Preview

INTERNATIONAL LABOUR OFFICE

STUDIES AND REPORTS
Series N (Statistics) No. 25
(Revision of No. 19)

THE

INTERNATIONAL STANDARDISATION
OF LABOUR STATISTICS
A Review of the Statistical Work of the International Labour Office
and of Various International Statistical Conferences

MONTREAL,

1943

PUBLISHED BY THE INTERNATIONAL LABOUR OFFICE,

3480 University Street, Montreal, Canada
Published in the United Kingdom for the INTERNATIONAL
by P. S. King & Staples, Ltd., London.

LABODR OFFICE

Distributed in the United States by the INTERNATIONAL.LABOUR OFFICE,
Washington Branch, 734 Jackson Place, Washington,^.C.

PREFACE
Good labour statistics are essential to both the framing and
the administration of progressive social policy. To serve this
purpose fully the statistics of employment, unemployment, wages,
cost of living, family expenditures, industrial accidents and other
aspects of industrial life and labour must be accurate and complete; and they must be comparable both within and between the
different countries.
Progress towards this goal is necessarily gradual but it can be
hastened by pooling the experience of the several countries and by
formulating international standards based on the lessons of that
experience. The work already accomplished in this direction
through the International Labour Organisation forms the subject
of the present report; and it is hoped that by presenting in convenient form the recommendations of successive international
conferences of experts on labour statistics the report will serve
to furnish practical assistance to statisticians and to stimulate
further progress towards the international standardisation of labour
statistics.
Since the publication in 1934 of the first edition of this report,
summarising the work of the International Labour Office and of a
series of International Conferences of Labour Statisticians in this
field, a number of new developments have taken place. The most
important of these is the adoption by the International Labour
Conference in 1938 of a Draft Convention concerning Statistics
of Wages and Hours of Work. This Convention established a set
of minimum standards on these subjects to which the different
Governments may adhere by appropriate action, thus converting
into a definite obligation the production and publication of statistics in these fields in accordance with these standards. Besides
this Convention, the present revised edition of this report adds
the recommendations of the International Labour Office Committee of Statistical Experts on international comparisons of wages,
food costs and rents, as well as extracts from the reports prepared
by the Committee of Statistical Experts of the League of Nations,
on which the Office is represented, on the statistics of the gainfully

INTERNATIONAL STANDARDISATION OP LABOUR STATISTICS

occupied and of housing. In the revision the analysis of the work
of the Office in the whole field of labour statistics is amplified and
brought up to date.
As before;, the report includes the resolutions and recommendations on wag;e statistics of the two Conferences called by the Social
Science Research Council, and those on housing statistics of the
Committee of Statistical Experts of the International Union of
Towns, in all of which the International Labour Office collaborated.
In the revision, the resolutions are arranged by topics and
chronologically under each topic, instead of chronologically according to the dates of the respective meetings, as in the original edition.
It is believed that the present arrangement, by bringing together
all the resolutions under each topic, will make the volume of greater
service to all those interested in the international standardisation
of the different branches of labour statistics. A key is given in
the Appendix showing the topics considered at each of the International Conferences of Labour Statisticians. An index is added
to make the materials conveniently accessible.

CONTENTS
Page
PRBFACB

PART I
THE WORK OF THE INTERNATIONAL LABOUR OFFICE AND OF THE
INTERNATIONAL STATISTICAL CONFERENCES IN THE
STANDARDISATION OF LABOUR STATISTICS
INTRODUCTION

CLASSIFICATION OF INDUSTRIES AND OCCUPATIONS

II.

EMPLOYMENT AND UNEMPLOYMENT

III.

WAGES AND HOURS OF WORK

Money Wages
Index Numbers of Wages
Total Wages and Salaries
Hours of Work

16
19
20
20

IV.

COST OF LIVING

INTERNATIONAL COMPARISONS OF REAL WAGES AND COST OF LIVING.

Wages
Elements of Remuneration other than Wages
Cost of Living
Food

25
26
27
28

Rents

VI.

FAMILY LIVING STUDIES

VII.

INDUSTRIAL ACCIDENTS

VIII.

HOUSING

IX.

COLLECTIVE AGREEMENTS

INDUSTRIAL DISPUTES

XI.

EMIGRATION AND IMMIGRATION

XII.

YEAR BOOK OF LABOUR STATISTICS

CONCLUSION

PART II
DOCUMENTARY TEXTS
I.

THE CLASSIFICATION OF INDUSTRIES AND OCCUPATIONS FOR PURPOSES OF LABOUR STATISTICS

1. Resolutions of the First International Conference of Labour
Statisticians, Geneva, October 1923
2. Resolution of the Second International Conference of Labour
Statisticians, Geneva, April 1925

49
50

II.

INTERNATIONAL STANDARDISATION OF LABOUR STATISTICS
3. Resolutions of the Third International Conference of Labour
Statisticians, Geneva, October 1926

4. International Scheme for the Presentation of D a t a by Industry
(or Occupation), framed by the International Labour Office..

5. Excerpts from the Report of the Committee of Statistical
Experts of the League of Nations on Statistics of the Gainfully-occupied Population

6. Minimum Nomenclature of Industries recommended by the
Committee of Statistical Experts of the League of N a t i o n s . .

STATISTICS O F EMPLOYMENT AND UNEMPLOYMENT

7. Resolutions of the Second International Conference of Labour
¡Statisticians, Geneva, April 1925
III.

IV.

STATISTICS O F W A G E S AND H O U R S O F W O R K

8. Resolutions of the First International Conference of Labour
¡Statisticians, Geneva, October 1923

9. Resolutions of the First International Conference of Statisticians convoked by the Social Science Research Council of the
United States of America, January 1929

10. Resolutions of the Second International Conference of Statisticians convoked by the Social Science Research Council of the
United States of America, May 1930
11. Convention concerning Statistics of Wages and Hours of Work
adopted by the International Labour Conference a t its 24th
Session, June 1938

12. Proposed Recommendation suggested by the Fifth International
Conference of Labour Statisticians, Geneva, October 1937. . .
13. Proposals Calling for Further Action

95
96

C O S T OF L I V I N G I N D E X N U M B E R S

14. Resolutions of the Second International Conference of Labour
Statisticians, Geneva, April 1925
15. International Convention relating to Economic Statistics, of 14
December 1928, adopted by the International Conference relating to Economic Statistics called by the League of N a t i o n s . .
16. Recommendation of the International Conference relating to
Economic Statistics called by the League of Nations
V.

INTERNATIONAL
L::VING

COMPARISONS

OP R E A L W A G E S

AND COSTS

100
100

OF
101

17. Resolutions of the Second International Conference of Labour
Statisticians, April 1925
101
18. Resolutions of the First International Conference of Statisticians convoked by the Social Science Research Council of the
United States of America, January 1929
101
19. Resolutions of the Fourth International Conference of Labour
Statisticians, Geneva, May 1931
104
A. Statistics of Elements of Remuneration Other than Wages
107
20. Resolution of the Committee of Statistical Experts of the International Labour Office, First Session, Geneva, 12-15 December 1933
107

CONTENTS
B. Statistics of the Cost of Living

vii
108

21. Resolutions of the Second International Conference of Statisticians convoked by the Social Science Research Council of the
United States of America, May 1930
C. Statistics of the Cost of Food

108
109

22. Resolution of the Committee of Statistical Experts of the International Labour Office, First Session, Geneva, 12-15 December 1933

109

D. Statistics

of Rents

23. Resolution of the Committee of Statistical Experts of the International Labour Office, First Session, Geneva, 12-15 December 1933

Ill

24. Resolutions of the Committee of Statistical Experts of the International Labour Office, Second Session, Geneva, 16-18 December 1935
112
25. Resolutions of the Committee of Statistical Experts of the International Labour Office, Third Session, Geneva, 6-7 October
1937
114
VI.

VII.

M E T H O D S OP FAMILY L I V I N G E N Q U I R I E S

26. Resolutions of the Third International Conference of Labour
Statisticians, Geneva, October 1926

116

27. Recommendation of the Committee of Statistical Experts of the
International Labour Office, First Session, 12-15 December
1933

118

STATISTICS OF INDUSTRIAL ACCIDENTS

28. Resolutions of the First International Conference of Labour
Statisticians, Geneva, October 1923
VIII.

HOUSING STATISTICS

29. Recommendations of the Committee of Statistical Experts convoked by the International Union of Towns ( M a y 1 9 2 8 ) . . . .
30. Excerpts from Recommendations of the Committee of Statistical Experts of the League of Nations
IX.

STATISTICS OF COLLECTIVE AGREEMENTS

31. Resolutions of the Third International Conference of Labour
Statisticians, Geneva, October 1926
X.

STATISTICS OF INDUSTRIAL D I S P U T E S

32. Resolutions of the Third International Conference of Labour
Statisticians, Geneva, October 1926
XI.

116

STATISTICS OF MIGRATION

33. Resolutions of the International Conference of Migration Statisticians, Geneva, 3-7 October 1932

119

119
122

122
128
134

134
138

138
143

143

A P P E N D I X : Key to Topics considered at each International Conference
of Labour Statisticians

149

INDEX

153

PART I
The Work of the International Labour Office and of the
International Statistical Conferences in the
Standardisation of Labour Statistics

INTRODUCTION
From the outset, the International Labour Office, having as
one of its duties "the collection and distribution of information
on all subjects relating to the international adjustment of conditions of industrial life and labour" 1 undertook to compile and
publish statistics in these fields. Obviously, if these labour statistics were to have significance, the first task was to study the methods
of compilation in the different countries with a view to bringing to
light differences in techniques employed and to appraise the importance of these differences. Such a comparative study naturally
suggests ways of improving the statistics by adopting the best
methods, and an immediate objective, therefore, was to obtain a
consensus of opinion on which methods were the best. An objective
scarcely less important was a greater degree of comparability in
the statistics of the different countries. The result of comparing
labour statistics in different countries was thus to emphasise the
importance of international standardisation.
The work of the Office in this direction has followed a number
of different techniques. In the first place, a series of International
Conferences of Labour Statisticians has been called by the Office
to make recommendations as to the best methods of compiling
data on the various topics in the field of labour statistics. These
have been supplemented by a conference on housing, called by the
International Union of Towns, two conferences called by the
Social Science Research Council of the United States, in all of
which the Office collaborated, as well as a conference on migration
statistics called by the Office. All these conferences of statisticians
have adopted resolutions specifying statistics which should be
gathered on labour topics. In this connection may be mentioned
the work of the Committee of Statistical Experts of the League of
Nations, appointed in accordance with the terms of the League's
Convention concerning Economic Statistics; this Committee, on
which the Office is represented, has issued a number of reports
dealing with international standardisation of statistics, including
two on topics of labour statistics discussed in the present report.
1

Article 9 of the Constitution of the International Labour Organisation.

