essential for, the general purpose of family budget
enquiries.

INCOME

With regard to income, it is important to know the sources
from which the family revenue is derived. It is of interest to
know how total income compares with total expenditure, and
thus to see to what extent savings are being effected. From
this point of view it should be noted that indications as to income may be incomplete, especially in the case of income derived from subsidiary sources. Either through carelessness, or
because of deliberate omission with the object of concealing
part of income, the recorded total may be less than the actual
total. This may explain some of the cases shown in different
enquiries of an excess of expenditure over income for certain
groups of families. Information as to income is also of value
to enable the results of an enquiry to be classified according to
income groups.
Usually the earnings of each member of the family are to be
recorded separately. These data are of interest as showing to

— 29 —
what extent the family is maintained by the chief wage earner,
and the relative importance of the earnings of the wife and of
the children. In some enquiries — for example, the Swedish
enquiry of 1913-1914 — in the case of the husband's earnings,
a distinction was drawn between earnings from ordinary employment and subsidiary earnings. Earnings cover both payments in money and payments in kind. In the casé of payment
in kind its approximate value should be stated, e.g. for free
dwelling, lighting, fuel, vegetable garden, board or other food,
clothing, travelling, etc.
Information with regard to any supplementary source of
income is necessary. Receipts for old age or other pensions
should be given, as should also income from insurance funds.
Income from persons who are provided with board and lodging
in the household is important 1 . Where there is a vegetable
garden, or where domestic animals are kept, the value of any
produce should be stated. Theoretically also, the value of
services within the household rendered gratuitously by different
members of the family, e.g. cooking, laundry work, etc., should
be included as income and balanced by equivalent amounts on
the expenditure side of the budget 2 . It is, however, impossible
to estimate the value of such unpaid services and, consequently,
in practice they must be omitted. The effects of differences
in various countries in standards of living and habits with
regard to free services within the household interfere to a greater
or less extent with the international comparability of the results
of family budget enquiries. Similarly the value of various services rendered freely by public authorities, e.g. education, is not
recorded in family budgets. Differences in the value of such
1
The housing accommodation provided for lodgers or sub-tenants
implies a higher rental for t h e household t h a n where no lodgers or subtenants are included. Differences in the percentage which rent forms
of t h e total expenditure of different families m a y be accounted for by
differences in habits with regard to t h e provision of accommodation
for lodgers. This m a y be especially true as between different countries.
For strict comparability, the income derived from lodgers and subtenants, or a large p a r t of it, should be deducted from t h e total rental
to give a basis of comparability for expenditure on rent. In t h e United
States enquiry of 1918-1919, families were not to be selected who had
sub-tenants other t h a n lodgers, while the number of lodgers was n o t
to exceed three. Families with boarders were not to be selected.
2
This point was referred to by Prof. Wagemann during the discussion
on International Comparisons of Rela "Wages at the Second International Conference of Labour Statisticians. See INTERNATIONAL LABOUR
O F F I C E : The Second International Conference of Labour
Statisticians;
Studies and Reports, Series N (Statistics), No. 8, p. 24. ; Geneva, 1925.

— 30 —
services in various countries impair the international comparability of the results of budget enquiries. Receipts from
investments, and also the value of presents received, whether
of food, clothing, or other commodities, should be recorded.
EXPENDITURE

The information required with regard to expenditure is of
necessity more detailed than that on income. The various
items on which the families expend their income may be combined into a small number of fairly clearly defined groups —
for example, food, housing accommodation, clothing, fuel and
light — and also miscellaneous items of expenditure which may
be combined together either into one general group or into
several special classes or sub-groups.
In most family budget enquiries the forms or books in which
the information is to be recorded specify each of the most important items of expenditure, these being classified into groups
on the lines indicated in the preceding paragraph. Space is
provided for the recording of expenditure on other items in
each group as well as those specified. In a few enquiries, however, blank forms are given in which each item of expenditure
is recorded in the order in which it occurs, no attempt being
made to classify the items. Forms are provided for each day,
and from these records the final results are compiled and the
items classified into groups by the authority conducting the
enquiry. There appear to be advantages in using forms on
which the chief items are specified. The recording of unnecessary details is avoided and the work of compiling the
results facilitated.
The problem of the grouping of the items may be considered
either in connection with the information to be obtained or
with the compilation of results. Where the items are specified
and grouped in the forms distributed the problem is closely
connected with the information to be obtained, and in the
compilation of results the grouping follows automatically that
determined by the schedules. If the different items of expenditure are recorded by the householders in the order in which
they occur, the problem of grouping is encountered when the
results are being compiled. In the present study the problems
of grouping are considered mainly in connection with the
compilation of results.

' — 31 —
For each item of expenditure, as far as possible, householders
are, in most enquiries, asked to state both the quantity or number of articles purchased and the cost or expenditure. In the
case of certain items of expenditure, of which housing accommodation is the chief, expenditure only can be given in practice. For all items where it is possible, both quantity and cost
should be given in detailed enquiries. It is especially necessary
to do so for articles the quality of which varies and which
cannot be clearly defined. Where the object of the enquiry
is to provide weights for the calculation of the cost-of-living
index numbers the relative importance of the different items
can be determined if only expenditure on each is given. Even
for this purpose it is preferable if quantities are also stated,
especially for the food group.
In the case of food, enquiries differ considerably in the degree
of details requested. Some specify a small number of items
such as bread, milk, flour, meat, fish, fruit, vegetables, etc.
Others ask for greater detail and divide meat, fruit, vegetables,
and other similar sub-groups into a number of separate items.
Thus in the Swedish enquiry of 1913-1914 separate information
was requested for the following items of meat : beef, veal, mutton and lamb, pork, other fresh meat, preserved meat (beef
and pork), sausages, etc. In the case of bread, information
was requested with regard to hard rye bread, soft rye bread,
wheaten bread, other table bread, biscuits, sweet bread, cakes,
etc. Other groups were similarly sub-divided.
In the United States enquiry of 1918-1919 even more details
were requested, altogether 128 separate articles of food being
specified. It seems desirable that separate data should be
obtained for various items of meat, vegetables, and other
groups of food. Also, in the case of meat, in countries where
frozen or chilled meat is consumed, these varieties should be
distinguished from fresh meat. Sausages should be given
separately and not combined with other meat, as often a given
weight includes a considerable proportion of other ingredients.
Where different qualities of bread are consumed, the chief
kinds should be specified ; especially should the distinction be
drawn between wheaten and rye bread. The details will differ
somewhat from country to country according to habits of consumption. It should be added that the cost of meals bought
and consumed away from home should be recorded as a
separate item.

-ï=SS55="

— 32 —
In the case of fuel and light, separate data should be given
for coal, coke, wood, gas, oil, electricity and candles in countries in which their consumption is important. In the United
States enquiry of 1918-1919, in recording expenditure on
articles of fuel and light, the distinction was drawn between
those used for heating, those for cooking, and those for lighting.
The case of housing accommodation presents special problems.
First, what is to be included ? The most satisfactory course
appears to be to take the rent of an unfurnished apartment,
including charges for water supply and those rates which are
levied specially for the upkeep of roads, sewers, etc., directly
connected with the dwelling. Charges for central heating and
for lighting are to be excluded. All local rates not specially
connected with the dwelling, even though levied in respect
of occupancy, should be excluded. Thus, rates for the maintenance of various local government services, which, in some
countries, are levied in respect of occupancy, are more appropriately included under the heading " Taxation ". It is not
always possible, under certain systems of payment, to separate clearly those charges directly connected with the dwelling
from certain elements of local taxation. In all cases where
the rent of a house includes rates covering such elements,
some indication should be given as to their nature.
A special problem arises in the case of families who pay no
rent because they own the house in which they live. A solution adopted in some enquiries is to ask for a statement of the
rents paid by tenants living in similar dwellings. In other
enquiries, no such statement is asked for, the view being taken
that only actual expenses should be recorded. In these cases the
house owner is asked to state the amount of interest which he
pays in respect of any mortgage on the house, or the amount
which he pays periodically towards the purchase of his house
on the instalment plan. Also any expenses incurred for the
repair of the house, insurance, and other amounts paid on
account of ownership would be recorded. Neither of these
solutions is altogether satisfactory. A more satisfactory
course appears to be either to avoid the problem altogether
by selecting only those families who live as tenants and pay
rent for the houses in which they live 1 or to regard the house
1
An alternative would be to include both tenants and owners, but
in compiling the results to classify them separately.

— 33 —
as capital.
In this case, the income side of the budget would
include an estimate of the revenue which an independent
owner of such a dwelling would receive, while on the expenditure side the rent which a tenant would pay for the dwelling
would be recorded. This method is logically necessary in
order to assimilate the case of a houseowner who lives in his
own house with that of the owner of a house who rents it to
a family while he himself lives in a house which he rents from
a third party. It also effects assimilation with that of a man
who, instead of buying the house in which he lives, continues
to pay rent for it, and invests the sum which would have
been necessary to purchase the house in securities the interest
on which figures in the income side of the budget. Cases of
families living rent free may be dealt with similarly.
Furniture resembles housing accommodation in certain respects ; for example, both are durable in character. There is
one essential difference. Whereas in most countries families
generally live in houses which belong to other people, and
pay rent to the owners for the privilege, comparatively few
families hire their furniture, the great majority of families owning the furniture which they use. Because of the durable
character of furniture, its cost to different families during the
period of an enquiry is likely to be less than that which would
result if the cost of each article of furniture in the household
were distributed evenly over the period of the durability of
the article, and the proportionate part allotted for the period
of the enquiry. To attempt such a distribution would, however,
be impossible in practice even if it were considered desirable.
From the practical point of view, the most satisfactory course
appears to be to consider only expenditure on furniture incurred
during the period covered by the enquiry, and to take no account of furniture purchased previously. Expenditure during
the period of the enquiry would include payments made towards
the purchase of furniture on the instalment plan. If a large
number of budgets are obtained, there will be considerable
compensation between families whose expenditure on furniture is merely for repairs, renewals and additions and those
who incur heavy expenditure during the period of the enquiry
for their initial stock of furniture. Such compensation would
not be complete in practice, largely because those purchasing
furniture for the setting up of a home would often be excluded
as not being typical families.
3

— 34 —
The practice indicated above has been adopted in a number
of enquiries. In the Australian enquiry of November 1913 and
certain other enquiries, the head of the family was asked to
include under the heading " Miscellaneous and Special Expenditure " the cost of renewals and of repairs to furniture. The
Swedish enquiry of 1913-1914, and other enquiries, asked for
the cost of furniture, furnishings, utensils, and crockery to
be recorded in a separate group. In the United States enquiry
of 1918-1919, some sixty articles of furniture and furnishings
were specified, and the number of articles purchased during
the year covered by the enquiry, together with their cost, were
to be recorded. In the Austrian enquiry of 1912-1914, in addition to the inclusion of expenditure on furniture and furnishings in the daily records of expenditure, an inventory was to
be given of the articles in possession of the family. This gave
some indication as to the " capital " which the furniture
represented. It was incomplete from this point of view,
however, as no valuation of the articles was attempted.
With regard to clothing, as was seen above, the question of
the durability of the articles is of some importance especially
in the case of enquiries covering a short period only. Articles
of clothing being, however, much less durable than dwellings
and than most articles of furniture, the difficulty is more easily
overcome. It is unnecessary to repeat here the points already
outlined above in connection with the duration of enquiries.
Suffice it to say that the difficulty has been avoided by asking
householders to record only the proportionate cost for the
period of the enquiry of articles of clothing the durability of
which is greater than that period. A simpler method is that
of trusting to the compensatory effects which result from the
averaging of the data obtained from a large number of families.
This method, which doubtless gives quite satisfactory results
in the case of clothing, has been adopted in most enquiries.
In view of the individual character of expenditure on clothing
some enquiries have asked for details as to the articles bought
and their cost for each member of the family. This information
enables expenditure on clothing to be classified according to
age and sex. The data necessary for this purpose were obtained, for example, in the United States enquiry of 19181919, and to some extent in the India (Bombay) and Irish
enquiries of 1921-1922 and June 1922 respectively.
In the case of other items of expenditure which are often

— 35 —

classified into one " Miscellaneous " group, the chief problems
are the degree of detail which householders should be asked to
record, and the system of grouping to be adopted. The latter
group of problems is dealt with in a later section. As regards
the degree of detail, some enquiries ask for each item of expenditure to be recorded. In a considerable number of enquiries,
however, a comparatively small number of the more important
miscellaneous items is specified. Where this is done, householders are asked to state their expenditure on the items specified, and then, without enumerating the different items, to
give the total amount spent on various items of lesser importance. This course appears to be the more satisfactory. The
variety of items of expenditure is so great that if ail are recorded
a considerable amount of unnecessary detail is introduced,
which renders the compilation of results more difficult and
adds little, if anything, to the value of the enquiry. The minor
items differ considerably from one family to another and the
details for one family are of little general interest as they
are unlikely to be representative of the minor expenses of
other families. The cost of many of the items is also comparatively unimportant and constitutes an insignificant part of total
expenditure. Such items may well be combined instead of
each being specified separately. From the results of enquiries
already undertaken in any country, it would be possible to
decide which items are sufficiently important to be given separately. Thus, apart from a few items of special interest, householders might be asked to give separate data only for items
the cost of which is above a given amount.
Among the miscellaneous items for which separate information has been asked in different enquiries, mention may be
made of the following : tobacco, cigars and cigarettes ; alcoholic drinks ; railway, tram and other fares ; insurance ;
education ; books and newspapers; laundry; medical attendance,
dental treatment, medicine, etc. ; recreation and amusements ;
charities and other gifts; church collections and subscriptions
to societies, etc., including trade unions ; post, telegraph and
telephone ; rates and taxes ; lottery tickets ; interest on debts ;
domestic servants. In the case or railway, tram and other
fares, it is desirable to ask for separate information for expenditure incurred by the head of the family or other member
in proceeding to or from work, and for expenditure on excursions or holidays. In certain enquiries, e.g. the United States

— 36 —
enquiry of 1918-1919 and the Swedish enquiry of 1913-1914,
distinction is made on these lines. In the United States enquiry,
in the case of street-car fares, three sub-headings were given,
namely, rides to work, rides to school, and other rides. In the
Swedish enquiry, the cost of journeys to work and to market
were to be given together ; journeys to school were to be
recorded separately, as were also other journeys. In the case
of insurance, certain enquiries ask for separate information
with regard to life, accident, sickness, fire, old age. In view
of the importance of insurance, development on these lines
appears desirable 1 . In the Austrian enquiry of 1912-1914,
compulsory insurance was distinguished from voluntary insurance.
The United States enquiry of 1918-1919 is exceptional in the
amount of detail asked for in each group. In the miscellaneous
group over sixty separate items were specified. Thus for
sickness and disability seven headings were given, namely :
physician, surgeon, oculist; medicine; nurse; hospital; dentist;
eyeglasses ; other expenses. Six items were specified in the
case of contributions, gifts and dues, namely: church and
other religious organisations ; labour organisations ; lodges,
clubs, and societies ; charity ; patriotic contributions ; gifts
to persons outside the family. The remainder of the schedule
is similarly detailed. In a country like the United States,
where the standard of living is high, consumption is more varied
than in countries where lower standards prevail, and consequently the amount of detail which may be with value recorded
is greater. For the purpose of most enquiries it appears unnecessary to adopt such an elaborate series of questions as that
used in the United States enquiry. On the other hand, in
number of enquiries, very few questions are asked regarding
miscellaneous expenditure, and it is probable that in such cases
an appreciable amount, of expenditure is unrecorded.

1
The recording of the'amounts paid in the case of compulsory insurance schemes is rendered [difficult by the fact that the payments are
generally retained out of wages and not made directly by the worker..