INTERNATIONAL STANDARDISATION OP LABOUR STATISTICS

A second lim; of development has been the appointment by the
Governing Body of the International Labour Office of a Committee
of Statistical Experts, to advise on international comparisons of
wages and cost of living, as well as to make preparation for international conferences: the resolutions of this Committee have broken
new ground in this difficult field of international comparisons. A
third development in the technique of international standardisation, is the adoption by the International Labour Conference of a
Convention concerning Statistics of Wages and Hours of Work.
A fourth type of effort consists of direct and indirect aid to the
statistical offices of the different governments including missions
to different countries, and other less formal methods of consultation and collaboration, which have carried forward the work of
international standardisation. A fifth group of efforts consists in
collaboration with other international organisations in the standardisation of statistics. Finally, the whole series of statistical publications of the Office, culminating in the Year Book of Labour Statistics,
tend to stimulate interest in this standardisation and record its
progress. Each of these lines of development will be discussed
briefly.
The first of a series of international conferences of official statisticians engaged in the compilation of statistics relating to labour
was called in 1923 by the International Labour Office following a
suggestion to this effect made by certain governments. This was
followed by other conferences in 1925, 1926, 1931 and 1937. All
the principal industrial countries sent representatives.
The decisions of these conferences in no way committed the
governments, nor was it intended to prepare a code of strict and
complete rules for the classification and compilation of statistical
data. Their aim was merely to point out the general lines to be
followed in the different branches of labour statistics and to make
suggestions as to the direction in which improvements should be
made when opportunity arose. These suggestions, representing
the considered views of the responsible experts in the field, would,
it was believed, have great weight in influencing the future development of labour statistics in the direction of better data and of
greater uniformity.
At the first Conference (1923) the agenda covered statistics
of wages and hours of labour, statistics of industrial accidents and
the classification of industries and occupations; the agenda of the
second Conference (1925) comprised, in addition to a second discussion on the classification of industries and occupations, the
question of cost-of-living statistics and the statistics of employment and unemployment; the agenda of the third Conference

INTRODUCTION

(1926) dealt with methods of compiling family budgets, the statistics of collective agreements and those of industrial disputes. The
fourth Conference (1931) dealt with the question of international
comparisons of real wages, and the fifth (1937) formulated a proposal for a Convention concerning statistics of wages and hours of
work, and referred it to the International Labour Conference
which, in 1938, adopted the Convention.
These conferences, together with those on migration and housing
to be discussed in the next paragraphs, covered almost the whole
field of what is commonly known as "labour statistics". 1 On each
of the above-mentioned subjects the Office made and circulated to
the delegates a preparatory technical study of the aims and objects
of the statistics under discussion and the methods adopted in the
different countries, together with suggestions for improving their
value and comparability. 2
Independently of the conferences of official labour statisticians,
certain other conferences of similar nature, in which the Office
closely collaborated, were held in 1928, 1929 and 1930. At each
of these meetings the International Labour Office was invited to
take part: it undertook the secretarial work and the Statistical
Section of the Office supplied memoranda on the subjects of the
agenda. The first of these was a meeting of a committee of statistical experts (from 10 European countries) called by the International Union of Towns at Munich in 1928, in order to lay down
guiding lines for statistics of housing.8 The two others were convened in 1929 and 1930 by the Social Science Research Council
of the United States of America, to study the statistics of wages,
and experts from six industrial countries—the United States,
Canada, France, Germany, Great Britain and Italy—were invited. 4
Closely related to these was a special conference of statisticians
1
See Karl PRIBRAM: "The Scope of Labour Statistics", in International
Labour Review, Vol. XIV, No. 4, Oct. 1926, pp. 476-488.
2
See below under each topic for detailed list of references, including the
discussions of the conferences.
' In the case of the Committee on housing statistics, the memoranda of the
experts and the resolutions adopted were published in the official organ of the
International Union of Towns (L'Administration locale, May-June, 1928). See
also below under Housing, p. 35.
4
In the case of these two Conferences, copies of the memoranda supplied by
the representatives of each of the countries and the memoranda presented by the
International Labour Office, together with a report on the existing wage and costof-living statistics in each of the countries represented, were published in a
volume entitled International Wage Comparisons, issued in 1932 in Great Britain
by the Manchester University Press, and in the United States by the Social
Science Research Council. The resolutions of the first Conference, together with
a short account of its work, were published in the International Labour Review,
Vol. XIX, No. 4, Apr. 1929, pp. 522-539; the resolutions of the conferences are
given in full in Part II of the present report.

INTERNATIONAL STANDARDISATION OP LABOUR STATISTICS

dealing with emigration and immigration called by the International Labour Office1 in October 1932.
In connection with these conferences and their resolutions,
the work of the Committee of Statistical Experts of the League
of Nations, in which the International Labour Office participated,
should be mentioned. 2 This Committee was established and constituted in accordance with the provisions of Article 8 of the International Convention relating to Economic Statistics of 1928,
which was adopted by the International Conference relating to
Economic Statistics called by the League of Nations. The membership of the Committee, appointed by the Council of the League,
included statisticians from the different countries with well developed statistics, and representatives of the International Labour
Office and the International Institute of Agriculture. The practice
normally adopted by the Committee was to submit preliminary
recommendations concerning statistical methods to the Council of
the League oi Nations for comment by the Governments and to
formulate definitive proposals in the light of the comment received.
Two of the reports of this Committee, those on the gainfully occupied population and on housing are in the field covered by the
present study; and in Part II excerpts from these reports, including
definitions and classifications recommended, together with the
reasons for the recommendations, are presented.
A second line of development in the work of international
standardisation was the appointment by the Governing Body of
the International Labour Office, in accordance with a recommendation of the Fourth International Conference of Labour Statisticians,
of a Committee of Statistical Experts. This Committee was organised as a committee of technical experts to advise the Governing
Body on matters particularly relating to problems of international
comparisons of wages and cost of living, to prepare for International Conferences of Statisticians and to consider special problems
referred to it by the Governing Body. Though its functions were
thus somewhat narrowly defined, the Committee has adopted a
series of recommendations which have pointed the way to fruitful
lines of development in the difficult problems of international
comparisons of wages, cost of living, food costs, and rents. Its
1
For a discussion of this Conference, see "Report of the Conference on Migration Statistics", in International Labour Review, Vol. XXVII, No. 1, Jan. 1933,
pp. a 1-24.
For a review of the work of this Committee, see Michel HUBER: "Le Comité
d'experts statisticiens de la Société des Nations (1931-1939)", in Revue de l'Institut international ¿a Statistique, 7th year, livraison 2/3, 1939, pp. 117 e/ seq. A brief
account is given in Charles K. NICHOLS: "The Statistical Work of the League of
Nations in Economic, Financial and Related Fields", in Journal o} the American
Statistical Association, Vol. 37, No. 219, Sept. 1942, pp. 336-342, esp. 339-340.

INTRODUCTION

work, furthermore, in preparing for the Convention concerning
Statistics of Wages and Hours of Work is of noteworthy importance.
The Committee held meetings in December 1933, December 1935,
and October 1937; plans for a further meeting in the spring of
1940 had to be dropped owing to war conditions. The agenda of
the first meeting included international comparisons of the cost
of food and rent, family living studies, and elements of remuneration other than wages; at the second meeting the agenda covered
international comparisons of rents, and the proposal for an international convention concerning statistics of wages and hours of
work, which was subsequently discussed by the Fifth International
Conference of Labour Statisticians in 1937 ; and at the third meeting,
the topics discussed included again international comparisons of
rents, the systématisation of statistics of wages and hours of work
published by the Office, and a proposal for an international enquiry
on wages and hours of work in the textile industry. The last mentioned project had been referred to the Committee of Statistical
Experts by the Governing Body of the Office, following a discussion
of the subject by a special committee of the Tri-partite Technical
Conference on the Textile Industry which was convened by the
Office and which met in Washington in April 1937.1
The third line of development of the techniques for international
standardisation is represented by the adoption by the International
Labour Conference in 1938 of the Convention concerning statistics
of wages and hours of work. This Convention incorporates a
minimum programme for statistics in this special field and the
States Members of the International Labour Organisation were
invited to ratify it and to signify by so doing their willingness to
collect the statistics on this minimum plan and submit them to
the Office.2
The preparation of this Convention involved three stages:
first, a consideration in detail by the Committee of Statistical
Experts of a draft of the proposed articles of the Convention prepared by the Office ; secondly, the approval of the Convention by
an International Conference of Labour Statisticians called for the
purpose and including experts from the different Member States, and
1
For a discussion of these three meetings, see Michel H U B E R : "Le Comité
d'Experts statisticiens du Bureau international du Travail (1933-1939)", in
Revue de l'Institut international de Statistique, 9th year, livraison 3/4, 1941, pp.
104-112. For the report of the Committee on Statistics of the Tripartite Technical
Conference on the Textile Industry, see The World Textile Industry, Economic
and Social Problems, Vol. II, Studies and Reports, Series B, No. 27 (Geneva,
1937), pp. 273 et seq.
* See J . W. Nixon: "An International Convention on Statistics of Wages
and Houre of Work", in Journal of the Royal Statistical Society, Vol. CI, Part
IV, 1938, pp. 730-735.

INTERNATIONAL STANDARDISATION OF LABOUR STATISTICS

finally, its adoption by the Twenty-Fourth Session of the International Labour Conference at which the Governments, workers
and employers of the Member States were represented.
A fourth phase of the activity of the International Labour
Office has been the sending of experts to various countries for
consultations with regard to the further development of labour
statistics. In particular a series oí visits to various American
countries arranged during 1940, 1941 and 1942 may be cited as an
illustration., In this type of effort may be included also the numerous
less formal activities in the way of specific aid to the statistical
offices of the different countries, supplying documents, references,
or special memoranda on different topics as requested, and serving
as a clearing house of information on the standardisation of labour
statistics.
A fifth phase of activity has been the collaboration of the Office
with other international organisations interested in international
standardisation or in specific fields of labour statistics. Besides
the Committee of Statistical Experts of the League of Nations,
the work of which has already been mentioned, the Office has collaborated with the Health Organisation, the Permanent Committee on Road Traffic, and the Institute of Intellectual Co-operation of the League, and with other international organisations and
committees, including in particular, the International Statistical
Institute and the recently formed Inter-American Statistical
Institute. The exact scope and influence of each of these forms
of collaboration might well be made the subject of special study,
but for present purposes it perhaps suffices to call attention to the
variety of these activities and the importance of their influence
as a whole.
Finally, the series of publications by the International Labour
Office in the field of statistics and statistical methods is itself an
important medium of international standardisation. These include reports on statistical methods in the various fields, as described in detail in the succeeding sections; other studies present
the statistical materials available on specific topics. Many articles
in the International Labour Review are devoted to statistical problems. The section of the Review devoted to statistics presents
current data on cost of living, employment and unemployment,
hours of work and wages. The Year Book of Labour Statistics,
which dates formally from 1935-361, presents in comparable form
the data from as many countries as possible on each of the main
fields of labour statistics. From a study of the statistics published
1
See below, pp. 41, 44, for a discussion of the predecessors of the Year Book
of Labour Statistic).

INTRODUCTION

in the Review since 1921, and in the Year Book of Labour Statistics
during the period since 1935, the progress made in the international
standardisation of the statistics of labour can be shown. As more
and more countries publish statistics on various topics the scope
of the tables in the Year Book is extended. The Year Book thus
serves both to measure progress, and to stimulate the development
of statistics in the various countries in the form adopted in this
annual volume.
This account of the work of the International Labour Office
would be incomplete and give perhaps a false impression if it were
to disregard the primary element in the whole picture, namely, the
desire on the part of the several countries to perfect their own
statistical measurement of the status, progress and problems of
labour. The International Labour Office has served as an agency
through which this desire could find expression and could be translated into effective action.
The demand for more and better statistics arising in different
countries from the broadening recognition of their importance has
been reflected in the adoption by various international meetings
of experts of resolutions calling for improvements in data on the
subjects with which they have been concerned. Thus the Conference of the American States Members of the International
Labour Organisation held at Santiago de Chile in 19361 urged the
development of statistics on the cost of living and family living
studies according to plans to be developed by the Office, and a
similar Conference held in Havana in 19402 called upon the Office
to take steps to secure uniformity in all labour statistics and urged
the Member States to ratify and give effect to the Convention concerning statistics of wages and hours of work. In a series of
resolutions on various topics the International Labour Conference
at several sessions has called for better and more uniform labour
statistics in the different countries. 8 But the force of the whole
movement is perhaps best exemplified in the series of International
1
See Labour Conference of the American States which are Members of the International Labour Organisation (Santiago de Chile, January 1936), Record of Proceedings: Resolution concerning the cost of living in American countries (No. 9),
p. 293. Other resolutions affecting statistics dealt with the question of uniformity
in methods of investigation into the conditions and costs of the nutrition of workers (Resolution No. 11, ibid., p. 295); the establishment or extension of central
bureaux of social statistics, particularly with regard to wages, cost of living,
unemployment, industrial accidents, occupational diseases, and labour disputes
(Resolution No. 23, ibid., p. 300), and with regard to statistics of the agricultural
population (Resolution No. 25, ibid., p. 301).
2
Second Labour Conference of the American States which are Members of the
International Labour Organisation, Hanava (Cuba), 21 Nov. to 2 Dec. 1939, Record
of Proceedings, p. 267, Resolution No. XX.
3
For the text of resolutions of the International Labour Conference on points
dealing with statistics, see The International Labour Code 1939, pp. 543-4, 555-6,

INTERNATIONAL STANDARDISATION OF LABOUR STATISTICS

Conferences of Labour Statisticians in which officials responsible
for labour statistics in the various countries have met for the common purpose of unifying the course of development and advancing
the international standardisation of labour statistics. The results
of the movement are shown in the expansion and steady improvement during the past twenty years of the labour statistics published
in the different countries.
In the following pages, each branch of labour statistics is
discussed in detail, and references are given to the various
statistical publications of the International Labour Office.

I
CLASSIFICATION OF INDUSTRIES AND OCCUPATIONS:
THE GAINFULLY OCCUPIED
This topic of classification of industries and occupations is in a
sense preliminary and of general character since it relates to the
way in which labour statistics in the various fields may be classified
or organised. The gainfully occupied or the employed, for example,
may be classified according to the industry in which they are employed or the occupation which they perform; wages may be arranged according to the industry or occupation of the wage earner;
strikes according to the industry to which the establishment affected
belongs. For purposes of international comparison, it is obviously
desirable that the materials available in the several countries be
classified or organised along similar lines.
In preparation for a discussion of the classification of industries
and occupations at the First International Conference of Labour
Statisticians in 1923, the Office published a report summarising the
classifications in use in a number of countries and presenting the
principles underlying the different uses of such a classification.1
The Conference in its resolutions emphasised the distinction
between classification by industry and by personal occupation and
pointed out that, for purposes of classifying the occupied population, the primary classification should be according to industries
with subdivisions by occupations, suggesting, however, that to
give full details by occupation a second separate classification, by
occupations, may also be necessary.
The Conference drew up a list of main divisions—primary
production (agriculture, mining, etc.), secondary production (manufacture and construction), and services (transport, commerce,
administration, etc.). 2
1
Systems of Classification of Industries and Occupations, Report prepared for
the International Conference of Labour Statisticians, Studies and Reports, Series
N, No. 1 (Geneva, 1923).
2
For the discussions at this Conference, see International Conference of Labour
Statisticians: Report, Studies and Reports, Series N, No. 4 (Geneva, 1924), pp. 8-30.

INTERNATIONAL STANDARDISATION OF LABOUR STATISTICS

At the Second and Third Conferences in 1925 and 1926 attention was given to the list of industries; the Second Conference recommended that the Office draft a provisional list for use in labour
statistics 1 , and the Third suggested that the details of the specific
content of each industry group as used in the classifications of the
different countries be compiled as a further step in the direction
of a uniform list.2 In response to the resolution of the Second Conference the Office prepared a provisional list of the principal branches
of economic activity, for the purpose of presenting in a uniform
order the data available in respect to wages, employment and
unemployment. The list (latest revision) is given in Part II (pp. 5152).
A further study of the subject has been undertaken by the
Committee of Statistical Experts of the League of Nations—on
which the International Labour Office is represented—to which
the whole question of preparing a scheme of classification of industries was referred in accordance with the provisions of the International Convention relating to Economic Statistics. 3 This Committee has considered the problem of industrial classification in
relation to the gainfully occupied population. On this subject its
final report in 1938 recommended a uniform definition of the "gainfully occupied" and two lists of industries—a minimum list and
a detailed list—for purposes of classifying the gainfully occupied
in a uniform manner for international comparison. The minimum
list is given in full in Part II (pp. 65-66), together with the pertinent resolutions of the Committee, and the recommended definition.
With regard to a similar list of occupations for purposes of international comparisons, the Committee decided against any attempt
(for the time being) to draft such a list since it considered that an
occupational classification was primarily of national rather than
of international statistical interest.
In this connection, the statistical work of the Office relating
to the gainfully occupied population may be discussed. A series
1
For the discussions at the Second Conference, see Report on the Proceedings
of the Second International Conference of Labour Statisticians, Studies and Reports,
Series N, No. 8 (Geneva, 1925), pp. 40-70. In this connection reference may be
made to R. H. COATS: " T h e Classification Problem in Statistics", in International
Labour Review. Vol. X I , No. 4, Apr. 1925, p p . 509-525.
2
For the discussions at the Third Conference, see Report on the Proceedings
of the Third International Conference of Labour Statisticians, Studies and Reports,
Series N, No. 12 (Geneva, 1926), pp. 86-91.
3
This Convention was adopted in 1928 by the International Conference relating
to Economic Statistics called under the auspices of the League of Nations.
Manley O. HUDSON (ed.): International Legislation: A Collection of Texts of
Multipartite International Instruments of General Interest beginning with the
Covenant of the League of Nations, Vol. IV, 1928-1929, pp. 2575-2589, esp. 25822583 (Washington, Carnegie Endowment for International Peace, 1931).

CLASSIFICATION OF INDUSTRIES AND OCCUPATIONS

of brief articles in the International Labour Review during the
period 1935 to 1940 reviewed the statistics of the occupational
censuses in the different countries as these became available, pointing out differences in definitions of gainfully occupied and presentting a series of tables, so far as possible, according to a uniform
plan. 1 Tables showing the gainfully occupied in the different countries of the world, based on a uniform definition including unpaid
family workers, so far as the data in the original sources permit,
are published in the Year Book of Labour Statistics beginning with
Volume II (1937). Tables are also given showing the gainfully
occupied classified by industry or occupation groups in accordance
with the classifications adopted in the sources. The 1941 Year
Book includes a table of the percentage gainfully occupied according to the age groups recommended by the Committee of Statistical Experts of the League of Nations. 2

1
See, for example, International Labour Review, Vol. X X X V I I I , No. 5, Nov.
1938, pp. 706-714, and references there cited.
* See also "A Comparison of the Gainfully Occupied Population by Sex and
Age in the Various Countries of the World", in International Labour Review, Vol.
X L I , No. 5, May 1940, pp. 541-550. Mention may be made of two special articles: J. W. N I X O N : "On the Statistics available concerning the occupied population of the world and its distribution", in Revue de l'Institut international de Statistique, 1938, livraison 3; and "Statistical Survey of Non-Manual Workers",
in International Labour Review, Vol. X X X I V , No. 2, Aug. 1936, pp. 244-274.
See also below, Part I I , p. 64.

II
EMPLOYMENT AND UNEMPLOYMENT
UNEMPLOYMENT

The Office: prepared a report on Methods of Statistics of Unemployment for the Second International Conference of Labour Statisticians in 1925.1 The resolutions of this Conference recognised
that the best basis for unemployment statistics was formed by the
comprehensive schemes of unemployment insurance, but where
these did not exist, statistics from workers' organisations and from
public employment offices would furnish valuable data. In countries
where satisfactory information from these sources was not obtainable, the Conference recommended that special enquiries should
be made from time to time, preferably at the time of the general
population census. 2

Mention should be made also of the Recommendation adopted
by the International Labour Conference in 1935 concerning
unemployment among young persons, which urged that unemployment statistics should show sex, age, and occupation, persons
who had never been in employment being classified separately and
according to the occupation for which they had been trained or
in which they had applied for employment. 3
The statistics of unemployment depend to a large extent upon
the systems in force for relieving unemployment, and on the definition of "unemployed". With the extension of the scope of unemployment insurance, the scope of adequate statistics of unemployment has continually increased.
1
Studies and Reports, Series N, No. 7 (Geneva, 1925). See also, Methods of
Compiling Statistics of Unemployment: Replies of the Governments, Studies and
Reports, Series C, No. 7 (Geneva, 1922). Mention should also be made of the
provision embodied in Article I of the Unemployment Convention, 1919, according to which each Member State ratifying the Convention undertakes to communicate to the Office all available statistics and other information concerning
unemployment. See The International Labour Code 1939, p. 543.
2
For the discussions at this Conference, see Report on the Proceedings of the
Second International Conference of Labour Statisticians, Studies and Reports, Series
N, No. 8 (Geneva, 1925),pp. 48-67.
3
For the text of this part of the Recommendation see The International Labour
Code 1939, p. 541.

EMPLOYMENT AND UNEMPLOYMENT

The Office publishes statistics of unemployment in the International Labour Review and in the Year Book of Labour Statistics.1
In general, two main problems occupy the principal attention: the
numbers unemployed—wholly and partially—and the percentages
unemployed based on the sum of the employed and the unemployed.
Special problems include unemployment trends by industry or
occupation, unemployment among women and young persons, and
seasonal movements in the percentages of unemployment.2
In publishing the periodical statistics of the numbers unemployed, the Office emphasises that they relate to recorded unemployment only, and are not a complete census of the unemployed ; that
therefore the figures are more useful in indicating whether unemployment is increasing or decreasing in any country than in showing the absolute level of unemployment at any given date.
In this connection the Office reviewed the results of censuses
of unemployment in various countries in articles in the International Labour Review in 1933 and 1935.3
In recent years the Office has devised an international index of
unemployment in which the fluctuations of some 15 important
countries (recently 13) are combined by a suitable system of weighting into an index number to show the general movements in the
level of unemployment.4
1
See also, Statistics of Unemployment in Various Countries 1910 to 19ZZ,
Studies and Reports, Series C, No. 6 (Geneva, 1922), and Unemployment 19201923, Series C, No. 8 (Geneva, 1924).
For a discussion of the different types of statistics, see the introductory notes
to chapter on "Employment and Unemployment" in the Year Book of Labour
Statistics. For detailed notes on the methods of the statistics, see I.L.O. YearBook, 1934-35, Vol. I I : "Labour Statistics", pp. 169-183, and corresponding
references in the Year Book of Labour Statistics.
2
See for example, "Movements of Unemployment and Employment in Different Industries and Occupations", in International Labour Review, Vol. X X V ,
No. 3, Mar. 1932, pp. 399-414; Henri F u s s : "Unemployment and Employment
among Women", ibid., Vol. X X X I , No. 4, Apr. 1935, pp. 463-497; and "An
International Index of Unemployment", ibid., Vol. X X X I I , No. 1, July 1935,
pp. 118-119.
s
"Some Recent Censuses or Estimates of Unemployment", in International
Labour Review, Vol. X X V I I I , No. 1, July 1933, pp. 46-61, and Vol. X X X I I ,
No. 6, Dec. 1935, pp. 826-850. Also Robert R. NATHAN: "Estimates of Unemployment in the United States, 1929-1935", ibid., Vol. X X X I I I , No. 1, J a n . 1936,
pp. 49-73.
i
For a full account of the methods adopted and the results obtained, see
John LINDBERG: "An Attempt to Construct International Measures of Unemployment", in International Labour Review, Vol. X X V I , No. 4, Oct. 1932, pp.
491-512; and idem, "Some Problems in the Construction of Index Numbers of
Unemployment", ibid., Vol. X X I X , No. 4, Apr. 1934, pp. 472-499; "National
and International Index Numbers of t h e General Level of Unemployment",
ibid., pp. 557-571; and "World Index Numbers of Unemployment", ibid., Vol.
X X X I X , No. 1, Jan. 1939, pp. 118-129, and "World Level of Unemployment",
Ibid., No. 6, June 1939, pp. 812-813.

INTERNATIONAL STANDARDISATION OP LABOUR STATISTICS
EMPLOYMENT 1

Comparatively little attention has been given by the Conferences to the subject of employment as such. The censuses of the
gainfully occupied including employers, persons working on own
account, as well as the employed and unemployed give the fundamental data on the employed as a part of the gainfully occupied
population ; the problems of the statistics of the gainfully occupied
have been discussed briefly in the preceding section, and are dealt
with in the excerpts from the report of the Committee of Statistical
Experts of the League of Nations, reproduced in Part II (pp. 5266). For the rest, the statistics of the employed have been treated
in the Conference resolutions almost as if they were implied in
the statistics of unemployment: the calculation of percentages unemployed implies corresponding percentages employed, and totals
of the two groups together. With specific reference to employment,
the Second International Conference (of 1925) included a short
resolution to the effect that "information as to the state of employment should also be published periodically, if possible monthly,
based on returns made by a representative number of employers".
Reference may be made also to resolutions dealing with actual
and normal hours of work. Apart from these the subject of statistics of employment is also referred to implicitly in resolutions
dealing with industrial disputes and industrial accidents, in describing the exposure data required. In order to calculate frequency
and severity rates of industrial accidents, for example, data on the
total man hours (or man days) of employment or average number
of persons employed in the corresponding industries and occupations, must be available. These resolutions, while they do not
describe in detail the methods by which the statistics are to be
obtained, give certain definite specifications as to the objectives
sought.
Statistics of employment are published by the Office in the
International Labour Review and in the Year Book of Labour Statistics. The statistics are derived from returns from sickness insurance
statistics of persons employed or covered by the insurance, from
unemployment insurance statistics and from employers (establishments).
Some of the special problems of employment data are indicated
in the introductory remarks to the section of the Year Book of
Labour Statistics dealing with statistics of employment.
The statistical series showing fluctuations are of two main
types: (1) those showing changes in the numbers of employed
1

See also "Gainfully Occupied", p. 9; "Hours of Work", pp. 20-21.