IV
COMPILATION OF RESULTS

Where very detailed records are kept by householders who
inscribe daily each item of income and expenditure, the preliminary work of compilation may be undertaken by the householders themselves. They may be requested to make weekly
or other summaries, special provisions being made for this
purpose in the forms. Where agents in the different districts
assist in conducting the enquiry they may be charged with the
duty of preliminary compilation. The systematic checking
of the data, the work of tabulation and the calculation of averages lies, however, with the central department responsible for
the enquiry.
In a number "of enquiries the chief features of each budget
obtained have been published — for example, in the Amsterdam enquiries, the Austrian enquiry of 1912-1914 and the
Norwegian enquiry of 1917-1918. This course is evidently
practicable only in cases where the number of budgets obtained
is comparatively small. Even where each budget is published
separately, however, averages have also been calculated, while
in enquiries in which the number of budgets collected is large,
the only satisfactory method of presenting the results is to
calculate averages. Further, since the main object of collecting a considerable number of budgets is to cancel out exceptional
features in the consumption of individual families, averages are,
for most purposes, of greater value than individual budgets.
CLASSIFICATION

In presenting the results of enquiries, general averages
covering all the families are usually calculated. It is desirable,
however, if an adequate number of budgets is obtained, for
separate averages also to be given for different divisions of the

— 38 —

total number of families. In many enquiries, separate averages are given for the families living in different districts. Thus,
in the United States enquiry of 1918-1919, data of income and
expenditure are tabulated for the different towns or other
districts. Similarly, in the Swedish enquiry of 1913-1914,
data are given for six separate divisions of the country and also
for a considerable number of separate towns. In the Danish
enquiry of 1922 separate data are given for Copenhagen, for
provincial towns, and for rural districts. The results of the
Swiss and certain other enquiries are also tabulated by locality.
In a number of enquiries, however, only general averages for
the whole country are given. This course was taken, for
example, in the case of the British enquiry of 1918. It is unnecessary to calculate the results separately by districts where
the object of the enquiry is to obtain weights for the calculation
of national index numbers of the cost of living.
In addition to classifying the results according to the geographical distribution of the families, the results of many family
budget enquiries are given by grade of family. Thus, in a
number of enquiries — for example, the Swiss enquiries —
separate averages are given for the families of state officials
and other non-manual workers, for those of skilled and for
those of unskilled manual workers. In the British enquiry
of 1918, the Norwegian enquiry of 1918-1919, and others
data are given separately for families classified into groups
somewhat similar to those mentioned for Switzerland. In
other enquiries the families are classified according to their
income or expenditure. For example, in the United States
enquiry of 1918-1919, data are given separately for families
with the following annual incomes :
Under
#900 and
#1,200
#1,500

#900
under #1,200
„
#1,500
„
#1,800

#1,800 and under #2,100
#2,100 „
„ #2,500
#2,500 „ over.
All incomes.

In the Bombay Labour Office enquiry, the families are classified separately according to income groups, differing by 10 rupees
per month, from incomes below 30 rupees per month to those
of 90 rupees and over. The Swedish enquiry of 1913-1914,
the Austrian enquiry of 1912-1914, and others have given
separate averages for families within different income or expenditure groups. In the Swedish enquiry, in addition to the

— 39 —
classification of all families by expenditure, separate classifications, by expenditure are given for manual workers and for
non-manual workers 1 . In the Swiss enquiries, classification
by income groups is given along with that based on the grade
of worker.
Where the object of the enquiry is to give detailed information
as to the standard of living of different sections of the community, it is desirable that such separate data for groups based
on income or on category of worker should be published.
Classification by income is in some respects more exact than
that according to grade of worker. Taken alone this method
of classification is insufficient. The income of a subordinate
clerk may be the same as that of a skilled manual worker.
By tradition and custom, however, non-manual workers, even
though their income is small, distribute their expenditure differently from manual workers. If, therefore, satisfactory comparisons are to be made between the expenditure of different
groups of families on various items of expenditure, it would
seem necessary to combine classification by income groups
with classification into the two groups of manual and nonmanual workers.
As in the case of geographical distribution, no separate
classification by income or category of worker is necessary if
the results are to be used for the compilation of a single cost-ofliving index number. In some countries, however, e.g. Belgium and Switzerland, separate index numbers are calculated on
the basis of the consumption of different categories of workers.
Thus, in Belgium, separate indices are calculated for four
categories of working class families, falling within different
income groups, and also for lower middle class families. In
order to provide the weights necessary for the calculation of
these separate indices, the results of the 1921 enquiry were
classified into four groups of workers' families and one group
of lower middle class families.
A further classification of results is that according to size
of family, averages being given for families classified by size
as well as for all families for which budgets are obtained. This
is desirable in any detailed study, as the conditions of families.
1
It appears to be of little importance whether the families are classified, as in some enquiries, according to income, or, as in others, according
to expenditure.

— 40 —

consisting of two or three persons may differ considerably
from those of large families. These differences would be hidden
if only general averages for all families were given.
In classifying families according to size, a special problem
arises. It is not sufficient to be told that the average number
of persons in a given group of families is, for example, 4.9.
Since the consumption of adults is generally greater than that
of children and the consumption of boys and girls between the
ages, say, of 5 years and 15 years is greater than that of infants,
it is necessary to know the age and sex of persons who make
up the total of 4.9. A " person " is not a satisfactory unit, arid
the question arises as to how to convert families of different
size into terms of a more exact unit. Such a unit, rather than
the number of persons in the families, may be used as basis for a
classification of results according to size of family. In certain
enquiries this course has been taken. The problem of the
reduction of families of different size to terms of a common
unit is considered in a special section which follows those
dealing with the tabulation of income and expenditure.
INCOME

The results of different enquiries vary somewhat with
regard to the details they give of the income of the families
covered. In the case of income and also of expenditure, the
amount of detail available depends largely on the questions
which the families were asked to answer. For detailed enquiries
it is desirable that data, classified according to district, grade,
and size of family should be published for all items of income
which are of sufficient importance or interest to be given separately. Data under the following headings have been given in
various enquiries :
Earnings of husband.
Earnings of wife.
Earnings of children.
Income from boarders and lodgers.
Income from insurance, pensions, etc.
Income from garden, etc.
Income from investments and other capital.
Income in the form of gifts.
Other income.
The first three appear essential if comparisons are to be made
between the economic conditions of different grades of families.

— 41 —
They help to answer questions with regard to the extent to
which the wives and children in families in different income
classes supplement the earnings of the husband. A separate
statement with regard to income from boarders and lodgers
is also desirable because in certain cases the income from this
source constitutes an appreciable contribution to the family
income. Whether the other items specified should be given
separately or should be combined into the group, " Other
income " depends on the relative importance in each case.
It might be decided to give separately all sources of income
which produce on the average more than a given proportion,
e.g. 0.5 per cent, of total income.
As an example of the more detailed type of table showing
income, the data published in the Swedish enquiry of 19131914 may be given. In a number of other enquiries similar
details, with minor variations, are shown.
A V E R A G E INCOME P E R H O U S E H O L D O F 1,355 F A M I L I E S IN S W E D E N , 1913-1914
Average expenditure per
family in crowns
N u m b e r of households
Average n u m b e r of consumption units p e r household

Under
1,500

1,500
to
1,750

1,750
to
2,500

2,500
to
4,000

AU
households

317

345

572

121

1,355

2.27

2.43

2.74

3.16

2.59

Cr.

Cr.

Cr.

Cr.

Cr.

Husband's 1 ordinary
earnings
1,211.6 1,423.2 1,669.9 2,300.5 1,556.1
Husband's supplementary earnings
47.7
128.2
62.2
28.3
75.9
35.8
22.4
42.3
39.3
Wife's earnings 2
80.0
23.3
14.3
76.7
56.2
Children's earnings
162.4
Income from boarders
13.5
29.3
64.4
62.9
43.5
and lodgers
Sickness benefits, pen13.8
27.9
26.6
22.6
23.6
sions, etc.
41.4
14.7
19.3
25.9
23.0
Gifts
15.0
1.5
2.6
3.9
4.0
Interest on securities
68.0
26.8
37.3
56.7
45.8
Other income
1,346.9 1,646.4 2,042.3 2,881.0 1,853.7
Total income
1
In t h e original source separate figures are given of income in money a n d income in
kind.
* I t is of interest t o note t h a t in t h e case of families w i t h t h e highest expenditure, based
on t h e higher earnings of t h e h u s b a n d , t h e a m o u n t s earned b y t h e wife a n d also by t h e
children were greater t h a n in t h e case of families in t h e lower expenditure groups.

In addition to these data, the Swedish statistics show for
corresponding categories of income the average income per
household of small families consisting of 1.8 units of consump-

— 42 —
tion only, i.e. husband and wife, of two groups of families of
intermediate size with 1.9-2.9 units of consumption, and of
large families consisting of 3.0 or more units of consumption.
Similar averages are given per unit of consumption. In addition
to the data covering all households, averages on the same lines
are tabulated separately for the 933 working-class families
and for the 422 families of non-manual workers covered by the
enquiry. Data are also given for the chief provinces and for
the chief towns of the country.
EXPENDITURE

The problems encountered in tabulating data with regard
to expenditure are the amount of detail to be given and the
classification into groups of items for which separate information
is shown. The results of various enquiries differ considerably
in the amount of detail published. The main features have
been considered under the heading of " Scope of Information "
(Chapter III, page 26). When the data are collected,
however, it may be found that certain items mentioned separately in the form distributed constitute so small a part of total
consumption that it is unnecessary to record them separately.
A similar course could be taken here to that suggested in the
case of income, and a minimum proportion of total expenditure
be set up. Only those items expenditure on which at least
equalled the minimum proportion would be quoted separately.
With regard to grouping these separate items, the practice
of various enquiries differs considerably. In most compilations
a few large groups, such as food, housing accommodation, fuel
and light, clothing, and a group of miscellaneous items are
adopted. Some enquiries give a separate group for furniture,
furnishings and other household requisites, whereas others
include these in the miscellaneous group. In a considerable
number of enquiries there is no miscellaneous group, but instead separate figures are shown for each of a number of groups
which in other enquiries are given separately, but for which
a combined miscellaneous total is also calculated. For the
purpose of international comparability, no difficulty would
arise from differences in grouping, provided that each country
would give separate figures for the same sub-groups or items.
The combination of these into larger groups on different principles in the various countries would not prevent strict comparability of results.

— 43 —

In the case of food, some enquiries limit the group strictly
to food items, together with tea, coffee, cocoa, and similar
non-alcoholic drinks. Others combine alcoholic drinks, and
also tobacco, cigars and cigarettes, with food into a general
group of food and so-called luxury consumption. There is no
objection to this combination provided separate figures are also
given for each part. It would appear preferable, however, to
give alcoholic beverages and tobacco as separate groups or
as separate parts of the miscellaneous group, and not to combine
them with food. The Australian enquiry of November 1913,
the United States enquiry of 1918-1919 and other enquiries give
alcoholic beverages and tobacco in the miscellaneous group.
Few difficulties arise in the case of the fuel and light group.
Most enquiries give either the quantities and expenditure, or
expenditure only on the chief items. In the United States
enquiry, data are given separately for families living in houses
and those living in flats and apartments.
The group "Housing accommodation" is more difficult.
As was indicated in Chapter III, it is not always possible to
separate clearly the rent for the dwelling from certain forms
of local taxation levied in respect of occupancy. In such cases
an attempt should be made when compiling the results to state
as precisely as possible what is included under the heading
" Rent ". Again, as has already been indicated, it would be
of value, where the habit of keeping lodgers or of sub-letting
is common, if rent could be divided into two parts, that incurred
in respect of the family and that necessitated in respect of the
extra accommodation for lodgers and sub-tenants.
The practices adopted in various enquiries may be outlined.
Many enquiries merely give the amount of the rent. The
Australian enquiry of November 1913 gives separate figures
for interest on mortgages, instalments on the hire-purchase
system, and rents. In the Swedish enquiry of 1913-1914, two
sub-groups are given, one for rent and the other for cost of
repairs and expenditure for removal from one house to another.
Expenditure for removal is given as a separate item of the miscellaneous group in the case of the United States enquiry of
1.918-1919. The United States enquiry gives separate statistics
of the rents of houses and of flats and apartments. In each
case the average rent of the household is given and also the
average rent per room and per person.

— 44 —
In the case of furniture, furnishings and other household
requisites, it seems preferable that these should be given as a
separate group than that they should form part of the miscellaneous group. This course is taken in certain enquiries — for
example, the United States enquiry and also the Swedish enquiry,
which however has no large miscellaneous group. Such a
group of " Household requisites " would cover articles of
furniture, including carpets, bedding, table cloths, cutlery,
crockery, kitchen utensils, towels, curtains, etc. Difficulties
arise with regard to such articles as lamps, gas mantles, electric
bulbs, and gas stoves. These might be classified in the heating
and lighting group. It appears preferable, however, to include
them amongst household requisites, and to limit the heating
and lighting group to wood, coal, gas, electricity, oil, and other
combustibles. Various laundry utensils might also be grouped
under the heading " Laundry ". It seems desirable, however,
in cases where laundry work is done by the family to give
durable articles such as wash-tubs, wringing machines, and irons
in the group of household requisites. On the other hand soap,
soda, and other cleaning materials may be more appropriately
placed as a separate division of the miscellaneous group.
The " Clothing " group is fairly straightforward. It generally
includes expenditure for repairs of clothing. Some enquiries
also combine expenditure for laundry with that for clothing.
It seems preferable, however, to make a separate group for
expenditure on laundry or to regard it as a subdivision of the
miscellaneous group, as laundry work is concerned not only
with the washing of clothing but also of bed-linen, table cloths,
towels, and other household articles. The Swedish enquiry
of 1913-1914 gives a group for laundry and cleaning with a
sub-heading for soap, soda, and other cleaning materials which,
in certain enquiries, are grouped elsewhere, In the United
States enquiry separate figures are given in the miscellaneous
group for " Laundry sent out ", and for " Cleaning supplies,
soap, etc. " This enquiry, too, classifies expenditure on clothing in great detail, separate figures being given by income
groups for each article of clothing for husbands, male children
of various ages, wives, and female children of various ages.
The Bombay Labour Office gives data by income groups for
different articles of clothing separately for men, women, and
children. In this enquiry, bedding is included in the clothing
group.

— 45 —
As has been indicated, the miscellaneous group differs very
considerably according to the published results of different
enquiries. Reference has been made to the different practices
with regard to alcoholic drinks, tobacco, furniture, and furnishings, etc. In the chapter on "Scope of Information", a list
was given of some of the chief groups of miscellaneous items
for which information has been asked in various enquiries.
Each of these groups however consists of a number of separate
items, and there is variation in the methods adopted in
different enquiries for grouping them. As has already been
indicated, comparability of the results of different enquiries
is not impaired if adequate details are given for sub-groups.
Thus, for example, the Swedish enquiry of 1913-1914 includes
expenses for " Haircutting and shaving " together with medical
attendance in the group " Hygiene ". As however separate
figures are given, it is possible to make comparisons with other
enquiries which give this item as a separate part of the miscellaneous group.
It may be added that in many enquiries, the number or
weight of different items purchased is given as far as possible
in addition to expenditure. Thus the number of pounds of
bread and of flour purchased are given, also the number of
coats and collars, the number of chairs, the quantity of coal,
and so on. This is desirable in detailed enquiries, as the quantities, being independent of price changes within a country as
well as of differences in currency between different countries
form a more satisfactory basis of comparison than expenditure,
Where the object of the enquiry is merely to provide weights
for the calculation of index numbers of the cost of living, expenditure is quite satisfactory. All that is necessary for this purpose is the relation between the amounts spent on different
items and groups to showthe importance of each in consumption.
It should be noted that for rent and certain other items, expenditure only can be given.

METHODS OF REDUCING DATA FOR FAMILIES OF DIFFERENT
SIZE TO TERMS OF A COMMON UNIT

The incomparability, to which reference has already been
made, arising from the fact that information collected in family
budget enquiries refers to families of different size, could be

— 46 —
avoided by collecting data only from families of a given size.
Such a method is, however, impracticable, especially in view of
the difficulty experienced in certain countries of obtaining an
adequate number of budgets even if families of different size
are included. The alternative is to adopt some unit of consumption and to reduce the consumption of families of different
size to terms of the unit chosen. The unit may be, for example,
the consumption of an adult man, an infant or a family of
given size. Thus, it might be found that the consumption of
a man is ten times that of a child under two years of age, that
children between two and five years of age consume twice as
much as children under two years of age, and so on. If such
relationships were established for persons of different age and
sex, it would be possible to reduce data for families of different
size to terms of the consumption of an adult man, of a child
under two years of age, or of some other unit.
The relationship which exists between the consumption of
persons of different sex and age differs from one group of commodities to another. For food one series of relationships
may be applicable, whereas for housing the proportions may be
very different. It does not appear, therefore, altogether
satisfactory to adopt a general scale and to apply this in turn
to different groups of commodities with which it has little or
no relation. The more satisfactory course appears to be to
adopt a separate scale for each group of commodities in respect
of which relative consumption differs. Thus, one scale will
be applicable to food, a second to clothing, a third to housing
accommodation, and so on. Unfortunately, there are practical
difficulties in the way of adopting this plan. The first is the
difficulty of elaborating a satisfactory scale for each group of
commodities. For such groups as food and clothing the construction of a scale is comparatively easy, as consumption in
the case of these two groups is individual in character. With
regard to other groups — for example, housing accommodation,
fuel and light — it is difficult to determine the proportion of total
expenditure which is necessitated in any family by its different
members. The second difficulty is that if four or five different
scales are used in compiling the results of a family budget enquiry, the process of compilation is complicated. In view,
therefore, of these difficulties the alternatives seem to be either
to apply one scale for food and a second scale to other groups
of commodities, or to take the somewhat unsatisfactory course

— 47 —
of constructing a single scale in which allowance is made for
all groups. Of these alternatives the former seems preferable, as at least it would ensure satisfactory results in the case
of the food group, for which a reasonably accurate scale can be
constructed. This plan has been already adopted by the
Basle Statistical Office in compiling the budgets obtained
in the Basle district during post-war years.
In determining the relation between the consumption of
persons of different sex and age, two main methods are possible,
that of economic research and that of medical research. Of
these the latter is applicable to the food group only. It is the
method which has been most frequently used in the establishment of scales showing the relative consumption of persons
of different age and sex. The relations have, however, been
established in practically all cases for other purposes than for
use in compiling the results of family budget enquiries, and as
has been seen are not altogether suitable for this purpose.
The results apply to the food group only and there are objections
to applying scales based on food to other groups of expenditure
and even to income, with which they may have little or no
relation.
The method of economic research has received less application in practice 1 . It consists in obtaining statistics of consumption for a number of families, the cost of each item consumed by each member of the family being recorded. In
obtaining from the different families the information which
would be required for the purpose, difficulty arises in the case
of commodities which are " consumed " in common by the
whole household. For such commodities it is difficult to determine how much of the total expenditure is necessitated by
different members of the family, and estimates must take the
place of actual statistics. The difficulty is greatest in the
case of housing accommodation, fuel and light, furniture, and
certain miscellaneous items. It is not possible to determine
with any high degree of accuracy in such cases what part of
total expenditure is due to the different members. For food,
which is a matter of individual consumption, the information
would be easier to obtain, although difficulties arise owing to
1

An interesting example of its use was that by Mr. Edgar Sydenstricker and Mr. Willford I. King in an investigation into population
and income in the case of cotton mill workers in South Carolina. Journal of Political Economy, Vol. XXIX, 1921, pp. 571-594.

— 48 —
the common preparation of food for the whole household.
Consumption of clothing is the most individual in character
and would be the easiest to deal with.
Some of the chief scales which have been used in family
budget enquiries for the purpose of reducing families of different
size to terms of a common unit are given in the table below.
V A R I O U S SCALES F O R C O N V E R T I N G D A T A F O R F A M I L I E S O F D I F F E R E N T
SIZE INTO TERMS OF A U N I F O R M CONSUMPTION U N I T »
Scales based on
other groups of
commodities
as well as food

Scales based on food only
Age
Engel's
omet
system

Atwater

Lusk

Amsterdam

United
States

German
(modified) *

Australian

Under 1
100
15
15
10
20
30
50
20
50
15
1 and under 2 110
10
20
30
30
50
15
2
„
3 120
10
35
40
35
50
15
3
„
4 130
10
35
40
40
50
40
4
„
5 140
20
35
40
45
50
40
5
„
6 150
20
35
40
50
70
40
6
„
7 160
55
70
75
20
50
50
60
70
75
7
„
8 170
30
50
50
65
70
75
8
„
9 180
30
50
50
70
83
75
9
„
10 190
30
50
50
75
83
90
10
„
11 200
40
65
60
80
83
90
11
„
12 210
40
65
60
85
83
90
12
„
13 220
90
40
65
70 60 100 83 90
100 90 100 90 50
13
„
14 230
80 70
80 70
14
„
15 240
50
80 70
80 70
15
„
16 250
70 60 80 70
90 80
16
„
17 260
70 60 80 70
90 80
17
„
18 270
90 70 100 80
100 80
18
„
19 280
90 70
19
„
20 290
100 80
20
„
21 300
21
„
22 310 300
22
„
23 320 300
23
„
24 330 300
1
figures
given, t h e lower represents t h e consumption of a woman and
24 W h e„r e t w o 25
340are300
t h e higher t h a t of a m a n .
25 *and
over
350
300
T h e original German scale was applied only t o " normal " families, t h a t is, families
of unchanged composition during t h e year of t h e enquiry, consisting of p a r e n t s with dependent children (under 15 years of age) or of h u s b a n d a n d wife without children. Consequently, t h e only ratios necessary were those for children u n d e r 15 a n d for adults. The
scale for such children was t h a t given in t h e t a b l e . The consumption of persons over 15
years of age was t a k e n as 100 for males and 80 for females. T h e modifications cover t h e
years from 15 t o 19.

They are shown in two groups, the first where the basis of the
scale is the consumption of food only, and the second where
the consumption of other groups of commodities have also

— 49 —
been taken into consideration. Engel's quet system, the
scales of Atwater and Lusk, the Amsterdam and the United
States scales are in the first group. Two scales, the German
and the Australian, are given in the second group.
Of the scales given in the table, the " quet " system elaborated by Dr. Ernst Engel is one of the earliest and best known 1.
The unit of the scale is the consumption of a newly-born child,
the unit being named " quet " in honour of Quetelet, the results
of whose researches Engel used in his investigations. Increase
in food consumption is considered to correspond closely with a
person's increase in height and weight. Data of the average
height and weight of persons of different sex and age were
compiled, and the product of height and weight served as basis
of the scale. The " quet " scale was used in connection with
the Swiss family budget enquiry of 1912, the Hungarian
enquiry of 1917, and the Belgian enquiry of 1921.
The Atwater scale is based on dietary studies made in New
York in 1895 and 1896 by Professor Atwater and Mr. Wood,
slight changes having been subsequently introduced. It was
used by Mr. Rowntree, for food requirements, in connection
with the investigations which he carried out in York in 18991901. A variation of the scale was used in the compilation of
the results of the enquiry conducted by the Osaka Municipal
Bureau of Labour Research in 1919-1920.
The Lusk scale was established by Professor Lusk of the
Medical School, Cornell, U.S.A., on the basis of physiological
requirements of food. It was used in connection with the
British family budget enquiry of 1918, the Egyptian enquiry
of 1920, and the Bombay Labour Office enquiry of 1921-1922.
It was also adopted by the Inter-Allied Scientific Food Commission, of which Professor Lusk was a member, for converting
the populations of different countries into terms of average
adult males.
The Amsterdam scale, which is based on food, was constructed
in connection with the enquiry conducted in that town in March
1917. It has been used in the compilation of the results of
various other enquiries subsequently conducted in Amsterdam,
1
The scale appears to have been first used by Dr. Engel in
Der Werth des Menschen, Berlin, 1883. It is also used in his
Die Lebenskosten belgischer Arbeiter-Familien früher und jetzt,
1895. See Bulletin of the International Statistical Institute,
Part I, 1895, pp. 4-8.

his work
study on
Dresden,
Vol. IX,
4

— 50 —
and also in connection with enquiries of The Hague Municipal
Statistical Office.
The United States scale is based on food studies, and in
United States enquiries has been applied to the food group
only. It appears to have been first used in the United States
enquiry of 1901-1902. In addition it was used in the United
States enquiry of 1918-1919, in the Swedish enquiries of 1916,
1917 and 1918, in the Swedish agricultural enquiry of 1920, in
the Danish enquiries of 1915, 1916 and 1922, in the Norwegian
enquiry of 1918-1919, and in the South African enquiry of
1925.
The German scale was drawn up by the German Statistical
Office for use in its family budget enquiry of 1907 1. It is based
not only on food consumption but, to a certain extent, on other
groups of expenditure. In addition to use in the German
enquiry of 1907, the scale (modified) was applied in the Stockholm enquiry of 1907-1908, the Finnish enquiry in 1908-1909,
the Norwegian enquiry of 1912-1913, the Austrian enquiry of
1912-1914, the Swedish enquiry of 1912-1913, the Norwegian
enquiry of 1916, the Swiss enquiries of 1919 onwards 2, various
Czechoslovakian enquiries, and the Polish enquiry of 1922.
The Australian scale was constructed in connection with the
family budget enquiry conducted in November 1913 by the
Commonwealth Bureau of Census and Statistics. It is based
on a study of data of a physiological, economic and anthropometric character 3.
An examination of different scales shows wide variation
from one to another. The differences are no doubt due in part
to differences in the methods adopted in their construction.
They are probably due in part, however, to differences in climate
and standards of living. Some scales are much more detailed
than others, e.g. the " quet " system is more elaborate than
1
In t h e unofficial enquiries conducted into the family budgets of
H a m b u r g teachers in 1903 and of metalworkers in Germany in 1908,
other scales were used. These and the Statistical Office scale were
compared by Dr. Ritzmann in an article on scales for comparing d a t a
for families of different size, published in t h e Archiv für Soziale Hygiene,
Leipzig, 1910, pp. 225-275.
2
The Basle Statistical Office, in publishing separately t h e budgets
for t h e Basle district which were obtained in these enquiries, applied
t h e " quet " scale for food and the German scale for other groups of
commodities.

* See

COMMONWEALTH B U R E A U OF CENSUS AND STATISTICS :

and Industrial

Report, No. 4.

Labour

— 51 —
the Lusk scale. Some scales distinguish between the consumption of the different sexes only in the case of adults. Others
make this distinction also for adolescents. In most scales the
consumption of male and female children is regarded as the
same x.
Scales differ considerably as to the age when maximum
consumption is attained. In this respect the " quet " scale
is the highest, maximum consumption being reached by a
woman at the age of 20 and by a man at the age of 25. In
most scales, however, maximum consumption is considered
as being reached at an earlier age, in certain cases as low as
14 or 15 years of age. It this connection it may be noted that
the age when maximum consumption is reached is likely to vary
from country to country. Also maximum consumption of
food may be reached at an earlier age than maximum consumption of clothing and other items. None of the scales
given in the table makes allowance for the reduced consumption
of aged persons 2. The question of making this allowance
is of little importance in connection with family budget studies.
Differences in the scales may be illustrated by comparing
the relative consumption of children of a given age or of a
family of given size according to the different scales. Thus,
the consumption of children of one year of age, according to
the United States scale, is only 15 per cent, of that of an adult
male, whereas in the Lusk scale it is 50 per cent. According
to the German scale, children of 6 years of age have a consumption equal to 20 per cent, of that of adult males, while
according to the Australian scale the percentage is 50. With
regard to families of given size the table below shows the consumption of two families expressed in terms of the consumption
of an adult male according to the scales given in the table
above. Family 1 consists of husband, wife, a boy of 15 years
of age, a girl of 13 years, and a child of 11. Family 2 consists
of husband and wife, a boy of 12 years, and two other children
aged 9 years and 5 years respectively.
1
In a scale used by the Danish Statistical Department in connection
with its family budget enquiries of 1897 and 1909, however, the consumption of females from the earliest years is given as lower than that
of males in the same age group.
2
Certain scales have made such allowance. In the Zuntz scale, for
example, which was constructed shortly after Engel's scale, and is
based on food requirements, the consumption of men over 60 years of
age is considered to be 10 per cent, less than that of males between the
ages of 15 and 60.

— 52 —
Family 2
units

amily 1
units

Scales

Based on food : Quet
Atwater
Lusk
Amsterdam
All groups
United States
German
Australian

3.83
3.80
4.49
4.50
4.70
3.40
3.95

3.46
3.30
3.86
3.80
3.95
2.70
3.30

Reference has been made to the difficulties in establishing
scales for other groups than food. Certain data are available
to show relative consumption of clothing, expenditure on which
is of an individual character thus making it possible to determine to a large extent the expenditure incurred in respect of
each member of the family 1 . In the enquiry conducted by
the Osaka Municipal Bureau of Labour Research in 1919-1920,
although the United States scale was used in modified form
not only in the case of food, but also for other groups of expenditure, a special scale was constructed for clothing. This
scale, which was not used in connection with the enquiry, is
as follows :
Males

75 years to 90 years
61 „
75
„
46
)
60
„
31
45
„
26
J
30
„
21
25
„
16
20
„
11
15
„
1
10
„
j

J

ï

J

30
40
60
70
100
70
50
30
20

Females

10
40
40
40
60
80
90
60
30

In one family budget enquiry — that conducted in t h e
United States in 1918-1919 — data were obtained with regard
to expenditure on clothing of each member of the family.
On the basis of this information, it is possible to calculate a
scale showing the relative expenditure on clothing of persons
of different age and sex. The following table shows the yearly
expenditure of various age and sex groups, together with relative
figures :

1

Difficulty arises where the same articles of clothing are used by
different children in the family as they pass through the same age group-

— 53 —
Males
Age group

Parents
Children, 15 years and over
„
12 and under 15
>>
8 ,,
,, 12
4 „
„ 8
,,
under 4

Yearly expenditure

$
71.38
81.77
47.08
39.39
31.31
22.09

Females

Relative Yearly ex- Relative
figures
penditure
figures

t
100
115
66
55
44
31

63.55
96.59
49.93
39.30
32.53
23.68

89
135
70
55
46
33

The above data are based on all families covered by the
enquiry. Similar data are available for seven different income
groups ranging from families with under $ 900 per annum to
families with $2,000 and over.
The relative figures given above for the United States show
that maximum expenditure on clothing was that of children of
15 years of age and over, the expenditure of husbands and
wives being less than that of this group. The high relative
expenditure of unmarried persons of 15 years of age and over
may be due in part to their desire to wear attractive clothing.
It is also probably due to the fact that many persons in this
group work, and instead of pooling the whole of their income
with that of the family, a part of it is reserved for individual
expenditure, of which clothing constitutes a considerable part.
For housing, furniture, fuel and light, expenditure is common
to all members of the family, and, as already indicated, it is
difficult to determine the cost involved for each member.
Also for these groups, scales based primarily on food requirements are particularly unsatisfactory. Few families require
increased housing accommodation during the first few years
after the birth of an additional child. On the other hand,
as children become older the necessity of taking account of
differences in the sex of the children involves considerable
increase in housing accommodation if satisfactory conditions
are to be maintained.
Very few attempts have been made to construct scales for
housing requirements. The following, which was used by
Dr. Bowley in his book Livelihood and Poverty may be quoted.
The scale shows the relative requirements of accommodation
of persons of different sex and age x, the figures being :
1
The scale allows for the need for more space in houses where there
are adolescents of both sexes.

— 54 —
Adults
Boys (14 to 18 years of age)
Girls (14 to 16
„
„ )
Children (5 to 14 „
„ )
Children (under 5)

100
75
75
50
25

Expenditure on fuel and light and on furniture and furnishings
may be regarded as being on about the same scale as that for
housing accommodation.
It may be added that in some only of the family budget
enquiries are the results calculated on the basis of a common
unit, in the remainder averages per family or per head being
given. Where the results are required merely to provide
weights for the calculation of cost-of-living index numbers
it is perhaps unnecessary to introduce the refinement of the
common unit. It is, however, desirable to do so if a high
degree of comparability is required.