EMPLOYMENT AND UNEMPLOYMENT

persons and (2) those showing changes in the total number of
hours worked. Index numbers of employment are given for different industries for a limited number of countries in the I.L.O. YearBook prior to the appearance of the Year Book of Labour Statistics
and for a somewhat larger number of countries in the latter publication.
As in the case of unemployment, a world index of employment
has been constructed by the Office, to show the trends and changes
in the general level of employment. 1

1
See John LINDBERG: "Some Problems of International Employment Statistics", in International Labour Review, Vol. XXXV, No. 5, May 1937, pp. 608-642;
and "International Index Numbers of Employment", ibid., Vol. X X X V , No. 5,
May 1937, pp. 724-733; also "World Indexes of Employment", ibid., Vol. X L ,
No. 1, July 1939, pp. 125-134.

Ill
WAGES AND HOURS OF WORK
MONEY W A G E S

The subject of wages was one of the topics on the agenda of the
First International Conference of Labour Statisticians in October
1923, for which the Office submitted a report on Methods of Statistics of Wages and Hours of Labour.1 This Conference adopted resolutions covering the types of wages or earnings statistics to be collected, as well as on index numbers of wages as discussed below.
The two Conferences of Statisticians convoked by the Social
Science Research Council2 in 1929 and 1930, considered the question
of wage data in conjunction with the index numbers of wages and
the difficult problem of international comparisons, a subject which
is discussed further below. Following these, the Fourth International Conference of Labour Statisticians in 1931, in a series of
resolutions addressed to the Office, reviewing the problems of international comparisons of real wages, included as part of this broader
topic details as to the wages data to be collected. Of special importance was its recommendation that the question of the supply
of data on wages and the cost of living required for international
comparisons of real wages be considered with reference to the
framing of a convention that would bind governments ratifying
it to collect and supply information at regular intervals. Following up this suggestion, the I.L.O. Committee of Statistical Experts,
itself organised in accordance with a recommendation of this Fourth
Conference, considered the plan for a convention at its second
session in December 1935; the Fifth International Conference of
Labour Statisticians in September-October 1937 formulated a
1
Studies and Reports, Series N, No. 2 (Geneva, 1923). For the discussions
at this Conference, see International Conference of Labour Statisticians, Studies
and2 Reports, Series N, No. 4 (Geneva, 1924), pp. 30-47. _
See International Wage Comparisons. Documents arising out of Conferences,
held at the Intel-national Labour Office in Jan. 1929 and May 1930, convened by
the Social Science Research Council of New York. Manchester University Press
(1932). An account of the First Conference is also contained in "An International Conference on Wage Statistics", in International Labour Review, Vol.
XIX, No. 4, Apr. 1929, pp. 522-539.

WAGES AND HOURS OF WORK

draft Convention concerning statistics of wages and hours of
work for the consideration of the International Labour Conference ;
and the 24th Session of the International Labour Conference
meeting in June 1938 adopted the Draft Convention which is given
in full in Part II (pp. 85-94). The Convention set forth a minimum
programme of wages and hours statistics to be compiled by each
ratifying State, the results of which were to be forwarded to the
International Labour Office. By this means it was hoped to
promote the adoption of this minimum programme by the various
countries. 1
The texts given in Part II include also certain supplementary
resolutions which were adopted by the Fifth International Conference of Labour Statisticians, but which by reason of the limitation of the agenda could not be considered by the International
Labour Conference. They include also the recommendations,
based on resolutions adopted by the Fifth Conference, which were
referred by the International Labour Conference to the Governing
Body of the Office to be sent out by the Office to the different
Governments for further consideration, in accordance with the procedure prescribed in the Convention concerning statistics of wages
and hours of work.
Prior to the First Conference in 1923 the Office had commenced
the study of wages and wage changes in different countries. The
International Labour Review for January 1923 contained the first
of a series of such studies 2 , while in 1922 a first report on Wage
Changes in Various Countries, 1914-1921 was published. 3 In these
special reports, emphasis was laid upon the difference between
wage rates and earnings, and on real earnings as distinct from
1
For a discussion of this Convention, see J. W. N I X O N : "An International
Convention on Statistics of Wages and Hours of Work", in Journal of the Royal
Statistical Society, Vol. CI, Part IV, 1938, pp. 730-735. See also " T h e TwentyFourth Session of the International Labour Conference", in International Labour
Review, Vol. X X X V I I I , No. 3, Sept. 1938, pp. 350-355; and International Labour
Conference, Twenty-Fourth Session, Geneva, 1938, Record of Proceedings, pp. 34-35,
292-295, 646-670. By 1 August 1942, the Convention was in force for Denmark,
Sweden, South Africa, Australia, Norway, Netherlands, New Zealand, Switzerland, Egypt and Mexico. The Convention is divided into six Parts; Members
ratifying it may exclude from their acceptance (a) any one of the Parts II, III
or IV; or (b) Parts II and IV; or (c) Parts III and IV (Article 2). Exclusion is
effected by appending a declaration to ratification; such a declaration may be
cancelled at any time by a subsequent declaration. Pursuant to this provision,
Australia, New Zealand and South Africa have excluded Part II from their
acceptance; Denmark, Egypt, Norway, Sweden and Switzerland have excluded
Part I I I ; and Egypt, South Africa and Switzerland have excluded Part IV. Part
II covers earnings and actual hours of work and Part I I I time rates of wages and
normal hours of work in mining and manufacturing industries; Part IV deals
with wages and hours of work in agriculture.
2
International Labour Review, Vol. VII, No. 1, Jan. 1923, pp. 80-95.
3
Studies and Reports, Series D, No. 2 (Geneva, 1922); No. 10, Wage Changes
in Various Countries, 1914-1922 (Geneva, 1923); No. 16, Wage Changes in Various
Countries, 1914-1925 (Geneva, 1926).

INTERNATIONAL STANDARDISATION OF LABOUR STATISTICS

money earnings. In these early articles and reports, wages and
hours of labour were treated country by country. This arrangement was continued in articles in the International Labour Review.1
The number of countries supplying information gradually increased, and in 1929 an international table covering 12 countries
showing changes in the general level of wages within each country
was presented. 2 In 1931 international tables 8 were given showing
the movements of wages in the principal industries and occupations in different countries. Data on wages in a number of specific
occupations in the principal cities as of October of each year have
been obtained since 1924 by means of special enquiries addressed
to the statistical offices of the different countries and have been
published in the International Labour Review.4 Since the publication of the Year Book of Labour Statistics (and its predecessor
Vol. II: Labour Statistics of The I.L.O. Year-Book, for 1934-35),
these tables have been incorporated in it.
Special monographs relating to specific industries were published
for the textile industry and agriculture. 6 In addition special enquiries have been made into wages and hours of labour in the coalmining industry, by means of questionnaires, in 1925, 1927, 1929,
1931 and 1937. The questionnaires, drawn up with the aid of a
special committee, were sent to the various Governments, together
with details of the terms used. By this means it was sought to
obtain internationally comparable data. 8 Special reports on the
world textile industry and on the world coal-mining industry included data on wages and hours of work in different countries in
1

May 1928 to May 1929.
International Labour Review, Vol. XX, No. 1, July 1929, pp. 113-125; Vol.
XXI, No. 1, Jan. 1930, pp. 117-127.
8
International Labour Review, Vol. XXIV, Nos. 2-3, Aug.-Sept. 1931, pp.
294-302; Vol. XXVI, Nos. 2-5, Aug.-Nov. 1932, pp. 255-267, 431-437, 562-577,
and 716-738.
4
Up to 1930 the data were obtained at more frequent intervals. The report
on the enquiry for October 1941 was published in the International Labour Review,
Vol. XLVI, No, 5, Nov. 1942, pp. 638-653; each such report refers to the preceding
report.
* "Wages and Hours of Work in the Textile Industry (Cotton and Wool)", in
International Labour Review, Vol. XXX, No. 3, Sept. 1934, pp. 387-418; "Statistics of Wages of Agricultural Workers in Various Countries, 1927-1934", ibid.,
Vol. XXX, Nos. 5-6, Nov.-Dec. 1934, pp. 692-708, 844-865.
6
The results of the first enquiry were published in a volume Wages and Hours
oj Work in the Coal-Mining Industry, Studies and Reports, Series D, No. 18
(Geneva, 19281, and those for 1927, 1929 and 1931 in the International Labour
Review, Vol. XX, No. 4, Oct. 1929, pp. 539-562; and No. 6, Dec. 1929, pp. 840855, errata, p. 903, and Vol. XXI, No. 1, Jan. 1930, pp. 89-108; Vol. XXIII,
No. 5, May 1931, pp. 644-666, errata, p. 937; and Vol. XXVIII, No. 3, Sept. 1933,
pp. 371-394. The volume on the coal-mining industry contained the results of a
further enquiry; see The World Coal-Mining Industry, Studies and Reports,
Series B, No. 31, 2 Vols. (Geneva, 1938). See also "The European Lignite Industry", in International Labour Review, Vol. XXII, No. 6, Dec. 1930, pp. 752-772;
Vol. XXIII, No. 1, Jan. 1931, pp. 25-Ì7.
2

WAGES AND HOURS OP WORK

these industries. 1 The Maritime Statistical Handbook2 published in
1936 included details of wages and wage rates for the merchant
marine personnel in all the maritime countries of the world. The
Office has also published from time to time summaries of the results
of the general wage censuses referred to in paragraph 5 of the
resolutions of the First Conference.8
INDBX NUMBERS OP W A G E S

The resolutions of the First International Conference of Statisticians of October 1923 included a recommendation to the effect
that "index numbers should be computed to show the general
course of changes in nominal wage rates and in actual earnings". 4
The two conferences of statisticians called by the Social Science
Research Council adopted a series of resolutions dealing with
index numbers of wages. The first Conference (1929) divided the
subject into three main topics and considered index numbers of
wages as measuring fluctuations (1) in one of the elements determining the standard of living, (2) in the remuneration per hour
of work, and (3) in the labour cost per unit of production. In detail,
its recommendations were designed to throw light upon six different aspects of wage index numbers: (1) changes in the standard
of living of the working classes as a whole, (2) changes in the fluctuations in the standard of living as between different industries and
occupations in the same country, (3) as an aid to forecasting economic conditions, (4) to measure earnings per hour in respect of
work of comparable character and efficiency, (5) for measuring
labour cost per unit of production, and (6) to measure variations
in the proportion of the national income formed by wages.
The second Conference called by the Social Science Research
Council (1930) emphasised certain specific points important in
the collection and organisation of the data, in particular stressing
the differences between rates and earnings, the importance of the
industries covered, the problem of including or excluding women's
rates in a general index, and the influence on the index of changes
in the relative proportions of the industries, etc., which are included
in the index number.
1
The World Textile Industry: Economic and Social Problems, Studies and
Reports, 2 Vols., Series B, No. 27 (Geneva, 1937); The World Coal-Mining Industry, 2 Vols., Series B, No. 31 (Geneva, 1938).
2
Studies and Reports, Series N, No. 21 (Geneva, 1936).
8
For a general discussion of the problems of wage statistics, see Umberto
R I C C I : " T h e Compilation of Wage Statistics", in International Labour Review,
Vol. X I I I , No. 4, Apr. 1926, pp. 489-505.
4
For a discussion of the resolution on wage index numbers and a synoptic
analysis of the methods of compilation adopted in different countries, see J. W.
N I X O N : "Index Numbers of Wages: A Survey", in Revue de l'Institut
internationale de Statistique, Vol. I, No. 1, Jan. 1933, pp. 39-53.