— 55 —

APPENDICES
APPENDIX I
Family Budget Enquiries Conducted in Different Countries

In the notes which follow information is given as to important family
budget enquiries which have been conducted in different countries.
Only enquiries which have been made since about the year 1900 are
included. The notes are further limited generally to investigations
undertaken by a department of the central Government or a local
administrative authority. Thus no account is given of the wellknown investigations of Mr. B. S. Rowntree in Great Britain, nor of
the considerable number of enquiries conducted by private individuals
or organisations in Germany, the United States, and other countries.
In a few cases, however, mention is made of private enquiries especially where, in the absence of official data, the results have been
utilised by the public authorities.
Indications are given as to the scope of the different enquiries, the
methods adopted for obtaining the information and the system of
compilation of the results. The countries are arranged in alphabetical
order as determined by the French names.
UNION OF SOUTH AFRICA
The Economic Commission, appointed in 1913 to enquire into the
cost of living, various working conditions, and the cost of production
in industrial centres in the Union of South Africa, collected about
100 budgets, especially with regard to food and rent. In connection
with the same enquiry, the Transvaal Chamber of Mines collected
and analysed 138 budgets, while the Commission also made use of
22 budgets, of families living in Johannesburg on approximately the
income of an artisan, collected by Mr. Alexander Aiken and analysed
in a paper which he read to the British Association in 1907. In
addition to these budgets, the budgets of 326 families of railway
employees with salaries varying from £240 to £300 per annum at
Pretoria, Bloemfontein, and Johannesburg served as basis for the
weights used until the end of 1920 in the calculation of the South
African cost-of-living index numbers. Certain other data on family
budgets were compiled during the pre-war period.
In July 1925 a Departmental Committee, appointed to enquire
into the cost of living in the Union of South Africa, collected a number
of family budgets. One of the main objects of the Committee was to

— 56 —
determine the expenditure required to maintain a typical family in
certain industrial areas on an average civilised standard of living
and on a minimum civilised standard of living. It was in order to
secure information as to actual consumption that the Committee
conducted a family budget enquiry. A detailed form was prepared
and widely distributed through the medium of inspectors of labour
and magistrates, with the assistance of trade union officials 1. The
families were asked to record information as to income and expenditure
for a period of a week or a month. Altogether 557 budgets were
secured, but of these 115 were received too late for tabulation, and
38 were discarded for various reasons. Of the budgets used by the
Committee, about one-fourth were those of the families of bank
officials, while the remainder covered the families of manual and
non-manual workers in a considerable number of different occupations. In analysing the results the budgets were classified into
three groups — those of families with incomes under £240 per annum,
between £240 and £360, and £360 to £500. Separate data were given
for each of these groups and also for each district. The American
scale was used for calculating averages per consumption unit.
SOURCES
UNION OF SOUTH AFRICA : Report of the Economic Commission, January

1914.
Social Statistics. Published by the OFFICE of CENSUS AND STATISTICS.

Report of the Cost-of-Living Committee, 1925.

GERMANY
Prior to 1907, enquiries into family budgets were conducted mainly
by private individuals and were limited to the conditions of living
of families in a given district or associated with a given industry.
A few enquiries had also been made by municipal or other local government statistical offices.
In 1907 an enquiry, covering most of the chief industrial districts,
was undertaken by the German Statistical Office in collaboration
with the municipal statistical offices. The municipal statistical
offices, often with the assistance of workers' organisations and other
similar bodies, secured the co-operation of the families. The families
were asked to keep detailed records of income and expenditure for a
period of one year. Certain municipalities made small money payments or a gift of books to those who supplied satisfactory records.
The records were kept in books, space being provided for entering
each day sums received and items of expenditure. The sources of
income for which separate information was to be given were specified.
In the case of expenditure, however, the form was blank, the householder entering each item as it occurred 2. The books included
pages giving model entries of income and expenditure as an illustration
of the manner in which the information should be recorded. Each
book covered a month, but was so arranged that the records for each
week could be readily detached. The records for each week when
completed were to be sent to the municipal statistical offices.
The enquiry covered the working class and also lower-grade officials
and teachers. Altogether, 4,134 books were distributed, but only
1

A copy of this form is given in Appendix II.
Copies of the forms used for the daily records of income and expenditure are given in Appendix II.
a

— 57 —
960 families supplied data for the whole year, while a number of
families supplied records covering periods of 6 to 11 months. Some
of the budgets were rejected as unsatisfactory. Of the 852 families
for which adequate information was obtained for 12 months, only
39 had expenditure of more than 4,000 marks par annum.
In the tabulated results, data of income and expenditure are given
separately for each family, and averages for each town and for the
whole country. Summary tables give separate averages classified
according to size of income and size of family. The data are also
tabulated separately by occupation ; also averages are given for
workers' families and for those of officials and teachers, classified
according to size and income. A further tabulation gives the expenditure for 391 normal families, i.e. families the composition of
which remained unchanged during the year, consisting of parents
with dependent children or of husband and wife without children.
The data for these families are reduced to terms of an adult man.
Reduction was effected by a special scale in which allowance was
made for other groups as well as food. In addition to tables showing
money expenditure, data are given, for the chief articles of food, of
the quantities purchased. In a supplementary table, data are given
for a number of families who kept records for periods of less than 12
months but more than 6 months.
The information obtained in the 1907 enquiry was supplemented
by the budgets of two higher grade officials who offered for publication information which they had compiled privately for 15 years
and 12 years respectively. Also five lower middle class families,
three of which had kept budgets for the 1907 enquiry, supplied budgets covering periods of two or three years, 1907-1909.
SOURCES
Supplement No. 2 to the Reichsarbeitsblatt, Berlin, 1909 : "Erhebung von
Wirtschaftsrechnungen minderbemittelter Familien im Deutschen
Reiche ".
Supplement No. 3 to the Reichsarbeitsblatt, Berlin, 1911 : " Zwei Wirtschaftsrechnungen von Familien höherer Beamter ".
ARGENTINA
The National Department of Labour has conducted several enquiries
into the income and expenditure of working class families. The most
recent was that of 1925, when data covering about 1,000 families
were compiled. Information as to income and expenditure was obtained by means of forms distributed to the families. In the case of
expenditure, data were requested for the main groups of items only
(food, rent, clothing, etc.), no details being asked for with regard to
expenditure on different items within each group. For housing,
questions were asked as to the nature of the accommodation. The
families were visited by inspectors of the Department.
When the results were compiled, the information supplied was
found to be adequate to establish the proportionate distribution of
expenditure for three groups only (food, rent and other expenses).
Information as to income is also given in the published results.
Comparisons are made with data for previous years.
SOURCE
: Cronica Mensual. February 1926.

NATIONAL DEPARTMENT OP LABOUR

— 58 —
AUSTRALIA
The first budget enquiry in Australia covered the whole of the
period from 1 July 1910 to 30 June 1911, the number of budgets
available as a basis for the compilation of the results being 212 out of
1,500 books which were distributed. The number of persons covered
was 999. The smallness of the number of budgets was believed to be
due to the labour entailed in keeping the desired records for the full
period of 12 months. In the case of rent the figures were checked by
data obtained direct from house agents. For certain items of food the
aggregate consumption was obtained from import, export and productive statistics, and the average weekly expenditure obtained from
them was compared with the figures obtained in the budget enquiry.
' A second enquiry was made in November 1913. The period covered
was one month, this being chosen in order to avoid the labour involved in keeping records over a long period. It was hoped that a large
number of families would supply information such as would permit
of a classification according to occupation, income, states and localities. Account books were distributed in October 1913 to householders throughout the Commonwealth 1. These provided for the
entering, each day, of particulars with regard to income and expenditure under certain specified headings. Altogether over 7,000 books
were distributed from the Commonwealth Bureau of Census and Statistics, but although wide publicity was given to the enquiry only
417, or less than 6 per cent., of the books distributed were returned.
Of these, 392 books only were sufficiently accurate for classification, the number of persons covered being 2,232. Had the response
to the enquiry been satisfactory, it was intended to make a similar
investigation during some other month shortly afterwards, so that
by combining and comparing the results of the two enquiries variations in expenditure due to seasonal influences might be eliminated.
In consequence of the small number of returns, no occupational
classification was adopted. Certain data are tabulated separately
by States. Many of the tables make a distinction according to size
of family between those with over 4 members and those with 4 members and under and also to a limited extent according to income groups.
In certain cases separate results are shown for families consisting of
father, mother, and two children under 14 years of age, and for
families consisting of father, mother, and three children under 14 years
of age. In certain tables comparisons are made between expenditure
in metropolitan and country districts.
As regards expenditure per head, data are given first without taking account of differences in consumption of persons of different age
and sex, and, second, taking these differences into account. The
relative expenditure on food of the different age and sex groups was
calculated by using the special scale named in the text of the present
Report as the Australian scale.
As regards expenditure on clothing and miscellaneous items, figures
per head are given, but no account is taken of differences according
to sex or age owing to the difficulty of computing satisfactory ratios
for groups of expenditure other than food.
In September 1917 the Government of Western Australia appointed
a Royal Commission of Enquiry into the cost, of the necessaries of
1

A copy of the chief pages of the book is given in Appendix II.

— 59 —
life in that State. In this enquiry one of the terms of reference instructed the Commission to investigate the effects of recent fluctuations
of prices upon the standard of living of wage earners and others. Forms
were widely distributed and serious attempts made to bring the importance of the enquiry before all classes. Nevertheless, the Commission reported that the response had been utterly disappointing and
the returns too few to admit of any safe general conclusions being
drawn.
Records of income and expenditure for 13 weeks were asked for, •
the householders being requested to return the data for each week
at the end of the week. Altogether, records were tabulated for 66
families, covering 380 persons, although for a number of these the
period covered was less than 13 weeks. The data were tabulated
on similar lines to those indicated above for the November 1913
enquiry.
Various other investigations had been conducted in certain towns
and districts in several States in connection with arbitration court
proceedings and the work of minimum wage boards. Mention should
be made of the investigations by the Royal Commission on the Basic
Wage, appointed in December 1919, which included among other
things an enquiry into the actual cost of living according to reasonable standards of comfort for a man with a wife and three children under
14 years of age.
SOURCES
Enquiry into the Cost of Living in Australia, 1910-191 i. Labour Report
No. 4. Expenditure on Living in the Commonwealth, 1913.
Labour Report No. 9.
Labour Report No. 13.
AUSTRIA
An enquiry was conducted in Austria by the Arbeitsstalisticher
Amt during the period 1912-1914. The object was to obtain information for as large a number of families as possible for a period of at
least a year. In addition to obtaining detailed information with
regard to income and expenditure, data were obtained with regard
to other aspects of family life, including housing, furniture, subtenancies, lodgers, insurance, debts.
The enquiry was restricted to Vienna, the families of workers and
those in a similar economic and social situation being selected. The
investigation was conducted with the help of a number of agents,
mostly women, who were instructed to assist the families in the fillingin of the forms, to check the information supplied, and to make supplementary enquiries where necessary. For this purpose the agents
kept in touch with the families at frequent intervals during the course
of the enquiry. For the purpose of recording the information, books
were distributed containing a form for each day for income and expenditure 1 . In the case of expenditure, information was asked for
with regard to the quantities purchased as well as their cost.
Of 280 families which undertook to keep accounts, 162 completed
records for a period of a least a year, and of these 119 were sufficiently
satisfactory for use. Seventy families completed a second year.
-The 119 families included 581 persons, or 4.64 persons per family.
A copy of the form used for the daily records is given in Appendix II.

— 60 —
For the purposes of comparison the German scale was used for converting families of different size to terms of a common unit. According to this scale the average number of consumption units ( = adult
males) per family was 2.8. In the tabulated results data are given
of income and expenditure, classified according to income groups and
size of family. Separate averages are given per family and per consumption unit. Information is given in great detail for each family
covered by the enquiry, not only as regards income, expenditure and
the quantities of different food commodities consumed but also as
to size of family, size of house and other data regarding living conditions.
SOURCE
Wirtschaftsrechnungen und Lebensverhältnisse von Wiener Arbeiterfamilien, 1912-1914. Vienna, 1916.
BELGIUM
Researches into family consumption were conducted in 1920 by
the Ministry of Industry and Labour on the basis of 1,028 budgets
collected in 1910 by the Institut de Sociologie Solvau. Figures showing the relative expenditure on the different items and groups of commodities were calculated and these were used in the compilation of
index numbers of the cost of living. The figures were, however,
considered unsatisfactory owing to the changes which had taken
place in habits of consumption during the period 1910-1920, and further researches were undertaken.
At the beginning of 1921 an enquiry was made into the expenditure on food of working class and lower middle class families 1 . The
families were chosen so as to be representative of such families in the
country as a whole. Arrangements were made for data to be supplied by 1,125 working class families and 375 lower middle class
families in many different industries and occupations. The choice
of the families was left to the agents who were responsible for collecting the information in the different districts. Each household was
asked to note from day to day the kind, the weight and the price of
the food commodities consumed during a period of 14 days. These
data, together with the information as to the income of the household
during the period covered, were entered on a special form. Actually
only 1,047 forms were obtained and of these it was necessary to reject
199 which had not been satisfactorily completed. Of the 848 satisfactory forms, 673 gave information with regard to working class
families and 175 to lower middle class families. Of the working
class families about 67 per cent, were those of workers in the mining,
metal, textiles, clothing, and wood and furniture-making industries,
the remainder being distributed in various other industries.
In order to compare the data for families of different sizes, the
" quet " system was used, and for the different food commodities
the quantities (in grammes) consumed per " quet " in a period of
14 days were calculated separately for working class families, for
1
The lower middle class was defined as consisting of officials and other
non-manual workers with annual incomes not exceeding 10,000 francs,
together with retailers not employing more than five workers, and
employers not employing more than ten workers.

— 61 —
lower middle class families, and for both these classes together
Comparison was also made between the quantities of different items
of food consumed per " quet " per fortnight by working class families
in 1921 with those in 1910, and certain deductions are drawn with
regard to changes in the standard of living during the period between
the two enquiries.
The enquiry into food consumption in 1921 was supplemented
during the same year by an investigation covering other groups of
expenditure, namely, clothing, housing . accommodation, heating
and lighting, and miscellaneous expenses including those in respect
of hygienic, social, and moral requirements, the main object of the
investigation being to provide weights for the calculation of cost-ofliving index numbers. The agents who had conducted the enquiry
into food consumption were instructed to secure information covering
a period of not less than a year. The data thus collected were regarded
as representative of conditions of consumption in the year 1921.
The relation between the amounts expended of different groups was
established for four classes of workers'families and also for lower middle
class families. The four categories of working class families were,
those with incomes of less than 20 francs per unit of consumption (i.e.
the " quet ") per fortnight, those with incomes of 20 but less than 30
francs, of 30 but less than 40 francs, and of 40 francs and over.
SOURCES
Revue du Travail, May 1922 (pp. 690-696) ; September 1923 (pp. 18861901), and subsequent numbers.
CANADA
Enquiries have been conducted into the consumption of families
in Canada, the results being used, in conjunction with similar data
for the United States and Great Britain, to provide weights for the
cost-of-living index numbers published by the Department of Labour
each month in the Labour Gazette. The weights represent the average
weekly consumption of a normal family of five persons, and indicate
approximately how a working man with an income of about $21
per week in 1913 would divide up his expenditure on food, fuel, light,
and rent.
Mention should be made of enquiries conducted by Minimum Wage
Boards in various provinces into the budgets of women workers maintaining themselves by their earnings. The results of these enquiries
have been used to establish minimum budgets on which to base minimum rates of wages.
SOURCE
The Labour Gazette (various numbers). Published by the DEPARTMENT
OF LABOUR.

DENMARK
Family budget enquiries were conducted in Denmark by the Statistical Department in 1897, when satisfactory information was obtained for 251 working class families, and in 1909 when 536 families
were covered, of which 310 were workers and 226 smallholders and
crofters. In these enquiries information for a whole year was obtained. During the war, enquiries were conducted in October 1915

— 62 —
and October 1916 for periods of 14 days, the results being considered
of value mainly for the food group. In 1916 an enquiry covering
the families of non-manual workers was conducted, information being
obtained for a year.
During the post-war period an enquiry was conducted in 1922
covering workers, employees, and independent artisans ; altogether
900 books were distributed, of which 624 were filled in. For some of
these, supplementary information was necessary, but in the meantime
data for 352 families were published covering the whole year. Among
these, 96 were for Copenhagen, 130 for provincial towns, and 126 for
rural districts.
The scale used in enquiries subsequent to that of 1909 for reducing
families of different size to units was that which has been adopted
in the United States enquiries.
The results of the enquiries are classified according to expenditure,
four separate groups of families being given as well as a general average
covering all families. Separate averages are given for families living
in Copenhagen, in provincial towns, and in rural districts, as well as
a general average for all the families together.
SOURCE
Statistiske Efterretninger, June 1912 and 22 January 1924.
ESTHONIA
In 1925 an enquiry covering working class families in Tallinn and
Narva was organised by the Central Statistical Office in collaboration
with the Narva Labour Exchange. Detailed information was obtained
regarding food consumption during one month (October-November).
Records were kept by the heads of the families on forms which specified
the chief items of food and included space for each day. In addition,
information was secured for the year ending October 1925, entries
being made on special forms of what the families remembered of their
income and expenditure during the year. These entries were generally
made by agents who conducted the enquiry. In addition to these
records, information was obtained with regard to the composition of
the household, the nature of the menus, and housing conditions. In
the selection of the families, the co-operation of workers' organisations
was obtained. In Tallinn, the General Sickness Fund also collaborated
for this purpose. An attempt was made to secure an adequate number
of budgets from families with different incomes.
The total number of families supplying satisfactory information on
income and expenditure was 187. In the case of food, 283 families
supplied information as to consumption during one month. Of these
202 supplied detailed records for the month of October. A number
of other families supplied useful information with regard to menus
and housing conditions. The results were classified according to the
size of family and income. Calculations were also made per unit of
consumption, a special scale, used in the Russian Department of
Vologda, being adopted. According to this scale relative consumption
was as follows :
units

Males, 18 to 60 years of age
Females, 17 to 55 years of age
Males, 60 years or 14 to 17 years of age

1.00
0.80
0.80

— 63 —
units

Females, over 55 years or 14 to 16 years of age
Children, 7 to 13 years of age
Children, 1 to 6 years of age
Children under 1 year

0.60
0.55
0.30
0.10

SOURCE
STATiSTiKA KESKBÜROO : Eesti Töölise Büdzhet, 1925 a. (Enquête
budgétaire parmi les ouvriers de Tallinn et Narva.)

RIIGI

UNITED STATES
In the Sixth and Seventh Annual Reports of the Commissioner of
Labour, published in the years 1890 and 1891 respectively, information as to family budgets was given for workers employed in a number
of important industries. In the Sixth Annual Report the information
related to 3,260 families (representing 16,581 persons) the heads of
which were employed in the iron, steel and allied industries. In
the Seventh Annual Report data were collected from 5,284 families
(representing 27,577 persons) the heads of which were employed in
the cotton, woollen, and glass industries.
In 1901-1902 the Department of Labour made another study
covering 25,440 working men's families (124,188 persons) located in
the principal industrial centres of the United States. The families
were selected without reference to industry, but it is believed that the
data were fairly representative of the conditions that existed among
wage earners in the whole country. The results were published in
the Eighteenth Annual Report of the Commissioner of Labour.
During the war the Bureau of Labour Statistics of the Department
of Labour made several less comprehensive local studies, the results
of which were published in the Monthly Labour Review. Mention
may be made of the study conducted in 1916-1917 in the District of
Columbia ; also of that conducted about the same time covering the
chief shipbuilding centres.
In 1918-1919 the Department of Labour, through the Bureau of
Labour Statistics working in co-operation with the National War
Labour Board, made an investigation into the cost of living in industrial
centres in the United States. The investigation covered white
families in 92 cities or localities in 42 States, the cities varying in size
from New York to small country places of a few thousand population.
The places were selected so as to get data representative of all parts
of the country. The points to be taken into consideration in the
selection of the families were stated on the forms used in the enquiry 1.
The number of families from which statistics were obtained was
12,096.
The data were secured by the personal visits of agents of the Bureau
to the homes of the wage earners and small salaried men, the information being obtained generally by interviews with the wife or other
member of the family.
The data on income and expenditure were compiled and published
in summary and in general tables. In the summary tables showing2
expenditure, figures are given separately for seven income groups
1

Extracts from the form are given in Appendix II.
These groups range from those with incomes below f 900 to those
with incomes of £2,500 and over.
2

— 64 —
as to the average number of persons per family, the equivalent in
terms of adult males * and the average yearly expenses per family.
Figures of expenditure are given separately for the food, clothing,
rent, fuel and light, furniture and furnishings, and miscellaneous
groups. Percentages are also given showing the relation between
the expenditure on each group and total expenditure. The data are
tabulated separately for each town and also for the United States
as a whole.
In the general tables, detailed information is given with regard
to the sources and amounts of family incomes in one year by cities
and income groups ; the average quantity of specified articles of food
consumed per family and per adult male in one year, by geographical
districts; the quantity of and expenditure for clothing purchased per
family in one year, by sex, age and income groups ; the number of
families living in houses or apartments having a specified number
of rooms and the average amount paid for rent in one year by cities
and income groups. Similar data are given for the fuel and light,
furniture and furnishings, and miscellaneous groups.
During the year 1923 the Department of Agriculture, in co-operation
with State colleges of agriculture and universities commenced a series
of studies on living conditions and the cost of living among farm
families in various localities of the United States, data covering a
year being obtained by means of personal visits to the homes. In
1924 reports were issued for selected localities in Iowa, Alabama,
Ohio, and Missouri. The number of families from which information
was secured in these States were 474, 187, 383, and 178 respectively.
During the year 1925 similar reports were issued for other localities.
One of these covered farm families of selected localities in Kentucky,
Tennessee, and Texas.
Although the studies do not cover identical periods, they are regarded as generally comparable. Within the areas chosen for study
in each State, information is obtained for all typical farm families,
that is, those having each an adult male working on the farm and an
adult female. Information is compiled separately for the families
of owners, tenants, and hired men. Expenditure included the value
of goods furnished by the farm as well as goods purchased. Figures
are given showing the proportion which the value of goods furnished
by the farms forms of total expenditure.
SOURCES
UNITED STATES DEPARTMENT OF LABOUR. BUREAU OF LABOUR STATISTICS : Cost of Living in the United States. Bulletin No. 357,

Washington, May 1924.
Monthly Labour Review, various dates.

UNITED STATES DEPARTMENT OF AGRICULTURE. BUREAU OF AGRICULTURAL ECONOMICS : Preliminary Reports on Living Conditions and

the Cost of Living in Farm Homes of Selected Areas of Iowa. Washington, 1924. (Similar reports have been published subsequently
for other States.)

FINLAND
In 1908-1909 a family budget enquiry was conducted with the
authorisation of the Board of Industry by Miss Vera Hjelt, factory
1
The scale, named in the text of the present Report " The United
States scale ", was used in calculating these equivalents.

— 65 —
inspector. The results are used, with certain minor modifications,
by the Central Statistical Office to provide weights for the calculation
of its cost-of-living index numbers.
In 1920-1921 an enquiry, which covered the twelve months from
1 March 1920 to 28 February 1921, was conducted by the then Central
Board for Social Affairs (now the Department of Social Statistics of
the Central Statistical Office). Account books were distributed, with
the assistance of the municipal authorities to families in 16 towns and
8 rural districts. The families included those of officials as well as
of manual workers. The former supplied 117 budgets and the latter
437.
Books, in which the records were to be made, were distributed to
the families. General questions were asked with regard to membership
of the family, housing conditions, etc. For income and expenditure,
columns were provided for each day. The separate items were specified in great detail. In the case of expenditure, both quantity and
cost were to be recorded.
In tabulating the results, expenditure per consumption unit was
taken as a basis of the classification, the German scale being used
for the calculation. Separate figures are given for manual workers
and for officials.
SOURCES
MINISTRY OF SOCIAL AFFAIRS : Sosialinen
Aikakauskirja ; Social
Tidskrift, No. 11, 1924.
Elinkustannukset tilinpitokaudella, 1920-1921 ; Levnadskoslnaderna under
bokfòringsperioden, 1920-1921. (Suomen Virallinen Tilasto XXXII.
Sosialisia erikoistutkimuksia 5. — Finlands Officiella Statistisk
XXXII. Sodala specialundersökningar 5.) Helsingfors, 1925.
FRANCE
In connection with the question of conditions of living of home
workers, a number of family budgets have been obtained by the
Ministry of Labour and Social Insurance. Thus in 1905, when an
enquiry was made into the wages and conditions of living of home
workers in certain needlework trades, 56 family budgets were obtained.
More recently, the regional committees established with the object
of studying changes in the cost of living used statistics of family
consumption to provide weights for the calculation of cost-of-living
index numbers.
In an unofficial enquiry conducted during 1923 by the Union
sociale d'Ingénieurs catholiques on the family and labour, information
on family budgets was obtained from different parts of France.
SOURCES
MINISTÈRE DU TRAVAIL ET DE LA PRÉVOYANCE SOCIALE, OFFICE DU
TRAVAIL : Enquête sur le Travail à domicile dans l'industrie de la

lingerie. Paris, 1907-1911.

STATISTIQUE GÉNÉRALE DE LA FRANCE. COMMISSION CENTRALE d'EruDES
RELATIVE AU COÛT DE LA V I E : Compte rendu des Travaux au cours

de l'année îgso. Paris, 1921.
Echo de l'Union sociale d'Ingénieurs catholiques, 1923.
HUNGARY
In 1917 an unofficial enquiry was conducted by Mr. Eugen DàlnokiKovàts. Forms were distributed to 300 families, who were asked to
enter information not for the date of the enquiry but for the pre-war
period. On account of the difficulties of giving such information,
5

— 66 —
only 32 reasonably satisfactory forms were returned. Of these, 19
were for families of lower-grade officials, and 13 for those of industrial
workers. The results have served as basis for the weights used in
the calculation of the cost-of-living index numbers compiled by Mr.
Eugen Dàlnoki-Kovàts.
In 1919 the Central Statistical Office of the Kingdom of Hungary
made preparations for an enquiry to cover 1,000 households in Budapest. This was to be followed by an enquiry covering industrial and
agricultural workers in the provinces. The development of events
in 1919 and the subsequent instability of economic conditions made it
undesirable to carry through the enquiry. The intention of making
an enquiry when conditions are stable has, however, not been given up.
In the beginning of 1926 the Statistical Office of Budapest undertook an enquiry covering 142 families of different grades, mostly
of workers but including also families of officials.
SOURCE
Communication of the Hungarian Correspondent of the INTERNATIONAL
LABOUR OFFICE.

INDIA (Bombay)
An enquiry was conducted in 1921-1922 by the Bombay Labour
Office into the consumption of the working classes of Bombay. Budgets were collected during the period May 1921 to April 1922, altogether 3,076 budgets being tabulated. These included 2,473 family
budgets and 603 budgets of men living alone *. The enquiry was
based on a larger number of budgets than any other similar enquiry
in other countries for any single city at one particular period. In
view of differences in religion, race, and caste of the workers, it was
found necessary to collect a greater number of budgets than would
have been necessary in a large centre of population in many other
countries. Of the budgets 90.3 per cent, relate to Hindus, 6.2 per cent.
to Mohammedans, and 1.8 per cent, to Christians. The occupations
covered include millworkers, 49.5 per cent. ; municipal workers, 14.8 per
cent.; dock workers, 12.2 per cent. ; railway workers, 8.9 per cent. ;
and engineering workers, 7.9 per cent. The extensive method was
used, but this was supplemented to a small extent by information
collected by the intensive method; the latter method was used partly
as a means of preliminary training of the investigators.
Data were collected with regard to size and earnings of the family,
cost and consumption of food, fuel and light, clothing, and miscellaneous items, and the cost of house rent, together with information
as to housing conditions. The results have been utilised to check the
relative importance given to each article in the cost-of-living index
number published by the Labour Office, for which the aggregate
expenditure method is used. It is intended to make use of the data
collected in the enquiry for the purpose of providing weights for the
cost-of-living index number calculations in place of those now used.
Separate forms were provided for family budgets and for the budgets of men living alone 2. For family budgets the agents were
1

The collection of budgets for men living alone appeared necessary
as many workers are not permanently resident in Bombay, but, leaving
their families in the villages, they work in the city for a period with the
hope of returning to the villages later.
2
A copy of the form for family budgets is given in Appendix II.

— 67 —
asked to select families consisting of husband, wife, and children ;
exceptional cases were to be avoided, and data were to be obtained
for representative families with different incomes — for example,
those of skilled and unskilled workers. Statements were to be obtained
from all working class districts within the area and from as large a
number as possible of representative trades. Families of workers in
full employment were to be selected, so that the information would
be representative of normal conditions.
The information asked for was to cover a period of one month.
In the case of clothing, an estimate was to be made of the length
of the period which each article would last, and information was to
be given as to the number of articles in use. For other items the
estimated cost per month was asked for.
In analysing the results, abnormal data were eliminated. The
budgets were classified by income and occupation; the income classes
range from those below 30 rupees per month to 90 rupees and over.
In addition to summarised data detailed statistics are given of
the quantities of the different items consumed. Representative and
sample budgets are also given for workers in different occupations.
At the end of 1923 and the beginning of 1924 an enquiry was conducted by the Bombay Labour Office into the consumption of European
families. The object of the enquiry was to provide weights for the
calculation of index numbers snowing changes in the cost of living
based on the consumption of Europeans. In conducting the enquiry
a selection was first made of various articles, 80 in number, which
enter into the domestic budget of the ordinary family living in Bombay
according to the European manner of living.
To obtain the quantities consumed of the different articles, the
" intensive " method was mainly used, a careful study of a family's
consumption over a period of several weeks being made. In addition
the " extensive " method was used, a number of family budgets
being obtained by the Labour Office. These data of family expenditure
were supplemented by figures prepared by various service organisations showing the relative expenditure on each article or on each
group of articles. From these data was constructed the consumption
of each article by a representative family which was taken to consist
of husband, wife, and two children, one child being assumed to be in
England for his education.
SOURCE
GOVERNMENT OF BOMBAY LABOUR OFFICE

Working Class Budgets in Bombay,
Bombay, 1923.
Bombay Labour Gazette, April 1924.

: Report on an Enquiry into
by G. Findlay SHIRRAS.

IRELAND
An enquiry was conducted in the summer of 1922 with the object
of providing weights for the calculation of cost-of-living index numbers.
A special form was drafted, and some 5,000 copies were despatched
to National School teachers in every school in the country, accompanied
by detailed instructions of the method in which they were to be
completed.
The households of wage earners were to be selected, but any type
could be chosen, no limit being fixed as to the size or composition
of the family, the number of wage earners in the family, or the income.

— 68 —
Thus it was considered that, taking the country as a whole, the budgets
would represent everyday conditions and avoid anything exceptional.
Where the household had a garden in which food was produced, or if
fowl, pigs, etc., were kept, this fact was to be noted on the form.
The agents collecting the data were also asked to give explanations
of exceptional figures.
The particulars asked for included information for each member
of the family as to age, sex, occupation, and earnings and other income,
during the seven days covered. With regard to the various items of
expenditure, the quantities and cost of articles consumed during the
week were asked for. In the case of clothing the total cost during
the year ending June 1922 was asked for, and where the articles
usually last more or less than a year the cost was to be calculated
on an annual basis. The data for clothing were to be given separately
for men, for boys, for women, and for girls. For fuel and light the
expenditure during the previous twelve months was to be given.
When the forms were collected it was found possible to use 308,
which gave completed budgets. The families covered were located
in 112 different towns, and covered wage earners in the principal
industrial occupations.
In addition to taking unweighted arithmetic averages of the results
a more complex method was adopted. It was found by analysis
of the census figures that the proportion of adults (i.e. persons over
14 years of age) to children (under 14) was approximately 5 to 2, and
an adjustment was made to bring the budgets into conformity with
this proportion in order that they might truly represent the actual
proportions of the population of the country, the assumption being
made that the percentage rise in the price of articles mainly consumed
by children might be higher or lower than that in the prices of articles
mainly consumed by persons over 14. The budgets were accordingly
classified into three groups, i.e. simple households consisting of not
more than two persons over 14 years of age, with one or more children
under 14 years of age ; complex households consisting of more than
two persons over 14 years of age with one or more children under
14 years of age ; the adult households in which there were no persons
under 14 years. It was necessary to multiply the last group by three
in order to bring the proportionate numbers of adults and children
in the budgets into conformity with the national proportion of adults
to children (i.e. 5 to 2). Averages were then calculated, and it was
found that the results obtained by applying the weight " 3 " to the
budgets of adult families and combining the figures with those for
simple and complex households were almost identical with those
obtained without applying a special weight to the adult households.
SOURCE
Report on the Cost of Living in Ireland. June 1922.

ITALY
Enquiries have been conducted in different districts with the object
of determining the average consumption of a typical family. These
data have been used in the calculation of the cost-of-living index
numbers published by various municipalities on the general lines
agreed upon at the Statistical Conference held in Milan in July 1920.

— 69 —
In addition to investigations into consumption during the post-war
period certain enquiries had been made before and during the war.
Mention may be made of the enquiry conducted by Professor Pugliese
in July 1913, the results of which are used in the calculation of the
Milan cost-of-living index numbers. In the enquiry, information was
obtained regarding the food consumption of 51 working class families
in different occupations in Milan. Forms were distributed and daily
records kept for periods of 15 to 20 days. Professor Pugliese made
supplementary enquiries in 1916 and 1917.

CITTA DI MILA.NO

numbers.

SOURCE '
: Bollettino municipale, No. 6, 1902, and. subsequent

: Il bilancio alimentare di 5i famiglie operaie
milanesi. By Professeor Angelo Pugliese. Milan, 1914.