INTERNATIONAL STANDARDISATION OP LABOUR STATISTICS

The Convention concerning statistics of wages and hours of
work includes articles prescribing index numbers of the general
movement of earnings per hour, and where possible per day, week
or other customary period, and of the general movement of rates
of wages per hour or per week to be compiled by the countries
ratifying the Convention (articles 12, 21).
With special reference to index numbers of real wages, that is,
indices combining indices of nominal wages with those of the cost
of living of wage earners' families, the First International Conference of Labour Statisticians in October 1923 included a general
recommendation to the effect that the changes in actual earnings
should be related to changes in the cost of living. The whole subject
of measuring; changes in the cost of living is treated in section IV,
(pp. 22-23) while that of international comparisons of real wages
and of the cost of living is discussed in section V (pp. 24-30) below.
In particular, the resolutions of the Fourth Conference of Labour
Statisticians in 1931, though directed specifically to the problems
of international comparisons of real wages, contain many recommendations applicable to the compilation of national indices of
real wages.
Index numbers of nominal and real wages are published at
quarterly intervals in the International Labour Review and in the
Year Book of Labour Statistics. At the date of their first publication only 12 countries were included; in the Year Book for 1941
they extended over 34 countries; in many cases with separate
series for different industries, for both sexes, for males and females,
and showing wage rates or earnings per hour, or earnings per day,
week or month.
T O T A L W A G E S AND SALARIES

The first Conference of Statisticians called by the Social Science
Research Council in 1929 adopted resolutions dealing with the
measurement of changes in the proportion of the national income
formed by wages, or by the remuneration of labour as a whole.
A special study of the statistics of total wages and salaries was
published in the International Labour Review for September 1936.1
Annual figures on the subject have been given in the Year Book of
Labour Statistics since 1937.
HOURS OF W O R K 2

Since the subject of hours of work has been dealt with at the
International Conferences of Labour Statisticians in close connec1
"Statistics of Total Wages and Salaries", in International Labour Review,
Vol.2 XXXIV, No. 3, Sept. 1936, pp. 391-410.
See also Employment, pp. 14—15.

WAGES AND HOURS OF WORK

tion with that of wages—on account of the relation between earnings and hours worked—the topic is included here under the
heading Wages and Hours of Work.
The First Conference adopted resolutions concerning the collection of data on normal and actual hours of work, giving the principles following which the statistics should be obtained. 1 Since in
practice normal hours correspond to full time wages, and actual
earnings to actual hours, the close relationship between hours and
wages is clear as well as the need for the collection of hours data
in connection with wages.
The Convention concerning statistics of wages and hours of
work followed the same plan with more detailed specifications of
the exact scope of the data to be collected.
Statistics of hours of work were published by the Office in a
series of reports for different countries 2 , and subsequently in the
statistics section of the International Labour Review and in the
Year Book of Labour Statistics. These figures give, in general,
average actual or full time hours of work per week (or per day, or
per month) as well as index numbers showing the trends; in some
countries distributions of hours of work are given; in many countries details are given by industries or occupations. 3
A special report on Hours of Work in Postal Services was published in 1935.4

1
1

See references cited in note 1, p. 16.
Hours of Labour in Industry: Germany, Studies and Reports, Series D, No. 4,
(Geneva, 1922); No. 5, Hours of Labour in Industry: Belgium (Geneva, 1922)
No. 6, Hours of Labour in Industry: France (Geneva, 1922); No. 7, Hours of
Labour in Industry: Great Britain (Geneva, 1922); No. 8, Hours of Labour in
Industry: Italy (Geneva, 1923); No. 9, Hours of Labour in Industry: Switzerland
(Geneva, 1923); No. 11, Hours of Labour in Industry: Netherlands (Geneva, 1923);
No. 12, Hours of Labour in Industry: Czechoslovak Republic (Geneva, 1924);
No. 14, Hours of Labour in Industry: United States (Geneva, 1925). See also references cited in notes 4-6, p. 18.
3
Normal hours of work were obtained for selected occupations in the October
enquiries beginning with that of 1932; see "Wages, Hours of Work and Other
Factors in the Remuneration of Workers in Certain Towns in October 1932: I",
in International Labour Review, Vol. XXVII, No. 6, June 1933, pp. 815-838.
Statistics of actual hours of work were published for the first time in the Review,
Vol. XXXI, No. 2, Feb. 1935, pp. 265-289.
The Year Book of Labour Statistics, 1941, includes 4 tables dealing with hours
of work; general level of hours of actual work; general level of normal hours of
work; hours of work by industries, and normal hours of work in 30 occupations
in October
1940.
4
Studies and Reports, Series D, No. 21 (Geneva, 1935).

IV
COST OF LIVING 1
At the First International Conference of Labour Statisticians the
subject of cost of living was approached in general terms m a resolution recommending that index numbers of the purchasing power
of wages should be calculated "by relating changes in actual earnings to changes in the cost of living", the two series being comparable.
The subject of Cost of Living Index Numbers was placed on
the agenda a.t the Second International Conference, and the Office
prepared a report on Methods of Compiling Cost-of-Living Index
Numbers.2 The Conference adopted a series of resolutions covering
the methods of construction, basis and other points for cost-of-living indices. 3 In particular, recommendations were made as to the
use of family living studies as a basis for weighting the indices for
the different elements in the cost of living.
Most of the other references to cost of living in the various
resolutions given in Part II (see index) are concerned specifically
with international comparisons of cost of living, which are discussed below. Reference should be made, however, to these on
account of the bearing they have on cost-of-living statistics in
each country.
The Office commenced the regular publication of cost-of-living
figures in the International Labour Review in its first number, in
January 1921. These utilised the series of cost-of-living index
numbers as developed in each country. Special articles on problems
of cost-of-living indices and notes on methods employed in different
countries have appeared in the International Labour Review from
time to time. 4
1
See also V. International Comparisons of Real Wages and Cost of Living,
p. 24; VI. Family Living Studies, p. 31.
2
Studies and Reports, Series N, No. 6 (Geneva, 1925).
3
For the discussions at the Conference, see The Second International Conference of Labour Statisticians, Studies and Reports, Series N, No. 8 (Geneva, 1925),
pp. 7-20, 35-37.
i
See Ugo G I U S T I : "Methods of Recording Retail Prices and Measuring the
Cost of Living in Italy", in International Labour Review, Vol. IV, No. 2, Nov.
1921, pp. 257-274; Felix K L E Z L : "Methods of Calculating Index Numbers",
Md., Vol. X, No. 2, Aug. 1924, pp. 236-262; and E. C. SNOW: " T h e Practical
Interpretation of Index Numbers", ibid., Vol. X I , No. 2, Feb. 1925, pp. 180-199.

COST OP LIVING

During the period prior to the great depression of the early 1930's,
the base used by most countries for cost-of-living indices was
1914, to show prices in relation to a pre-war level, and to facilitate
international comparison conversions were made by the Office to
this base where the original series used some other period.1
Beginning with the volume of labour statistics issued as Part
II of the I.L.O. Year Book in 1934-1935, which immediately preceded the first Year Book of Labour Statistics, the cost-of-living
figures as well as other economic series have been converted to a
uniform base of 1929 as the year of reference.2
The Year Book of Labour Statistics included data on the cost
of living in forty-three countries in 1935-1936, and in fifty-three
countries in the 1941 edition. In most years, figures on the trends
in costs of important sub-groups, food, fuel and light, clothing and
rents have also been presented.

1
The question of the year to be chosen as the base was considered by the
Second International Conference of Labour Statisticians which recommended
that the Governments, in collaboration with the I.L.O., agree to adopt the same
year as base for the calculation of a new series of index numbers, and suggested
the year 1930 as suitable for these purposes. Subsequently the first Conference
called by the Social Science Research Council in 1929 considered the same problem from the point of view of wage statistics; they emphasised the importance
of uniformity, but did not specify any particular year. The problem was considered also by the International Conference relating to Economic Statistics called by
the League of Nations, which referred the question to the Committee of Statistical
Experts of the League: this Committee at its second session in December 1933
held that it was, at that time, premature to establish a uniform base subsequent
to the war of 1914-1918 for purposes of economic statistics.
2
The Statistical Year Book of the League of Nations has also adopted the year
1929 as the year of reference for indices.

V
INTERNATIONAL COMPARISONS OF REAL WAGES
AND COST OF LIVING
In this section only those aspects of the topics of real wages
and cost of living will be discussed which relate to international
comparisons. 1 Obviously, however, the basic data already discussed under wages and cost of living are also relevant to comparisons between countries.
In the First and Second International Conferences of Labour
Statisticians {1923, 1925) the problem of international comparisons was raised in connection with the work of the Office in carrying further the international comparisons of real wages in selected
cities commerced by the British Ministry of Labour. In particular,
the Second Conference adopted a resolution embodying certain
suggestions for the work of the Office.2 The whole problem was
considered further by the first and second Conferences of Statisticians (1929, 1930) called by the Social Science Research Council.
The Fourth International Conference of Labour Statisticians (1931)
reviewed the whole subject and adopted a series of resolutions
addressed to the Office on various specific details of methods to
be followed in comparing, internationally, money wages and real
wages. In addition, as noted elsewhere, the Conference recommended the establishment of a Committee of Statistical Experts,
to advise the Office in its work on wages and costs of living. All
these resolutions, including those of this Committee dealing with
wages, food, and rent comparisons, are given in Part II (pp. 101115).
This outline of the development of the resolutions, however,
needs to be supplemented by a review of the development of the
statistics.
The enquiry of the British Ministry of Labour, which was taken
as the starting point, consisted in ascertaining the wages of adult
1
See Felix KLEZL: "International Comparisons of Real Wages", in International
Labour .Review, Vol. XII, No. 4, Oct. 1925, pp. 467-483.
2
For the discussions at the Second Conference, see The Second International
Conference of Labour Statisticians, Studies and Reports, Series N, No. 8 (Geneva,
1925), pp. 21-34, 37-39.

INTERNATIONAL COMPARISONS

male workers in a series of about thirty occupations in a large number of towns, together with the retail prices of a number of articles
of common consumption, in order to permit calculations of the
purchasing power of wages. In 1924 the Office undertook its first
enquiry of this type: the results, published in October 19241,
attempted to express the relative purchasing power of average
wages or earnings for a standard (48-hour) week in the different
cities with reference to London as a base.
The elements in this method included the wage data, on the one
hand, and the cost-of-living data—a list of important food items,
with an adjustment for rents—on the other, in conjunction with a
method of expressing the final results in terms of an index, with
London as base.
It was soon found, however, that the wage data were not homogeneous or strictly comparable in all respects: some of the data
were for earnings, some were for wage rates; in some cases they
referred to minimum, in others to predominant, rates; and moreover,
the money wages themselves did not always fully express the total
remuneration of the workers—the amount of which is affected by
payments for holidays, family allowances, allowances in kind, and
the value of the social insurance contributions.
In carrying out the successive studies, a series of improvements
was introduced in the techniques and methods—as discussed in
the reports of these investigations.
Tables of these international indices of wage comparisons were
published during the period from 1924 to 19302 in conjunction with
data on wages, retail prices of important food items, and percentage
spent for rent as shown in family budget studies. These indices
were presented as merely rough indications of relative wages, and
their defects were of course apparent both to the Office and to those
who might attempt to use them. The further development of the
subject and the various steps taken to advance the materials and
methods available will be discussed under the headings Wages,
Elements of Remuneration Other than Wages, Cost of Living,
Food and Rents.
WAGES

The first plan adopted to secure materials usable for purposes
of comparison involved obtaining wage data for a list of selected
1
"Comparison of the Levels of Real Wages in Certain Capital Cities", in
International Labour Review, Vol. X, No. 4, Oct. 1924, pp. 630-652.
* See in particular, in addition to article cited in preceding note: "Wage
Rates and Retail Prices in Various Cities", in International Labour Review, Vol.
X I , No. 1, J a n . 1925, pp. 103-110; Vol. X I , No. 4, Apr. 1925, pp. 561-572; Vol.
X I I , No. 1, July 1925, pp. 96-106; and "Comparison of Real Wages in Various
Countries", ibid., Vol. X X I , No. 4, Apr. 1930, pp. 557-566.