SOCIETÀ UMANIATARIA

JAPAN
An enquiry was conducted into the cost of living among skilled
labourers in Osaka by the Municipal Bureau of Labour Research of
that town. Information was obtained from actual records kept by
the families of skilled factory workers during the year June 1919 to
June 1920. Altogether 416 families began to keep records, but only
99 budgets were finally used. Where the information given appeared
to be incorrect — where the workers were living rent free, or where
the records were not kept for a whole year — the information was
not utilised.
The records were kept on blank forms for each month. Instructions with regard to the keeping of the records were given through
the intermediary of the labour leaders. In order to secure the interest
of the families, each family keeping records was given periodically
admission tickets to a variety show or moving picture hall, and at
the end of the year a piece of cotton cloth.
Information was asked for with regard to income, expenditure, the
estimated value of gifts received, remarks regarding guests, sickness,
increase of rent, etc.
Families of different size were reduced to terms of adult males
according to the United States scale. This scale was applied not
only to food, but also to other groups of expenditure. In addition,
however, to this scale, the report publishes the results of an attempt
to establish a more satisfactory scale for clothing.
In the tabulated results data are given for each family as to income
and the chief groups of expenditure. Income, classified according
to source and expenditure on different articles of food, clothing and
miscellaneous items are shown in detail. Averages are calculated by
income groups and size of family.
In addition to the Osaka enquiry, investigations have been conducted in other districts. It is announced that the Japanese Government is planning an extensive enquiry to be conducted during the
year 1 September 1926 to 31 August 1927, when it is hoped to secure
data from over 7,000 families of different classes in the chief
districts of the country.
SOURCE
: Report of Cost of
Osaka, 1921.

OSAKA MUNICIPAL BUREAU OF LABOUR RESEARCH

Living among Labourers in Japan.

— 70 —
LATVIA
The Statistical Office undertook an enquiry into the budgets of
workers' and officials' families in different towns in 1926. Books
were prepared for the recording of details of income and expenditure.
Each book covered a period of one month, and these were to be distributed successively to cover the year commencing 1 April 1926.
. The books included detailed questions with regard to the composition
of the family and its housing conditions. For the recording of income
and expenditure the system of blank pages was adopted, no items
being specified. The collaboration of about 500 families was secured
largely through the intermediary of workers' and other organisations.
The co-operation of other families was obtained by means of notices
in the press. Agents of the Government will keep in touch with
the families and advise them in the recording of the data.
SOURCE
Communication from the

LATVIAN STATISTICAL OFFICE.

NORWAY
Various official enquiries into family budget were made in Norway
before the war, including one conducted in 1912-1913 covering mainly
working class families in Christiania, Bergen, Trondhjem, Drammen,
Kristiansand, and Hamar. In 1912 the Central Statistical Office
conducted an enquiry into the family budgets of commercial employees
in different towns. Information was obtained from 19 employees with
low or average salaries. Separate data are given for each budget and the
German scale was used for reducing families of different size to a
common unit. In addition to these 19 budgets, information was
given on the basis of a budget kept by a lower-grade public official
during the seventeen years 1898-1914. For purposes of comparison,
certain budgets of families with higher incomes were given.
An enquiry covering the period September 1918 to September
1919 was conducted by the Norwegian Central Statistical Office.
Information was obtained for Christiania and Bergen, books being
distributed to the different families for the purpose. Altogether,
82 satisfactory budgets were obtained — 51 for Christiania and 31
for Bergen, of which 48 were of working class families, 20 of lowergrade officials and 14 of higher grade officials \ The 82 families
consisted of 444 persons, of whom 231 were adults (i.e. over 15 years)
and 213 were children. The average number of persons per family
was 5.4 or 4.12 units of consumption. The United States scale was
used for reducing families of different size to terms of a common unit.
Detailed tables give data for each family as to the number of persons,
their age and sex, income classified according to source, and expenditure on the different items. In certain tables data on income and
expenditure are classified according to the occupation of the head
of the family.
In addition to statistics per family, data are given
per unit of consumption. Summaries are given snowing averages
for the families of workers, lower-grade officials, and higher-grade
officials, and also for different groups of families classified according
to income. Comparisons are made with the results obtained in the
1912-1913 enquiry.
1

A small honorarium was paid to them.

— 71 —
SOURCES
Husholdningsregnskaper For Handéisfunkjonaerer M. V. Christiania,
1915
Husholdningsregnskaper, September 1918-September 1919. Christiania,
1921.
NEW ZEALAND
An investigation was carried out in 1910-1911 by the Labour Department on lines similar to those followed in the 1910-1911 enquiry by
the Commonwealth Statistician of Australia. The investigation
was limited to the four chief centres, and as far as possible to bona
fide workers. The agents of the Labour Department were instructed
to eliminate as far as possible from the enquiry households where
boarders were kept or where occupants other than the head were
breadwinners. Likewise, households free of rent or in receipt of an
annual income in excess of £250 per annum were to be excluded.
Nearly 1,800 account books providing for weekly entries of receipts
and expenditure were distributed. The instructions for entering the
information requested were accompanied by a specimen page properly
filled in. Nevertheless, only 69 satisfactory budgets were obtained
when the books were called up in 1911. The results were classified
by income, data being given separately for three groups, namely,
families with incomes under £143 per annum; those with incomes
between £143 and £169 per a n n u m ; and those with incomes over
£169 per annum. Classification by size of family was also introduced, two groups being given — those with more and those with
fewer than four members.
In 1919, the Census and Statistics Office, with the co-operation of
a Christchurch firm, undertook a similar investigation. Instead of
asking for budgets to be kept for a year, the period was reduced to
six months and in order to encourage householders to furnish returns
a pecuniary inducement in the shape of prizes for the most accurately
compiled returns was offered. Again, the results were disappointing,
for although several thousand books were printed only 109 returns
which could be considered of any practical value were received. The
class of person who filled in the books was clearly representative of
the more thrifty part of the population, and thus not truly representative of the whole.
In presenting the results a distinction was drawn between households in receipt of under £4 10s. per week, and those in receipt of
£4 10s. per week and over. Again, data are given separately for
families of different size.
SOURCE
Prices : An Enquiry into Prices in New Zealand, 1891-1919. Prepared
by the New Zealand Government Statistician.
NETHERLANDS (Amsterdam)
An enquiry was made in 1911 by the Socialistische Studieclub
with regard to the expenditure of seventy working class families in
different parts of the country. During and since the war the Central
Statistical Office published various budgets of families of different
classes in the Maandschrift, while similar data have been collected by
the labour inspection. The most important of the enquiries made

_

72 —

by the labour inspection was that in Amsterdam during a period of
four weeks in February and March 1917. In order to allow of results
comparable with those of the 1911 enquiry, the investigation in 1917
was made into the same class of family as was the earlier enquiry,
namely, that of intelligent workers in reasonably comfortable circumstances. The names of the heads of various families were obtained
from the offices of different socialist trade unions. Before making
the investigation, the heads of the families and their wives were asked
to attend a meeting at which the method of conducting the enquiry
was discussed. Altogether 51 households were included, but 8 withdrew before the conclusion of the enquiry. The officers of the labour
inspection conducting the enquiry, three in number, remained in
contact with the families during the whole period in order to assist
in the making of the records. Of the 43 budgets, 4 were found on
examination to be unsatisfactory, so that the final results are based
on data for 39 families. A new enquiry early in 1918 was made, the
same families being asked to co-operate. Actually 28 satisfactory
budgets were collected.
Further enquiries conducted by the Amsterdam Statistical Office
were made in August-September 1918, and from then onwards each
quarter to September 1920, the results being used to indicate changes
in the cost of living 1. A similar enquiry was conducted in March
1922. In this last enquiry the wives who collaborated received a
small payment for the work involved, while instead of the expenditure
being grouped by the families under the different headings as was the
case in the earlier enquiries, the expenditure was entered in chronological order in books provided for the purpose, and the results were
grouped by officials of the Statistical Office. The number of families
covered by these enquiries varied from 25 to 34. The results were
published within three weeks of the period covered. It should be
noted that most of the enquiries conducted in Amsterdam have dealt
with expenditure only and not income, owing largely to the belief
that the recording of income is likely to be inaccurate as the workers
fear lest the data recorded will be used in connection with taxation.
In addition to working class budgets, data were compiled for nonmanual workers, generally officials of the State or the municipality.
The names of those willing to co-operate were obtained by the intermediary of the unions of officials, and especially after an appeal in
their publications. An enquiry was conducted from 1 September
1918 covering a period of a year, 99 families agreeing to take part.
Books were distributed for the entering of expenditure each day,
and for income weekly or monthly. The Office engaged ten visitors
(two of whom were women) to keep in touch with the participants,
who had also attended a meeting at which the method of procedure
had been discussed. The Office assisted the families in other ways.
Thus balances were provided so that the commodities might be weighed.
Actually 82 families, covering 351 persons completed the enquiry.
In a special volume, the Statistical Office published the data obtained
from these families together with the information collected in March
1919 in respect of 32 working class families covering 164 persons. A
special scale, described in the text of the present Report as the Am1
It should be noted that as the items and quantities consumed
would change from one date to another, changes in their cost would
include the effect, not only of changes in the cost, but also in the standard
of living.

— 73 —
sterdam scale, was used for reducing families of different size to units.
The consumption of a conventional family consisting of husband,
wife, and three children of 8, 10 and 12 years of age, respectively, was
then calculated. The data are classified into groups based on total
expenditure ; also the figures for each household are tabulated
separately.
During the period 1 October 1923 to 20 September 1924, a new
enquiry was conducted by the Amsterdam Municipal Statistical
Office in order to obtain statistics with regard to normal consumption
free from the restrictions and price fluctuations of the years of the
war and those immediately succeeding it. About 260 families undertook to collaborate and of these 212 continued the records throughout
the year. These included the families of workers and officials in the
public service and of persons employed in private undertakings.
Daily records of expenditure were made, books with blank pages
for each day being provided. Each book covered one month. The
elaboration of the results follows similar lines to those indicated above
for previous enquiries.
In addition to the Amsterdam enquiries, family budget enquiries
have been conducted in other towns in the Netherlands including
The Hague.
SOURCES
Arbeidersbudgets Gedurende de Crisis. The Hague, 1917.
Les dépenses de 114 ménages de fonctionnaires et d'ouvriers. Communications statistiques du Bureau municipal de Statistique d'Amsterdam, No 73. Amsterdam, 1924.
Supplement to the Bulletin mensuel du Bureau de Statistique de la Ville
d'Amsterdam, November 1925.
POLAND
In November 1922 the Warsaw Municipal Statistical Office conducted
a family budget enquiry 1 . Forms were distributed to 16 families of
officials and 16 families of workers. On these the families were asked
to record their expenditure for the four weeks from 1 to 28 November
•1922. The quantities purchased as well as the cost was to be recorded. Various questions of a general character were asked, e.g.
the occupation of the head of the family, the number of members of
the family, together with their sex and age. Actually only 12 budgets were secured, 6 of officials and 6 of workers. Two of the officials
who sent in budgets were without dependants ; their records were
therefore not used, as they were not typical.
The results were tabulated to show quantities consumed and expenditure per unit of consumption. The German scale was used for
reducing families of different size to units. Separate averages were
given for the families of officials and of workers, as well as a combined
average.
Early in 1926 the Central Statistical Office decided to undertake
an extensive enquiry into the budget of working class families in
Warsow, Lodz and the mining districts. The appointment of agents
to assist in securing the data is provided for. Families completing
satisfactory records will receive a small payment, while prizes are to
be awarded for the best results.
1

Earlier in the year the results of enquiries conducted by Messrs«
and ZAZIARSKI were published.

KRUSZOMSKI

— 74 —
SOURCES
Statistique du Travail, March 1923 : article by Mr. Eugenjusz Budziszewski.
Communication of the Polish correspondent of the INTERNATIONAL
LABOUR OFFICE.

UNITED KINGDOM
During the summer of 1904, of the British Board of Trade collected
a number of working class family budgets. Forms asking for particulars as to income and the main items of expenditure (food and rent)
during one week were sent to a large number of members of workmen's organisations, co-operative societies and individuals. These
were asked to furnish the desired information or to obtain particulars
from fellow-workmen.
When the forms were collected it was necessary to reject a certain
number on the grounds of insufficiency, but 1,944 were left available.
The results of this enquiry, supplemented by those obtained in an
enquiry into rents, retail prices and wages conducted by the Board
of Trade in 1912, have been used to provide the weights for the calculation of the British cost-of-living index numbers.
During the summer of 1918, an enquiry was conducted by a special
committee known as the Working Classes Cost-of-Living Committee.
One of its main objects was to enquire into the actual increase since
June 1914 in the cost of living of the working classes. In this connection family budgets were collected to show the extent to which
changes had taken place in the habits of consumption of working
class families since the outbreak of the war. The method adopted
was similar to that of the Board of Trade in 1904, the attempt being
made to collect budgets from working class families in numbers
sufficient to minimise error and eliminate special peculiarities, and
over an area large enough to make local variations unimportant.
A detailed form, in some respects more precise and more elaborate
than the form used in 1904, was prepared and circulated 1, altogether
10,000 forms being distributed. Questions were included as to expenditure during one week, not only on food, but on all items entering
into the cost of living except clothing. Questions were not asked,
however, as to the income of the wage earners. The forms were
distributed as widely as possible in important centres of population
in Great Britain. No paid investigators were employed. The forms
were entrusted for distribution to the voluntary services of trade
unions, co-operative societies, medical officers of health, the Adult
School Union, guilds of health, and others.
The number of forms available for use was 1,306. These were
classified according to the occupation of the principal wage earner
of the family, into five groups : skilled, semi-skilled, unskilled, clerks,
and " on service " (military or naval). In reducing the average
results obtained, persons of different sexes and ages were reduced
tó a uniform standard unit. This course had not been followed in
the case of the 1904 enquiry. The scale adopted for the purpose
of reduction of the standard unit was that worked out by Professor
Lusk.
As was noted above, information with regard to clothing was not
asked for on the forms circulated. In this case estimates as to ex1

A -copy of the form is given in Appendix II.

— 75 —
penditúre were made on the basis of information supplied by cooperative societies and large distributing houses. The tabulated
results show separate averages for the families of skilled, semi-skilled
and unskilled workers for the different items and groups of expenditure.
SOURCES
Memorandum on the Consumption and Cost of Food in Workmen's
Families. Cd. 2337 of 1904.
Report of the Working Classes Cost-of-Living Committee. Cd. 8980 of
1918.
RUSSIA1
During recent years a number of family budget enquiries have
been conducted by the Central Office for Labour Statistics with the
assistance of local offices for labour statistics and also of local trade
union committees. Thus in December 1922 an enquiry was conducted
covering about 2,000 families in a number of the chief towns. The
families included those of unskilled workers, skilled workers, technicians, and non-manual workers in the chief industries and occupations.
The offices for labour statistics and the trade union councils in
different municipalities have also conducted enquiries. Thus in
Moscow a number of investigations have been made. In an enquiry
conducted in that city in November 1924, the budgets of 66 unmarried
workers and of 413 workers with families were obtained. Subsequently, enquiries have been conducted in Moscow for the period
December 1924 to February 1925, data being obtained for 174 families, and for the period March to May 1925 when information was
supplied by 175 families.
The enquiries are generally conducted by means of agents, who
distributes account books to the families and supervise the keeping
of the records. Entries are made daily of all items of income and
expenditure. In the tabulated results of certain enquiries, including
those for Moscow, averages per consumption unit as well as per
family are given.
SOURCES
Labour Statistics (various numbers). Monthly organ of the CENTRAL
OFFICE FOR LABOUR STATISTICS.

Bulletin of Labour Statistics (various numbers). Published by the MoscowTRADE UNION COUNCIL and the Moscow OFFICE FOR LABOUR
STATISTICS.

SWEDEN
An enquiry into family budgets was undertaken in 1907-1908 by
the Stockholm Municipal Statistical Office. The number of budgets
obtained was 150, in most cases from the poorer classes of the population. In the results details are given for each family as well as
general averages.
In 1913-1914 the Swedish Royal Administration of Labour and
Social Insurance conducted an extensive family budget enquiry with
the object of determining the average budget of the lower classes in
the towns and semi-urban districts in different parts in the country.
Books were distributed, in which the head of each household was
asked to enter details with regard to family income and expenditure

— 76 —
for the year covered 1. In the book the most important items of
receipts and expenditure were named, while blank spaces were provided for the entering of other items. Information as to the sex and
age of the different members of the family was asked for in order that
the data for families of different size might be reduced to a common unit.
The data obtained refers chiefly to the year 1914. The number
of books distributed was 2,325, and the number of families who kept
records during a whole year was 1,619. As 264 of the completed
records lacked adequate uniformity, only 1,355 were actually used.
Of these, 933 were kept by workers, the remainder being kept by
lower-grade non-manual workers in the public services and in private
industry. For reducing data for families of different size to a common
unit, the German scale was used.
In the tabulated results detailed figures are given showing family
income classified according to source, size of family, and income ;
also family expenditure classified according to size of family and income given for over 100 different items of expenditure. Data similarly classified are given also per unit of consumption. In addition
to tables covering all families from which information was obtained,
separate tables give data for manual workers and for non-manual
workers on the lines indicated above. Similar information is given
per unit of consumption for six separate divisions of the country and
also for a considerable number of separate towns.
In addition to statistics showing the amounts expended, data per
unit of consumption are given as to the quantities consumed of the
chief items of food, drink, and fuel and light, and also of the prices
paid in different districts for these items.
In May 1916, June and July 1917, and July 1918, supplementary
family budget enquiries were conducted, accounts for a period of
four weeks being asked for in each case. The number of budgets
obtained in these enquiries was 454, 422, and 390 respectively. The
object of the enquiries was to study the effect of the rise in prices
following the outbreak of the war. Special attention was given to
food, this being the only group for which it was considered that exact
figures could be obtained during a period as short as four weeks. In
the tables, data are given showing for each of the towns covered and
also for different income groups the amounts expended per week per
consumption unit on the different food commodities.
In 1920 an enquiry into the family budgets of workers in rural districts was undertaken in order to complete the information obtained
before the war with regard to town workers. The enquiry was conducted on similar lines to that of 1913-1914. Altogether 372 budgets of the families of agricultural workers and of artisans in rural
districts were obtained, covering generally the year 1920.
A detailed enquiry was conducted in 1923 in order to determine
post-war conditions. It was expected that the results of this enquiry,
which was conducted on similar lines to that of 1913-1914, would be
published in 1926.
SOURCES
Levnadskoslnaderna i Stockholm, 1907-1908. Stockholm, 1910.
Levnadskostnaderna i Sverige, 1913-1914. Stockholm, 1921.
Livsmedelsforbrukningen, 1914-1918. Stockholm, 1922.
Levnadskostnaderna Pä Landsbygden i Sverige, 1920. Stockholm, 1923.
1

Copies of some of the chief pages are given in Appendix II.

— 77 —
SWITZERLAND
The first important family budget enquiry undertaken in Switzerland was that conducted in 1912 by the Secrétariat ouvrier suisse 1.
Official enquiries have been made each year during the post-war
period. These enquiries were conducted by the statistical offices
of the Cantons of Zurich and Basle and of the Municipalities of Zurich
and Berne. The results of the enquiries were compiled at the Federal Statistical Office. In the 1912 enquiry and the four post-war
enquiries, the budget records were kept in a similar way, each family
entering day by day during a period of a whole year each item of income and expenditure. In the case of the 1912 enquiry, however,
the classification of expenses was not so detailed as in the case of
post-war enquiries.
In the different enquiries the families were distributed in somewhat.
varying proportions between large and small towns and other districts. In the case of the 1912 enquiry, most of the families were in
towns with less than 50,000 inhabitants, whereas in the post-war
period they were mainly in towns above that size.
The families from which budgets were obtained in the 1912 enquiry
were those of skilled and unskilled manual workers and of officials
and other non-manual workers or low or intermediate grades. In
the post-war enquiries data were also obtained from the families of
a few higher grade officials. The number of families which supplied
budgets was 791 in the year 1912. In the post-war enquiries the
number were smaller, being, in 1919, 1920 and 1921,277,225, and 323
respectively.
For the 1912 enquiry the " quet " was adopted as unit of consumption. For the post-war enquiries the German scale was
adopted for
reducing families to terms of units of consumption 2. The results
of the enquiry are tabulated according to category of worker, income and districts ; averages per family and also per consumption
unit are given.
SOURCES
: Informations de Statistique sociale ;.
Budgets de familles tenus en Suisse durant l'année 1921 comparés
avec des budgets tenus en 1912, 1919, 1920. Published 1923.
Le Marché du Travail, No. 6, July 1924, pp. 251-255.
Mitteilungen des Statistischen Amtes des Kantons Basel-Stadt, No. 45.
Haushaltsrechnungen von Basler Familien aus den Jahren 1912,
1919-1923. Basle, 1925.
OFFICE FÉDÉRAL

DU TRAVAIL

CZECHOSLOVAKIA
The Statistical Office of the Czechoslovak Republic has conducted
a series of family budget enquiries. The period covered by the en1
Provisional data with reference to this enquiry were available in
1914, but the final results were not published until 1922. See Die
Lebenshaltung schweizerischer Arbeiter vor dem Kriege. Ergebnisse der
Haushaltstatistik des Schweizerischen Arbeitersekreiariats, Zurich, 1922.
2
As noted in the text of the present Report, the Basle Statistical
Office in separate publications of the budgets of families living in the
Basle district has used the " q u e t " scale for food and the German
scale for other groups of commodities.

— 78 —
quiries have been twelve months, detailed records of income and
expenditure being obtained from a small number of families.
The results of enquiries were published as follows 1 :
In 1922, covering 10 families of workers.
In 1923
„
18
„
„
and 25 of officials.
In 1923
„
15
„
„
„ 10
In 1924
„
13
„
„
„
8
In 1925
„
15
„
„
„ 11
The results of these enquiries are tabulated separately for workers
and officials. Averages per head and per consumption unit are calculated. The scale used for reducing families of different size to terms
of a consumption unit is the German scale. The caloric value as well
as the quantity and cost of the different items of food is given. In
addition to averages, separate data are given for each family.
SOURCES
Zprávy Síátniho Úradu Statistickêho, 1922, No. [9 ; 1923, Nos. 1
and 77 ; 1924, No. 105 ; 1925, No. 115.

1

The results of certain earlier enquiries have also been published by

G O V E R N M E N T STATISTICAL O F F I C E .

APPENDIX

II

Specimen Forms Used in Various Enquiries

In order to illustrate the most important questions asked in different budget enquiries, and also the system of recording information,
specimen forms used in a number of important enquiries are reproduced below.
The forms given may be classified into two main groups : first,
those used in enquiries covering a long period during which the householders themselves kept detailed records ; second, those on which
information for a long period is given on the basis of what the householder remembers of the family income and expenditure and on receipts and other accounts in possession of the family, or records for a
short period are kept. In the first group, specimen forms are given
for the German enquiry of 1907-1908, the Australian enquiry of
November 1913, the Austrian (Vienna) enquiry of 1912-1914, the
Swedish enquiry of 1913-1914, and the Swiss enquiry of 1923. In
the second group are given specimen forms used in the British working
class cost-of-living enquiry of 1918, the United States cost-of-living
enquiry of 1918-1919, the Indian (Bombay) enquiry of 1921-1922,
and the Union of South Africa enquiry of 1925.
The Australian and Swedish forms resemble one another in that
pages are provided for each week with separate columns for each day
in the week and a final column for the weekly totals. The Swedish
enquiry, in addition to asking the usual questions about income, expenditure, and the composition of the family, also asks various general questions, particularly as to the nature of housing accommodation.
In both these enquiries, as in many others, the forms specify all the
ordinary items of income and expenditure, space being, however,
provided for " other items " in addition to those specified. The
items of expenditure are classified by groups, e.g. food, fuel and light.
In the German, Swiss, and Austrian enquiries, account books
were provided with separate pages for each day of the week. No
items of expenditure are specified, the householder being asked to
record each item as it occurred. Thus there is no separation of items
into different groups. In the Austrian and Swiss forms space is provided for a summary of the day's income and expenditure. As in
the case of the Swedish enquiry, special forms or pages were provided
in the Austrian, Swiss, and German enquiries for detailed questions
with regard to the general conditions of the family, its welfare, housing accommodation, health, recreation, etc.
In the second group of enquiries referred to above, one—the United
States enquiry— covered a period oí twelve months ; the remainder
covered snorter periods. Although, however, the information ob-

— 80 —
tained in the United States enquiry covers one year, daily records
were not kept. The form is constructed simply for annual totals,
these figures being recorded either by the householder concerned or
by the agent, on the basis of what the householder remembered of
income and expenditure and of receipts and other accounts in his possession. The form is characterised by the great detail in which the
different items of expenditure are specified, there being, for example,
no less than 162 different items of food, while, in the case of clothing, information is asked for separately for male and female members
of the household.
The British form deals only with expenditure, information being
asked for with regard to food, fuel and light, rent, and certain miscellaneous items. Expenditure on clothing was not covered by the
«nquiry. The Indian (Bombay) form covers income and expenditure, details being recorded of expenditure on food, fuel, and rent,
for the month covered by the enquiry. In the case of clothing, a
special page is devoted to estimates based on the durability and cost
of each article, from which the estimated cost per month is calculated.
The South African form also covers income and expenditure, data
covering either a week or a month being requested.

— 81 —
GERMANY
ENQUIRY

OF

1907-1908

Family Budget Enquiry Book, Page 1
Number
Christian name and surname of husband
(If preferred, a letter or number may be given instead of name)
Age of husband
Christian name and surname of wife
(if preferred, a letter or number may be given instead of name)
Number of children (including adopted children)
i of which
sons
.-:.. daughters
Age of children (to be given separately) :
(1) of the sons
(2) of the daughters
Number of other persons belonging to the household :
male
female
Their ages : male
female
(to be given separately)
Occupation and grade of the husband
Weekly earnings :

from
M.

Week
to
Pf.

from
M.

Week
to
Pf

Earnings of the husband
For overtime or subsidiary work
Earnings of the wife
Earnings of the children
Of which amount contributed to
the household . ..'
Income from sub-tenancies
Other income
Total weekly income of the family. ..
(including contributions of relations
to the family)
Yearly earnings of the husband, in Marks
Yearly earnings of the wife, in Marks
Yearly earnings of the children, so far as these are given to the parents,
in Marks
Other yearly income, including sub-tenancies, in Marks
Total yearly income of the family, in Marks
6

— 82 —
Page 2

Number
1907, Month
1. Week
1. Day of month
M.

Pf.

Income
(All income, whether monthly, weekly or daily, to
be recorded on t h e day it is received).
Of t h e h u s b a n d
Of the
Of t h e children, total
M
of which contributed to t h e household

Pf.

Other income

Number

Quantity Weight
in
in
pounds
litres

Expenditure
(to be noted daily, including
t h a t of t h e husband)

Note. — Entry is to be made daily of all income as received, and also of all expenditure,
not only that of the wife for the household, but also that of the husband, etc.
Each commodity, even if several are bought at the same time, is to be recorded separately.
In the case of expenditure covering a given period of time, e.g. house rent, rates, taxes,
coal, potatoes, etc., the period should be stated.

— 83 —
AUSTRALIA
ENQUIRY

OF NOVEMBER

1913

In this enquiry account books were distributed t o t h e householders.
T h e t w o chief pages of this book are given below :
M E M B E R S OF H O U S E H O L D

2 to 29 November 1913
State
Town or nearest post town
Occupation and weekly r a t e of wages of husband.
(State occupation very fully, as " C a r p e n t e r , j a m f a c t o r y " , " Labourer, saw mill " , etc.)
Occupation
"Weekly rate of wages £
s
d
Occupaton of wife
(If t h e wife performs household duties only, insert " Domestic duties")
M E M B E R S OF H O U S E H O L D

( I t is i m p o r t a n t t h a t t h e age and sex of each member of t h e household
who is n o t an adult, should be specified. If t h e number of members
of t h e household changes during t h e four weeks under review, please
give particulars as t o t h e duration of absence, etc.)
N U M B E R , A G E , S E X , AND OCCUPATION OF CHILDREN L I V I N G AT H O M E .

(In t h e column headed " Occupation " , insert whether a t school, n o t
at school, or, if a t work, t h e occupation. If a t work, state t h e r a t e of
wages paid.)
No.

Sex

Age

Occupation

Weekly rate of
wages (it any)

1.
2.
3.
4.
5.

,

6.
N U M B E R AND S E X OF B O A R D E R S

(State length of residence during period under review, and, if not
adults, state ages.)

N U M B E R AND S E X OF SERVANTS

(State length of residence during period under review, and, if not
adults, state ages.)

N U M B E R AND S E X OF OTHER M E M B E R S O F T H E H O U S E H O L D ( I N C L U D I N G
R E S I D E N T V I S I T O R S , ETC.)

(State length of residence during period under review, and, if not
adults, state ages.)

— 84 —
WEEK ENDING 8 NOVEMBER 1913

[The Total Income and Expenditure need NOT be inserted (last column and
last line).]
Sunday

Monday

Tuesday

d.

d.

INCOME

From Earnings of
„
„
„
other receipts
Total

husband
wife
children

1. E X P E N D I T U R E ON H O U S I N G

Rent
Interest on mortgage
Instalments on purchase of house
2. E X P E N D I T U R E ON F O O D

Bread
Milk (included tinned, etc.)
Flour
Tea, coffee and cocoa
Sugar
Butter and cheese
Eggs
Bacon and h a m
Meat (butcher's a n d cooked)
Rabbits, poultry, etc.
Fish (including preserved)
Potatoes
Fruit (including preserved)
Vegetables (including preserved)
Other food
3. H O U S E H O L D R E Q U I S I T E S

Candles
Soap, starch and blue
Kerosene
Gas
Electricity
Coal
Firewood
Other
4. CLOTHING, D R A P E R Y , BOOTS, E T C .
5. O T H E R E X P E N D I T U R E ON L I V I N G

Tobacco, cigars and cigarettes
Alcoholic beverages
Fares (railway, t r a m , etc.)
Insurance (life, fire, etc.)
Contributions to benefit societies
Education fees, school materials
Medical attendance a n d medicine
R a t e s and taxes
Sports and amusements
Charity, church collections, etc.
Wages (servants, grooms, etc.)
6.

MISCELLANEOUS AND SPECIAL
EXPENDITURE X

Total
1

Including renewals ot and repairs to house, furniture, etc.

— 85 —
AUSTRIA (Vienna)
ENQUIRY

O F 1912-1914

Day of the week .

Date

E X P E N D I T U R E FOR T H E D A Y

Quantity (Units, litres,
kilogrammes,
decagrammes, etc.)

Amount paid
Commodity purchased
Kr.

H.

Debt remaining to be paid
Kr.

H.

Kr.

H.

Expenditure for the present day
Debt remaining due for the iDresent dav

Amount remaining from
previous day
Nature of income :

Income for
the day :

Total remaining over from the previous
day, together with the income for the

86 —
SWEDEN
ENQUIRY

A. Composition

of

OF

family

Here insert all persons, each on a separate line, who have participated in the family
consumption during the financial year whether wholly or partially. These will include
children, even if they do not live at home, so far as they take their meals there. It will

No.

Position in family (For exam- Occupation (Particulars of
ple, husband, wife, son,
this even in the case of wife Year and day
and children, so far as they
daughter,
mother-in-law,
of birth
have an occupation produservant, boarder, lodger,
cing earnings in money.)
mealtime guest, etc.)

MEMORANDA

B. General information respecting dwelling, natural advantages, etc.
1. Does t h e family occupy (a) its own dwelling ?
(b) rented dwelling ?
(c) dwelling let free of rent ?
N . B . — "Where the family occupies its own house, t h e particulars
under " C " on the opposite page must be given.
2 . Size of dwelling (number of rooms and kitchen).
(¿>) electric
3 . Is t h e dwelling provided with (a) central heating
00 larder
light
(c)
gas
for
cooking
(d)
water
supply
(pantry)'
(/) cellar for provisions (pantry)
(g) share in
wash-house ?
Has the family the use of a plot of garden land (e. g. p o t a t o land) ?
(a) Is a separate rent paid for this ?
(b) About what quantity of t h e following garden produce has
been obtained during the financial year ?
Quantity

Approximate
value
Kr.

Potatoes
Other garden produce (state kinds and
if possible quantity and approximate
value)
Has the family received any form of wages other t h a n money e.g.
free dwelling, free lighting, fuel, foodstuffs, clothing, garden plot,
free carriage, free travelling, or t h e like ? State each of these
separately, indicating if possible, t h e approximate value in money.

— 87 —

1913-1914
during financial year
also include, not only actual members of the family, but also boarders, lodgers, guests at
meals, etc., taking care to indicate for each ot these the time and extent of their
consumption.

Time of consumption (Here
state whether the person in
question participates in consumption during the whole
of the financial year or only
for a part of same, e.g. a
whole year, 8 months, 6
weeks, etc.)