INTERNATIONAL STANDARDISATION OP LABOUR STATISTICS

occupations in specific industries at a given time in the chief cities
of the principal countries. This was commented upon in the Resolutions of the Second Conference (1925); the first Conference of
Statisticians called by the Social Science Research Council proposed
certain additions to the list. The Fourth Conference (1931) also
urged the extension of these materials.
For the rest, the Fourth Conference and the Convention concerning statistics of wages and hours of work urged the development of the basic statistics of time rates of pay and actual earnings
as fundamental to international comparisons of real wages.
The materials available and published by the Office have already
been described in the section on Wages. They include the wages
for 30 occupations in October of each year for the principal cities
in the different countries, as published in the International Labour
Review (since 1924) and in the Year Book of Labour Statistics, and
average wage rates and earnings, by industries and occupations
in the several countries, as published in the Year Book of Labour
Statistics. Special mention should be made of the investigations
of the coal mining and textile industries, with their information
on wages in these industries in different countries. 1
ELEMENTS OF RÉMUNÉRATION OTHER THAN W A G E S

This question was touched upon in one of the resolutions of the
Fourth Conference relating to wage comparisons between countries
operating under different systems of social insurance, family allowances, paid holidays and other points, and was referred by it to
the Committee of Statistical Experts. The Committee at its first
meeting (Geneva, 12-15 December 1933) adopted a resolution on
the subject as given in Part II, pp. 107-108.
Statistics dealing with the subject are meagre. T h e principal
materials available in Office publications are a special report dealing
with wages, hours of work and other factors in the remuneration
of workers in certain towns in October 19322, and the reports
1
5

See references cited above, notes 4-6, p. 18 , and note 1, p. 19.
"Wages, Hours of Work, and Other Factors in the Remuneration of Workers
in Certain Towns in October 1932: I" (Money Wages, Hours of Work), in International Labour Review, Vol. XXVII, No. 6, June 1933, pp. 815-838; II (Family
Allowances, Holidays with Pay, Employers' and Workers' Contributions to
Compulsory Social Insurance Schemes), ibid., Vol. XXVIII, No. 1, July 1933,
pp. 104-121; III (The Interpretation and Combination of the Data, Conversion
of the Data for the Purpose of Comparisons between different Towns and Countries), ibid., Vol. XXVIII, No. 2, Aug. 1933, pp. 265-284.
See also "Holidays with Pay for Private Employees", ibid., Vol. XXIV,
No. 6, Dec. 1931, pp. 713-741, and references therein cited.
For special problems of non-monetary remuneration of agricultural workers
see "Statistics of Wages of Agricultural Workers in Various Countries, 19271934", ibid., Vol. XXX, Nos. 5-6, Nov.-Dec. 1934, pp. 692-708, 844-865. Also,
for example, "Allowances in Kind given to Farm Labourers in the United States",
ibid., Vol. XX, No. 4, Oct. 1929, pp. 562-567.

INTBRNATIONAI, COMPARISONS

dealing with wages of workers in coal mines in the series of studies
already referred to.1
COST OP LIVING

The resolutions of the second Conference of Statisticians called
by the Social Science Research Council in 1930 dealt with the
question of international comparisons of costs of living; they recommended that new family living studies be made whenever necessary
to furnish recent data on family expenditure, and made certain
proposals as to the methods to be followed in making comparisons
of living costs.
The Fourth International Conference in 1931 also emphasised
the need for new family living studies. In regard to comparisons
of cost of living, these resolutions included a recommendation that
the Office ask the statistical authorities in each country to furnish
data on rents of working-class families in the towns for which data
on wages and prices were obtained. Certain resolutions were directed also to the specific methods of comparison. In addition, as
already noted, the Conference recommended that a Committee of
Statistical Experts be established to assist the Office in the work
of developing wage and cost-of-living statistics.
It may be noted that in the resolution of the Fourth Conference,
recommendations were phrased in terms of food, fuel and light,
and rents; for purposes of ascertaining weights, family living studies
of working-class families were recommended. These resolutions
thus recognised the practical difficulties in the way of any complete
comparison of costs of living on an international plane.
The recommendations of the Committee of Statistical Experts
have considered the problems, thus far, primarily in terms of the
two most important elements, food and rent, as discussed below
in the succeeding sections.
The figures of retail prices of a list of commodities in different
towns in various countries, usually as of October of each year, have
been obtained and published since 1924.2
During the period 1924 to 1930 when indices of comparative
real wages based on selected occupations in certain cities were
published by the Office, an attempt to take differences in costs of
living into account took two forms, the first based on the cost of
1
See citations in note 6, p. 18. See also "Annual Holidays with Pay for
Miners Employed in Coal Mines", in International Labour Review, Vol. XXI,
No.2 2, Feb. 1930, pp. 219-243.
See, in addition to references cited in note 1, p. 25, "Wage Rates and
Retail Prices in Various Cities", in International Labour Review, Vol. XI,
No. 1, Jan. 1925, pp. 103-110; No. 4, Apr. 1925, pp. 561-572; Vol. XII,
No. 1, July 1925, pp. 96-106, and similar articles in subsequent volumes of the
Review.

INTERNATIONAL STANDARDISATION OP LABOUR STATISTICS

baskets of a list of food items to represent food consumption of
wage earners' families in the different regions, and the second
based on this cost as adjusted to take account of differences in the
proportion of the family budget spent for rent.
The whole problem of comparative costs of living was raised
in the enquiry carried out by the Office to compare the cost of
living of a certain category of workers in Detroit (U.S.A.) and those
in fourteen European towns in January 1931.1 The methods adopted and the problems encountered in the course of the enquiry were
discussed in the report and in an article in the International Labour
Review entitled "An International Enquiry into Living Costs". 2
In its subsequent work on this subject, the Office has devoted
its attention to the problems of comparing food costs and rents as a
preliminary stage to the solution of the more complex problem of
the costs of living as a whole. 3
Food
In the first work undertaken by the Office in the international
comparison of real wages, as already noted, the source data available for comparing costs of living were limited to the costs of a list
of important foodstuffs and the final results were presented in two
forms, one based on relative food costs only and the other including
an adjustment for rent. 4 The method followed in these food cost
comparisons included calculations based on regional food budgets
averaged so as to give a single figure with London as a base.
In October 1929, the method of making comparisons of food
costs was modified by the use of an international basket. The
international basket was established by taking the average consumption in workers' families per consumption unit in the different
countries included in the comparisons; the basket of foods was
priced in each city and the ratios of these costs in the case of any
two cities gave the desired ratio of food costs. In the original list
of foodstuffs, some 25 articles were priced 5 , but in 1933 the list was
1
A Contribution to the Study of International Comparisons of Costs of Living,
Studies and Reports, Series N , No. 17 (Geneva, 1932). The enquiry is often
referred to as the Ford-Filene enquiry since it originated in a request for information made to the International Labour Office by the Ford Motor Company Ltd.
(London) and was financed by t h e Twentieth Century Fund (President—Mr.

Edward A. F I I . E N B ) .
2
By H. STAEHLB, in International Labour Review, Vol. X X V I , No. 3, Sept.
1932, pp. 313-363.
3
For an examination of certain methodological problems, see Robert M.
WOODBURY: "Quantity Adjustment Factors in Cost-of-Living Ratios", in Econometrica, Vol. 8, No. 4, Oct. 1940, pp. 322-332.
4
"Comparison of the Levels of Real Wages in Certain Capital Cities", in
International Labour Review, Vol. X, No. 4, Oct. 1924, pp. 630-652, esp. p. 652.
6
"Comparison of Real Wages in Various Countries", in
International
Labour Review, Vol. X X , No. 4, Oct. 1929, pp. 580-588.

INTERNATIONAL COMPARISONS

reduced to 14 by omitting the least important items. 1 Data on
the cost of this food basket in a list of cities were published each
year.
A study embodying the results of an enquiry undertaken to
ascertain the cost of living in various European cities on the same
standard as that of employees in the Ford factories in Detroit
adopted the method of the geometrical average of the Laspeyres
and Paasche indices, the so-called "ideal" formula of Fisher, as
the best measure of the relative cost of living (and hence also of the
cost of food) between two cities.2 Apart from the resolutions of the
Fourth Conference, dealing with the general problem of international comparison of real wages, which contain some provisions relating
specifically to data on food consumption and prices, and those of
the Third Conference relating to family budget studies, the problems
of food cost comparisons are dealt with in resolutions of the Committee of Statistical Experts to which the whole question was
referred by the Fourth Conference. A special report dealing with
the theoretical problems of international comparisons of food costs
was prepared for the first meeting of this Committee in 1933.3
This report suggested the establishment of regions of good comparability; the relative costs of food between cities within each
region would be based on a comparison of the costs of the regional
food basket in these cities.
Finally, the whole problem was reviewed in a report on International Comparisons of Food Costs, originally prepared for a
meeting of the Committee of Statistical Experts called for the
spring of 1940; when this meeting had to be abandoned the comments and suggestions received by mail from the members of the
Committee were noted and the report was published in the Statistics Series. This report proposed the group basket method, in
which the Fisher formula is used as a means of comparing each pair
of countries, substituting, however, in the calculations the group
basket to which the country is allocated on the basis of similarity
of proportionate consumption for the local basket of each country. 4
International indices of comparative food costs on this basis are
1
"An International Comparison of the Retail Prices of Certain Important
Foodstuffs, July 1929 to Oct. 1932", in International Labour Review, Vol. XXVII,
No. 4, Apr. 1933, pp. 530-538.
2
See note 1, p. 28.
3
International Comparisons of Cost of Living. A Study of Certain Problems
connected with the Making of Index Numbers of Food Costs and of Rents, Part I,
"International Comparisons of Food Costs", by H. STAEHLE, Studies and Reports,
Series N, No. 20 (Geneva, 1934).
4
International Comparisons of Food Costs, by Robert Morse WOODBURY,
Studies and Reports, Series N, No. 24 (Montreal, 1941). See also Robert Morse
WOODBURY: "International Comparisons of Food Costs", in International Labour
Review, Vol. XLIII, No. 2, Feb. 1941, pp. 153-173.

INTERNATIONA!, STANDARDISATION OP LABOUR STATISTICS

included in trie Year Book of Labour Statistics, 1941, for 25 countries
for prices and exchange rates of October 1938. Figures for a
smaller number of countries for October 1940 are given in the
International Labour Review for April 1942.1
Rents2
The Fourth International Conference of Labour Statisticians
in 1931 adopted a resolution containing specific recommendations
on the subject of rents for purposes of international comparisons
of costs of living. The International Labour Office Committee of
Statistical Experts, established in accordance with the suggestion
of the Fourth Conference, considered tlie problem of rents at each
of its three sessions and adopted a series of resolutions advising the
Office in its research activities in this field. In particular, a series of
standard housing units was defined in terms of the availability of
certain items of equipment—bathroom, toilet and running water—
and these in conjunction with the number of rooms in the dwelling
furnish a frame of reference to make possible the collection of statistics of rents related to size and quality of dwellings along uniform
lines in different countries. The theoretical problems of rent comparisons are discussed in a special report prepared for the Committee and in a special article in the International Labour Review}
The Office carried out a survey of statistics of rents in certain
towns in 1932, the results of which were published in the Review.
Subsequently, at the suggestion of the Committee of Statistical
Experts, rent statistics for the standard housing units, as established
by the Committee, were collected, the survey for 1936 covering a
number of towns in some 20 countries. The results were published
in the Review and in The Year Book of Labour Statistics, 1939.4
1
s

Vol. XLV, No. 4, Apr. 1942, pp. 472-474.
For the statistics of rents as part of the cost of living see above, pp. 22-23,
IV. Cost of Living; as part of expenditures, see below, pp. 31-32, VI. Family
Living Studies; as an element in housing statistics, see below pp. 35-36, VIII.
Housing. The Committee of Experts on Housing Statistics at its meeting in
Munich, 1928, adopted a definition of rent (see under VIII. Housing).
* International Comparisons of Cost of Living. A Study of Certain Problems
Connected with th-e Making of Index Numbers of Food Costs and of Rents, Part II,
"International Comparison of Rents", by R. GUYB, Studies and Reports, Series
N, No. 20 (Geneva, 1934), pp. 107-146; and R. GUYE: "A Further Contribution
to the International Comparison of Rents", in International Labour Review,
Vol. XXXIV, No. 5, Nov. 1936, pp. 569-593.
4
"International Statistics of Rents in Certain Towns", in International Labour
Review, Vol. XXVIII, No. 2, Aug. 1933, pp. 248-264; "Retail Prices and Rents
in Certain Towns and Countries in October 1933", ibid., Vol. XXIX, No. 6,
June 1934, pp. fi59-865; "Retail Prices and Rents in Certain Towns in 1934",
ibid., Vol. XXXI, No. 3, Mar. 1935, pp. 428-136; "Workers' Dwellings in Certain
Towns: An Enquiry into the Quality, Rents, and Conditions of Occupation of
Workers' Dwellings", ibid., Vol. XXXIV, No. 5, Nov. 1936, pp. 686-702; "Rents
of Workers' Dwellings in Various Towns, 1936-37", ibid., Vol. XXXVIII, No. 6,
Dec. 1938, pp. 840-861.

VI
FAMILY LIVING STUDIES
The subject of family budget enquiries was approached indirectly at the Second International Conference in 1925 in connection
with the problem of cost-of-living index numbers. The Conference
adopted a resolution to the effect that the weights should be based,
so far as possible, on working-class consumption in the districts or
among sections of workers to which the index numbers related,
and pointed out that the best method of obtaining such data was
through family living enquiries. For the Third Conference in 1926
the Office prepared a report on Methods of Conducting Family
Budget Enquiries1 and the Conference adopted a series of resolutions on the subject. 2
At the Fourth Conference, these resolutions of the Third Conference were referred to, with the recommendation that where enquiries had not been made during the preceding ten years new
enquiries should be undertaken on lines laid down in the resolutions.
In following up the work of the Third Conference, the Office
adopted a programme of analysing and publishing in the International Labour Review the results of the various important family
budget enquiries soon after their appearance; these analyses called
attention to the chief characteristics of these studies, and by presenting these results according to a uniform pattern suggested the
importance of covering at least the basic points which were dealt
with in these summaries. The analyses were themselves summarised
in November 1933 in a general review of the chief enquiries conducted in the previous ten years. 8 A special analysis of food consumption per consumption unit in wage earners' families according
to the results of family living studies appeared in the Review in
December 1933.4 A second general summary was published in
1
2

Studies and Reports, Series N, No. 9 (Geneva, 1926).
For the discussions of the Third Conference see Studies and Reports, Series N,
No. 12 (Geneva, 1926), pp. 12-37, 92-98.
* "Recent Family Budget Enquiries", in International Labour Review, Vol.
XXVIII,
No. 5, Nov. 1933, pp. 635-672.
4
"Food Consumption of Working-Class Families in Certain Countries", in
International Labour Review, Vol. XXVIII, No. 6, Dec. 1933, pp. 870-876.

INTERNATIONAL STANDARDISATION OP LABOUR STATISTICS

1939 covering not only the general expenditure but also food expenditure and consumption habits 1 ; this was followed in June 1940
by a special summary of enquiries relating to families of the unemployed2, and in April 1941 by a summary of enquiries relating to
families of agricultural workers. 8
In 1940 a revised report on Methods of Family Living Studies:
Income—Expenditure—Consumption
brought the material on
methods up to date; it included a special chapter devoted to food
and dietary surveys. 4 A special report, prepared for the 1942 PanAmerican Sanitary Conference at Rio de Janeiro 5 , included a survey
of dietary studies in American countries together with a chapter on
Methods of Dietary Surveys. 6
Comparative data on certain aspects of family living—sources
of family income, percentage expenditures on chief groups, food
consumption per consumption unit—are published in the Year
Book of Labour Statistics.

1
"An International Survey of Recent Family Living Studies": I (Income and
Expenditure), and II (Food Expenditure and Consumption Habits), in International Labour Review, Vol. X X X I X , Nos. 5-6, May-June 1939, pp. 662-705,
814-846.
2
"An International Survey of Recent Family Living Studies": III (Families
of Unemployed Workers), in International Labour Review, Vol. X L I , No. 6, June
1940, pp. 645-662.
» "An International Survey of Recent Family Living Studies": IV (Families
of Farmers and Agricultural Workers), in International Labour Review, Vol.
X L I I I , No. 4, Apr. 1941, pp. 474-491.
4
Studies and Reports, Series N, No. 23 (Geneva, 1940).
s
Robert Morse WOODBURY: Food Consumption and Dietary Surveys in the
Americas, Results—Methods, Report presented by the International Labour
Office to the Eleventh Pan-American Sanitary Conference held at Rio de Janeiro,
7 to 18 September 1942 (Montreal, 1942).
6
In addition to the references already cited, see J. W. N I X O N (Rapporteur):
"Preliminary Report of the Committee on Family Budgets", in Revue de l'Institut internationale de Statistique, 4th year, 1936, No. 1, pp. 4 5 - 6 1 ; and Workers'
Nutrition and Social Policy, Studies and Reports, Series B, No. 23 (Geneva, 1936).

VII
INDUSTRIAL ACCIDENTS
A report on Methods of Statistics of Industrial Accidents was
prepared for the First International Conference of Labour Statisticians in October 1923.'
This Conference adopted certain general resolutions on the
question of international standardisation of statistics of industrial
accidents, e.g. concerning the classification of accidents by industry,
cause, extent and degree of disability, location of injury and nature
of injury. 2
Recommendations were made also as to the form of calculation
of frequency and severity rates.
Subsequent conferences have not dealt further with the problems of industrial accident statistics. The activities of the Office,
however, include special articles in the International Labour Review*,
two reports on methods of compiling statistics of accidents in coal
mining and on railways4, and analyses of industrial accidents in
different countries,
A first analysis of industrial accident rates in 30 countries was
published in an article in the Review for December 19316; this
included details of the definitions and methods utilised in the different countries. A Report on Industrial Accident Statistics*, issued
in 1938, supplemented and expanded these figures, and brought
them up-to-date. Comparative tables were presented on accident
1
Studies and Reports, Series N, No. 3 (Geneva, 1923).
* For discussions at the Conference, see International Conference of Labour
Statisticians: Report, Studies and Reports, Series N, No. 4 (Geneva, 1924), pp.
49-63.
* J. W. NIXON: "Some Problems of Statistics of Accidents as Illustrated by
the British Statistics", in International Labour Review, Vol. XVIII, No. 6, Dec.
1928, pp. 731-739; idem, "The Measurement of Risk in Connection with Labour
Statistics", Vol. XVII, No. 5, May 1928, pp. 633-650; ibid., "Industrial Accidents", Vol. XXIV, No. 6, Dec. 1931, pp. 751-787.
* Methods of Compiling Statisticsof Coal-Mining Accidents, Studies and Reports,
Series N, No. 14 (Geneva, 1929), and Methods of Compiling Statistics of Railway
Accidents, Series N, No. 15 (Geneva, 1929).
'International Labour Review, Vol. XXIV, No. 6, Dec. 1931, pp. 751-787.
A preliminary survey covering 21 countries was published in the January 1931
issue,
Vol. XXIII, No. 1, pp. 87-110.
4
Studies and Reports, Series N, No. 22 (Geneva, 1938).

INTERNATIONAL STANDARDISATION OP LABOUR STATISTICS

rates in coal mining, metal mining, all mining, quarrying, industry,
agriculture and railways.1
Beginning; with 1940, comparative tables of fatal accident rates
in coal mining and all mining (including non-fatal rates in some
countries), industry and railways have been published in the Year
Book of Labour Statistics.

1
Reference may also be made to a statement adopted by the Committee of
Experts on Safety in Coal Mines (Geneva, 21-24 Nov. 1938) published in The
International Labour Code 1939, p. 737.

Vili
HOUSING»
Although the statistics of housing fall somewhat outside the
scope of labour statistics as this term is generally understood,
housing conditions are an important factor in the standard of living
and rents an important element in living costSi The latter topic
has already been discussed in connection with international comparisons of costs of living. Housing statistics are basic to the
development of a housing policy of national and local authorities
in their efforts to improve housing conditions.
The interest of the Office in statistics of housing has led to its
participation in the work of two Committees, the Committee of
Experts on Housing Statistics of the International Union of Towns
and the Committee of Statistical Experts of the League of Nations.
For the meeting of the former in Munich in May 1928 the Office
prepared a report on Methods of Compiling Housing Statistics.2
Its resolutions on statistics of housing are included in full in Part II,
which includes also extracts from the report of the Committee of
Statistical Experts of the League proposing a detailed plan of
housing statistics.
The resolutions of the two Committees in part overlap: both
give definitions of room, dwelling, occupant, and household; the
former includes a definition of rents, as well as a recommendation
concerning statistics of dwellings classified according to the amount
of rent paid. Both give detailed specifications for censuses of dwellings and for their classification according to certain important
characteristics—number of rooms, equipment, and occupancy.
The problem of standardisation of these statistics is important
not only among different countries but also in the same country
between the different branches of these statistics (censuses of buildings and dwellings, statistics of the housing market, and statistics
of rents), which are often within the competence of different departments.
1

See also Rents, p. 30 above.
* Studies and Reports, Series N, No. 13 (Geneva, 1928).

INTERNATIONAL STANDARDISATION OF LABOUR STATISTICS

The different statistics of housing as published in the different
countries have not yet been included in the scope of regular compilations by the Office. In a study of housing policy in Europe1,
however, the principal results of the censuses of dwellings as well
as statistics of housing construction for the years 1920-1927 were
given systematically, following the proposals of the resolutions
of the Committee of Experts in Housing Statistics. Reference has
already been made to the results of the surveys of rents in the principal towns of different countries in the section dealing with international comparisons of rents.
The importance of similar classifications of dwellings in housing
censuses and i:i current statistics of rentals paid must be emphasised
not only for comparisons of rental costs in different communities
but especially in international comparisons.

1
Housing Policy in Europe, Studies and Reports, Series G, No. 3 (Geneva,
1930); see also, R. GUYE: "Post-War Building Difficulties and Housing Policy
in Europe", in International Labour Review, Vol. XXIV, Nos. 2-3, Aug.-Sept.
1931, pp. 200-227.

IX
COLLECTIVE AGREEMENTS
The Office prepared a Report on Methods of Compiling Statistics
of Collective Agreements for the Third International Conference
of Labour Statisticians. 1
The resolutions on this topic adopted by the Conference are
somewhat similar to those on industrial disputes and recommend
detailed methods, definitions and classifications for these statistics. 2
Very few countries, however, have elaborated this branch of labour
statistics, which depends to some extent on the forms of trade
union organisation and the labour legislation of the country. 3
Annual figures on this topic were compiled by the Office up to
1935 showing, both for the number of new agreements concluded
during the year and the number in force during the year, the number
of establishments and workers covered. Only a few countries (11),
however, could be included in these statistics. 4

1
Studies
%

and Reports, Series N, No. 11 (Geneva, 1926).
For the discussions at this Conference, see The Third International Conference of Labour Statisticians, Studies and Reports, Series N, No. 12 (Geneva, 1926),
pp. 838-57.
For a discussion of the statistics in Germany, see W. WOYTINSKY: "New
Statistics of Collective Agreements in Germany", in International Labour Review,
Vol. XXIII, No. 4, Apr. 1931, pp. 506-532.
4
For notes, sources, etc., of these statistics, see The I.L.O. Year-Book, 193435, Vol. II: Labour Statistics, pp. 162-163, 245-247.