What part of consumption
(Here state whether the
person in question partici- Other particulars (Capates in the whole consumppable of furnishing
tion or only in part of the
guidance in judging
same, e.g. all except dwelthe composition and
ling, all except dinner each
consumption of the
day, only meals, only lodfamily.)
ging, only breakfast or supper, etc.)

MEMORANDA

6.

7.

1.

2.
3.
4.
5.

Have any gifts in kind been received during the year, e.g. foodstuffs, clothing, shoes, fuel, or the like ?
Indicate every gift of this kind separately, and, if possible, the
approximate value and quantity.
Has the family received any other kind of contribution to its means
of livelihood without expense to the family, e.g. in the form of
fruit picking or mushroom gathering, fishing, etc. ? State the
nature of these, and if possible the quantity and approximate
value.
C. Special particulars for family living in its own house
Property assessment
Kr.
Assessment of site
Kr.
Assessment of house
Kr.
What is the amount of the outstanding debt on
the property ?
Kr.
How much has been paid during the financial
year for amortisation (capital payments
plus interest) ?
Kr.
What is the rate of interest on the said debt ? ..
Kr.
How much has been paid during the financial year
for maintenance charges, taxes and the like
on the property ?
Kr.
D. Other particulars

— 88 —
EXPENDITURE

Week from

to
Sunday Monday Tuesday Wednes
day, etc.
Kr. Ore Kr. Ore Kr. Ore Kr. Ore

19.
Total for week
Kr.

Ore

Milk :
unskimmed
skimmed
cream
Butter
Margarine
[Here follow various other items of food.]

Quantity
Litre

Kilo

Total expenses for food

Other Expenses
Drinks :
spirits
beer, porter, etc.
small beer, mineral waters
Fuel and lighting :
birch wood
pine wood (not including waste wood)
other wood (including waste wood) and
peat
coal and coke
gas
paraffin
other lighting, matches
Rent : per month, per quarter
(strike out if not applicable)
Household effects (purchase and upkeep):
furniture, bed linen, kitchen utensils,
crockery, other household effects, tools
Clothes, mending of clothes, sewing materials
Shoes, purchase, and repairs
Washing and ironing (not done at home)
Soap, washing soap, soda, polishing material, and the like
Medical attendance and medicine
Baths
Children's education, school, books, etc.
Newspapers, reading, etc.
Payments for societies and insurance (give
particulars of these)
Gifts, support to dependants, etc.
Taxes
Consumption away from home (meals and
other refreshments)
Tobacco, snuff, cigars, cigarettes
Amusements and recreations
Travel :
journeys to work, journeys to market
journeys to school, etc.
other journeys
Interest on loan
Various expenses (give description of expenditure hereunder) :
Total of other expenses
Expenses for food
Total expenses

N. B . — Here indicate whether any special circumstances have influenced the income
and expenditure for the week (e.g. absence from home, guests, sickness, unemployment,
receipt of bonuses etc.) :

Litre
Bottle
Cord

Cbm.
Litre

— 89 —
INCOME

Week from

to..
2

•o SS

'S SS »

¿s

m'& ¡1 ¡il e
¡2 Pi

Kr.

Ore Kr.

Ore Kr.

19..

Ore Kr.

I! I

Ore Kr. Ore

3 oS
^•5

¡I

11

Kr. Ore Kr. Ore Kr. Ore Kr. Ore Kr.

Ore

W E E K L Y SUMMARY (OF INCOME A N D E X P E N D I T U R E )

Kr.

Cash in hand at the beginning of the week
Cash income during the
week
Received on loan
Withdrawn from bank
Repaid loans
Total

Kr.

Ore '

Cash payments during the
week
Repayment of loans (not
including interest)
Payments into bank
Money lent
Money in hand at the end
of the week"
Total

Ore

— 90 —
SWITZERLAND »
ENQUIRY

OF

1923

Information for household with control No

at I January 1923.

(a) Head of household : man'—woman'—married'—widowed'—divorced'
Chief occupation
Employer
Subsidiary occupation
Home: Town

Grade in occupation
Work place
Date of birth
Canton or State

(b) Wife : Occupation or employment
Date of birth
(c) Number of children
Christian name

Specify for each :

Date of birth

Occupation or employment

1.
2
3
4.
5
6
(d)

Other persons belonging to the household :
Position in household :
(Relative, servant, lodger,
etc.)

Date of birth

Earning capacity

1.
2

:

3
(e) Housing accommodation : Do you live in a dwelling for which you
pay rent * or in your own house ' ?
Floor
No. of rooms
Attics
What was the yearly rent or rent value in January 1923? Fr
Since when have you lived in this dwelling? (Month and year)
(/) Have you a garden or plot of land?

If so, what area.

sq. ms.

(g) Do you keep domestic animals?
rabbits, hens)
(ft) Remarks

If so, what kind ? (dog, cat,

1
Books with blank pages for each day closely resembling the German and Austrian
pages for daily records, given above, were distributed for the entering of details of income
and expenditure.
' Underline whichever applies.

— 91 —
UNITED KINGDOM
T H E W O R K I N G CLASSES COST O F L I V I N G C O M M I T T E E E N Q U I R Y
O F 1918

You are'asked to state the quantities and kinds of all foods consumed
by you in the week Friday, 31 May to Thursday, 6 June 1918, and to
state your usual weekly expenditure on other household necessaries.
Your answers will be kept strictly secret and confidential.
P A R T I. — T H E HOUSEHOLD

Members of household
(Give names, or, if preferred,
relationship to head of
household)

Sex
(M. or F.)

Occupations
(Describe these in detail)

Age if under
18 years

Earners

Is the head of the household living away from
home (a) on naval or
military service or (f>)
at work?
Non-earners

P A R T II. — E X P E N S E S O T H E R T H A N FOOD, P E R W E E K D U R I N G R E C E N T W E E K S
Amount
spent

Quantity

d.

Coal
Oil
Candles
Matches
Firewood

per
per
per
per
per

week
week
week
week
week

Rent (including rates)
Insurance, thrift clubs, etc
Gas or electric light
Tram, bus or rail fares
Soap, soda, blue and other cleaning materials .

.. cwts.
quarts
lbs.
.. boxes

per
per
per
per
per

week
week
week
week
week

— 92 —
PART III. — FOOD : ACTUAL CONSUMPTION DURING THE WEEK COMMENCING 31 MAY

A. — Food Consumed at Home
Quantity

s.

s.

d.

Total spent on food prepared
and eaten at home

Produce

C. — Meals Bought and Eaten Awt
from Home, such as Canteen or R
freshment House Meals or Scho
Meals for the Children

N o . oi
Tota
meals
in the cost in 1
weel
week

Husband
Wife
Children under 18 :
1
2
3
Other members of
household

Signed .
Address.
Date.

d

Other foods bought :
Kinds :

•

Eggs
Other produce :
Kinds :

Amoui
spent

Macaroni
Haricot and dried beans
lentils and peas
Eggs
Cheese
Honey
J a m or marmalade
Syrup or treacle
Sugar
Pickles and sauces
Condensed milk
Fresh milk
Fresh fruit
Dried, tinned or bottled
fruit
Potatoes
Fresh vegetables
bought :
Kinds :

Beef
Mutton
Pork
Bacon and H a m
Sausages
Tinned meat
Other meat, such as
poultry, liver, tripe
etc.
Fish.— 1. Fresh
2. Dried
3. Tinned
Bread
Flour
Cake, buns, etc.
Butter
Margarine
Biscuits
Suet
Lard, dripping, etc.
Tea
Cocoa
Coffee
Oatmeal and rolled oats
Rice, sago, tapioca, etc.

B. — Home or Allotment

Quantity

Amount
spent

— 93 —
UNITED STATES
ENQUIRY

O F 1918-1919

UNITED STATES
DEPARTMENT
OF LABOUR
Bureau of Labour Statistics - Washington 1
Cost of living in the year ending
1. Name of head of family
3. Street and number

Family, relation to
husband

6.

7.

Sex

Age

2. City and State
4. White or coloured
8.
9.
Weeks Industry and
in
occupation or
home
status.

10.
Meals
per
day

12.
11.
13.
Paid Weeks Rate
emfor B . ployed
of
and L.
wages

(a) Husband
(b) Wife
(c) Eldest child
(d) Next child
(e) Next child
(/) Next child
(?)
(/>)

(0
15. Summary of income :

16. Summary of expenses :

(a)
(ft)
(c)
(d)

(a)
(b)
(c)
(d)

(e)
(/)
(g)
(ft)

Earnings of family
Board and lodging
Net from garden, chickens, etc...
Gifts : money, food, clothing
etc.
Net from rents, interest, etc. .
Fuel picked up
Other (specify)
_.
Total .

Food
Clothing
Housing-rent
do. on owned home

(«) Fuel and light
(/) Furniture and furnishings
(g) Miscellaneous
(ft)

17. Amount of surplus or deficit
18. If surplus, how used? If deficit, how met?.

1

Various instructions and notes omitted.

Total

14.
Earnings

— 94 —

EXPENSES FOR FOOD DURING YEAR »

[Report all articles bought, whether paid for or not, and the market value of food raised
or received as a gift]
Quantity

Items

19.
20.
21.
22.
23.
24.

Beef, fresh steak
Beef, roast
Beef stew
Beef
Beef
Beef, salt, corned

f

lbs.
lbs.
lbs.
lbs.
lbs.
lbs.

Quantity

Items

Cost

Cost

;

25. Beef, dried
lbs.
26. Veal
lbs.
[Here follow other articles
of food (Nos. 27 to 162)]
163. Lunches and meals
bought outside
No.
164. Total cost of food

1
Similar pages are provided for recording the quantity purchased during the year and
the cost of fuel and light (11 items), furniture and furnishing (62 items), and miscellaneous
expenses (62 items).

EXPENSES FOR CLOTHING DURING YEAR '

Males *

Items

Husband

Other males (insert age of each person
as given on first page)
Age

Age
No.

Cost

No.

Cost

No.

Cost

Headwear :
193. Hats, felt
194. Hats, straw
195. Caps
Outer garments :
196. Suits : wool
197. Suits : Cotton
198. Coats (separate)
[Here follow other items of
male attire (Nos. 199 to
249.]
250. Other clothing
(note if over $ 5)
251. Total for males
1
On the form, instructions were given for the recording of expenditure for materials
bought and made up by the family ; also the value of any clothing received as gifts.
* Space was provided similarly to record expenses for articles of clothing worn by females.

— 95 —
EXPENSES FOR HOUSING DURING YEAR

165. Type : A. Single house : Detached
semi-detached
row
end of row with Windows in t h e side
166.

B. F l a t : Detached
semi-detached
end of row with windows in t h e side

167.

C. Apartment : W h a t floor
outlook on front
two sides

168. Material : Brick

elevator
side

rear

frame

169. Interior : Plastered
170. Rooms :

row
what floor

ceiled

Total number
cellar

171. Sanitation : W a t e r
water

wall board

also p a n t r y
b a t h room

attic

inside
in yard
running hot water

running

172.

W a t e r closet inside
sole use

173.

Stationary l a u n d r y tubs
connection

174. H e a t :

Steam
hot water
hot air
fireplace
latrobe
equipped for heating

175. Light :

Total windows
number of sleeping rooms without
outside windows
number of other rooms without outside windows

176. R e n t includes heat
kitchen stove

outside

Rented home :

sink

light.:

177. Owned house would rent for

privy
sewer

stove
number of rooms

hot water
per year.
Owned home :

g

178. Rent for year

184. Paid on principal

179

185. Paid on interest

180. Repairs

186. Taxes

181. W a t e r rent if separate
from rent

187. Special assessments

182
183.

^ ^
Total

/

$

188. Repairs and improvements
189. W a t e r rent
190. Insurance
191
192.

^ ^
Total f

— 96 —

INDIA (Bombay)
ENQUIRY

OF

1921-1922

Information for use in the Labour Office, Secretariat, Bombay.
Reference No
District
Family Income and Expenditure for Month of
192
Address
Caste of family
Race to which head of family belongs
Number in Family
Men

Women

Children
under 14

Number residing at above address
Dependants not living at above address
Family income of those residing at above address during month of
192
:

Earnings in above month
Occupations of wage
earners

Rs.

Men
Women
Children
Total amount of family income

Overtime
pay

Wages
a.

Rs.

a.

Additional
earnings
if any
Rs.

a.

Total
Rs.

a. .

— 97 —
EXPENDITURE ON FOOD, FUEL, LIGHTING AND RENT »

Commodities
(Note. — This list may
be altered according to
what is bought)

Total Total
quancost
tity
in
in
month month

Commodities
(Note. — This list may
be altered according to
what is bought)

Total
quan- Total
cost
tity
in
in
month month
R s . a.

Rs. a.

Rice
Wheat
W h e a t flour
Barley
[Here follow 25 other
items of food]

Coal
Charcoal
Firewood
Kerosene
Other fuel and lighting

Refreshments bought
and consumed away
from home

Total expenditure on
food, fuel and lighting
Rent per month

1
A separate page was provided for recording the cost per month of various miscellaneous items. These Included barber, medical fees and medicines, education, travelling
expenses to and from work, tobacco, liquor, amusements, cooking pots and other household necessities, payments to dependants not residing with the family, and interest on debts.

EXPENDITURE ON CLOTHING, SHOES AND OTHER NECESSARIES NOT
BOUGHT EVERY MONTH
Articles
(Note. — The
list of articles
should be
extended so as
to include all
those in use.)

Number
of articles
in use

Cost
per article
when
bought

Total cost
of
articles

Estimated
number
of months
that each
article
will last

Clothing :
[Here are
specified in
separate
groups,
articles of
clothing for
men,
women and
children.
Expenditure
for
bedding is
included on
this form.]
Total

Estimated
cost
per month
on average

— 98 —

NOTES OF

INVESTIGATOR

Description of dwelling :
Number of rooms occupied by family
Approximate size of each room
State whether the dwelling is in a good state of repair or is dilapidated

State of what materials the dwelling is built
Give details as to supply of water for drinking, and sanitary arrangements
State here whether the family buys on credit or cash system
If interest was charged on credit purchases, state, if possible, what
was the rate of interest
Religious ceremonies including marriages and funerals
Give any information available as to the occasional expenditure on
Festivals
Marriages
Funerals
State to what extent the money so spent was borrowed

— 99 —

UNION OF SOUTH AFRICA

E N Q U I R Y OF

1925

Department of Labour: Cost-of-Living

Committee

Budget oí Family Expenditure
Town

Magisterial District

Occupation of head of family
Salary or wage earned by head of family ¿g
Income from other sources, e.g. investments, earnings of other members
of family, rent of rooms let to lodgers, etc
Number of persons in family (excluding boarders or lodgers, but including relatives or friends (non-paying guests) resident with the
family for the whole period covered by this return)
Adults

Male children (state ages)
Female children (state ages)

State whether boarders or lodgers are kept, and if so state number
Materials of which outer walls of house are constructed (e.g. brick,
stone, concrete, wood and iron, etc.)
Number of rooms in house :
Bedrooms
dining or sitting room
pantry
bathroom

kitchen

— 100 —
Household expenditure during month or week ended (delete period
which does not apply)
1925
[Noie : Figures may be given for a period of a week or a month. If the former,
care should be taken to state whether a weekly or a monthly rental is given.]
Value
£.
Foodstuffs :
Bread (actually purchased)
Flour (including " seconds ")
[Here follows 30 other food articles, together with
space for the entering of unspecified articles of
food]
Fuel and light :
Coal
Bags of 200 lb.
Wood
Bags
Paraffin
Tin
Candles
Pkt.
Matches
Pkt.
Electric light
Units
Rent :
Sundries :
„
Clothing
„
Men's
„
Women's
„
Children's
Material purchased for making into clothes
Boots and shoes
Boot and shoe repairs
Furniture
Crockery and household utensils and linen (not
furniture)
Tobacco
Liquor
Soap, blue, starch, soda and other washing and cleansing materials
Washing (if not done at home)
Servants' wages
Newspapers, books and periodicals (including subscription to library or book club)
Train, t r a m or 'bus fares
Amusement and recreation
Holiday expenses
Rates and taxes (water, sanitary, etc.) actually paid
by the householder
Medical and dental attendance and medicines or,
alternatively, subscription to medical fund or
benefit society
Insurance premiums
Subscription to building society
Subscription to friendly society
Subscription to trade union
School fees, school books and other expenditure on
education
Other items
not household
detailed above
(specify)
Total
expenditure