X
INDUSTRIAL DISPUTES
A Report on Methods of Compiling Statistics of Industrial
Disputes was prepared for the Third International Conference of
Labour Statisticians. 1
The resolutions adopted by the Third Conference included
recommendations that full details of industrial disputes classified
by industry, method of settlement, by results, by importance (i.e.
number of establishments, number of workers and number of days
lost) should be published at least yearly in each country. 2 The
method of classification recommended for each of these characteristics was given in detail.
The Office has not attempted to analyse the available statistics
by all the criteria indicated above, but has limited itself to publishing figures showing the number of establishments, number of workers involved and number of days lost.8 In the Year Book of Labour
Statistics, 1939, figures for some 35 countries were included. In some
countries, however, industrial disputes have been forbidden by law.
The statistics are of interest chiefly in showing the loss caused by
this form of interruption of work.
A comparative table showing for 19 countries the loss caused
(in man days) by industrial disputes in certain branches of industry
—manufacture, mining, transportation—in relation to the persons
gainfully occupied in these branches was published in the International Labour Review in May 1938.*

1
8

Studies and Reports, Series N, No. 10 (Geneva, 1926).
For the discussions at this Conference, see Studies and Reports, Series N,
No.8 12 (Geneva, 1926), pp. 38-85.
See also "Industrial Disputes", in International Labour Review, Vol. XXVIII,
No.4 1, July 1933, pp. 92-99.
"Days Last through Industrial Disputes in Different Countries", in International Labour Review, Vol. XXXVII, No. 5, May 1938, pp. 674-678. For an
earlier analysis on similar lines, see E. J. RICHES: "International Comparisons of
the Time Lost through Industrial Disputes", in Economic Record (Melbourne),
Vol. IX, No. 17, Dec. 1933, pp. 226-234.

XI
EMIGRATION AND IMMIGRATION
As far back as 1921, the International Emigration Commission
set up by the First Session of the International Labour Conference
at Washington, 1919, passed resolutions recommending inter alia,
that the various Governments should communicate regularly to
the Office all available information concerning emigration and
immigration. As a sequel to this, the Fourth Session of the International Labour Conference in 1922 adopted a Recommendation
requesting the various Member States to communicate all available
information and also invited the International Labour Office to
make every effort to facilitate an international co-ordination of
migration statistics. 1 As a consequence of these decisions, the
Office compiled and published the data as articles in the International Labour Review and in the form of special studies. 2
A special Conference on Migration Statistics was called in
October 1932.3 For this purpose, a study of the methods adopted
in different countries was prepared and published. 4 The resolutions
adopted by the Conference dealt with the definition of a migrant,
the various methods of collecting the information, and the methods
of classifying the data. In the international tables compiled annually
by the Office, and published in the Year Book of Labour Statistics
it has not been possible to analyse the data by all the different
criteria indicated by the Conference. Emigration and immigration
figures are given separately, and subdivided as far as possible
1
For the text of this Recommendation, see The International Labour Code 1939,
pp. 555-556.
!
These studies were published as follows: Migration Movements 1920-1923,
Studies and Reports, Series O, No. 1; Migration Movements 1920-1924, Studies
and Reports, Series O, No. 2; Migration Movements 1925-1927, Studies and
Reports, Series O, No. 4. See also "Migration Movements", in International
Labour Review, Vol. X X , Nos. 2 and 5, Aug. and Nov. 1929, pp. 259-283 and
719-740.
3
An account of this Conference is contained in "An International Conference
on Migration Statistics", in International Labour Review, Vol. X X V I I , No. 1,
J a n . 1933, pp. 1-24.
4
Statistics of Migration: Definitions—Methods—Classifications,
Studies and
Reports, Series N, No. 18 (Geneva, 1932).

INTERNATIONAL STANDARDISATION OF LABOUR STATISTICS

into nationals and aliens, and into continental and intercontinental
movements. The different sources of migration statistics enumerated
by the Conference are also clearly indicated: port statistics; declarations of residence; frontier control; coupons detachable from documents (e.g. passports), etc.1

1
In addition, mention should be made of an article b y L. VARLEZ: "Seasonal
Fluctuations in Migration", in International Labour Review, Vol. X V I I , Nos. 4
and 5, Apr. and M a y 1928, pp. 486-514, 671-686; an article by Imre FERENCZI:
" A Historical Study of Migration Statistics", ibid., Vol. X X , No. 3, Sept. 1929,

pp. 356-384; NATIONAL B U R E A U OP ECONOMIC R E S E A R C H : International

Migra-

tions (New York, 1929), Vol. I of which contains statistics compiled on behalf
of the International Labour Office with introduction and notes by I. FERENCZI;
and World Statistics of Aliens, Studies and Reports, Series O, N o . 6 (Geneva,
1936).

XII
YEAR BOOK OF LABOUR STATISTICS
Finally, the recommendation of the Fourth Conference that an
annual volume on wages and prices should be published which
would be in essentials a development and amplification of the
tables published at regular intervals in the International Labour
Review has been realised by the publication of the Year Book of
Labour Statistics. The statistics in this volume are not confined
to wages and prices, but extend to other branches of labour statistics with which the Conferences of Labour Statisticians have dealt,
including employment and unemployment, cost of living, family
budgets, food consumption, industrial disputes, emigration and
immigration, and industrial accidents. In addition, a chapter on
the gainfully occupied gives basic figures for the labour supply,
and tables furnished by the Economic, Financial and Transit
Department of the League of Nations give information on exchange
rates, wholesale price indices, and world indices of production
necessary to furnish a background for the adequate interpretation
of the data in the other tables.
The Year Book, commencing formally with the 1935-36
issue, was a natural outcome of the work of the International
Labour Office in the field of labour statistics. The first
number of the International Labour Review (Jan. 1921), contained
statistical material on wholesale and retail prices, cost of
living and unemployment; a special article on wage changes
appeared in April 19231; and from January 1924 a separate section
in the Review was devoted to statistics. Supplementing this material
were the statistics published in the various special studies on employment and unemployment 2 , wages3, and migration. 4 The first volume
» Loc. cit., Vol. VII, No. 4, pp. 577-603.
Statistics of Unemployment in Various Countries 1910 to 1922, Studies and
Reports, Series C, No. 6 (Geneva, 1922); Unemployment, 1920-1923, No. 8
(Geneva, 1924).
' Wage Changes in Various Countries 1914-1921, Studies and Reports, Series
D, No. 2 (Geneva, 1922); No. 10, Wage Changes in Various Countries 1914 to 1922
(Geneva, 1923); No. 16, Wage Changes in Various Countries 1914 to 1925
(Geneva, 1926).
* Migration Movements 1920-1923, Studies and Reports, Series O, No. 1 (Geneva, 1925); No. 2, Migration Movements 1920-1924 (Geneva, 1926); No. 4,
Migration Movements 1925-1927 (Geneva, 1929).
3

INTERNATIONAL STANDARDISATION OF LABOUR STATISTICS
TABLE I.

SCOPE OP TABLES AND NUMBER OF COUNTRIES 1 COVERED

Subject
Population
Gainfully occupied
Structure of the gainfully occupied population:
industrial status, and industry or occupational groups 2
Sex and age groups 3
II.

Employment and unemployment 4
General level of employment
Indices of numbers employed:
A. Mines, industries, transport, commerce, etc
B. Industries
Indices of total hours worked in industries
Employment, by industries
General level of unemployment 5
Unemployment, by industries 6

III.

Hours of work
(ieneral level of hours of work
A. Hours actually worked per worker
B. Range of hours actually worked: percentage distribution.
C. Normal hours
Hours worked by industries
A. Average hours actually worked
B. Range of hours actually worked: percentage distribution.
C. Normal hours'
Hours of work in October (normal)

IV.

Wages
(ieneral level of wages
A. Hourly rates or earnings
B. Daily, weekly and monthly earnings 8
Wages by industry or occupation
A. Hourly rates or earnings
B. Daily, weekly or monthly earnings 9
Wages per hour in October, in selected occupations
Total wages and salaries

VI.

VII.

VIII.
IX.

Cost of living and retail prices
Cost of living, general indices 10
Retail prices of foodstuffs, fuel, in October 11
Family living studies
Sources of family income (wage earners' families)".
Direction of expenditures (wage earners' families) 12 .
Direction of expenditures b y income classes
Average food consumption 14
Migration
Emigration and immigration
Net intercontinental migration
Labour relations 16
Industrial disputes (strikes and lockouts).
Industrial accidents
Mining
1 ndustry
Railways

> As countries are counted those which had separate status at the time the Year Book was issued or
have been shown as separate countries in subsequent editions. Where dashes are shown, no table on the
subject was published in the Year Book. ' In 1937-1938 a special summary table showed the gainfully
occupied according to industrial status, with 19 and 24 countries respectively. « In 1937, the gainfully
occupied were shown for different age and sex groups;
in 1941, the percentages gainfully occupied by sex,
for three age groups, 15-19, 20-64, and 65 and over. 4 In addition, world indices
of employment were given
for 1937 to 1941, and world indices of unemployment for 1935-36 to 1941. 5 In addition national indices
(of degree of unemployment) are given for 16 countries for
1935-36,
1937
and
1938.
• In 1937 and 1938,
figures of unemployed by sex were given for 18 countries. 7 By industries or occupations. 8 Including earnings per shift, and insured daily or weekly wage. • Wages in agriculture, 21 countries in 1935-36, 22 coun-

YEAR BOOK OF LABOUR STATISTICS

IN THE YEAR BOOK OF LABOUR STATISTICS, 1935-36 TO 1941
N u m b e r of countries for which statistics on the specified topics are
shown in Year Books of
1935-36

1937

1938

1939

1940

1941

1942

—

18
17

22
—

29
—

30
—

33
IS

32
18

—
—
—
10
33
17

—
—
—
16
32
17

17
16
—
17
32
17

18
18
12
17
31
17

16
23
12
17
31
16

18
29
13
23
31
17

18
29
13
26
30
17

13
10
—
—
12

12
10
6
—
11

14
12
6
—
11

15
15
8
8
9

17
17
8
7
15

3
7
26

—
5
26

—
5
21

17
17
8
7
15
10
—
5
19

4
—
21

3
—
23

14
12
6
—
12
10
3
7
23

26
17
16
28
18
15
21
—

27
19
15
30
19
16
23
21

31
19
24
29
20
16
24
21

31
19
24
28
20
15
27
21

29
24
20
24
18
11
27
21

34
26
27
27
18
16
23
26

34
26
27
27
18
16
20
27

43
24

44
26

48
25

51
32

52
35

53
31

56
35

11
20
8
14

15
21
14
—

15
22
18
—

21
26
22
17

23
27
—
21

24
30
—
26

25
32
29
29

63
21

59
21

60
23

61
23

—

—
—

14
20
20

14
20
21

—
—

tries in 1938; wages in coal mines, 7 countries in 1938. i° Group indices for the same number of countries
were
published in 1935-36, 1937, 1938 and 1939; indices for food only in 1941; in 1942, for 54 countries.
11
Figures on monthly rents of workers' dwellings in 19 countries were published in the Year Book for 1939.
Indices of international comparisons of food costs
were published in the Year Book for 1941 for 25 countries;
12
in the Year Book for 1942, for 19 countries.
For non-manual workers' families in 10, 13 and 13 countries
in 1939, 1940 and 1941 respectively. 1S For non-manual workers' families in 11, 14 and 15 countries in 1939,
1940 and 1941 respectively. The distribution of food expenditures (vas given for 12 countries in 1935-36
and 16 in 1937. » In 1935-36, according to income classes. 16 In 1935-36, figures on collective agreements
in force were given for 7 countries, and for collective agreements, renewed, or modified for 10 countries«

INTERNATIONAL STANDARDISATION OP LABOUR STATISTICS

of the I.L.O. Year Book in 1930 contained a few tables on cost of
living, wages and unemployment, scattered through the text in
connection with the various topics. In subsequent volumes the
scope of these tables was expanded: in the I.L.O. Year Book for
1932 labour statistics appeared in 5 appendices with 8 tables; in
1933, an entire Appendix II was devoted to labour statistics, with
5 groups of topics; in 1934-35, Labour Statistics appeared as Volume
II of the I.L.O. Year Book: and in subsequent years, the Year Book
of Labour Statistics appeared as an independent publication.
Table I portrays the changes in scope of the tables of the Year
Book, showing the number of countries included in the principal
tables given in each edition. As a complement to this, and to give
a background of the expanding scope of these statistics, table II
shows the changes in the number of countries covered in the tables
of the International Labour Review from 1924 to 1935 (prior to
the publication of the Year Book). The increase in the number of
countries is not presented as in any sense an exact measure of the
improvement in the various statistics over the period; it reflects
in part a greater diligence on the part of the Office in obtaining
the data and a desire to include as many countries as possible
instead of merely the principal ones: it remains true, however,
that the improvement in these statistics is both real and important,
and that the drive for better statistics in the several lines in the
different countries is the principal factor in the broadening scope
of the tables. A point of especial interest throughout the series is
TABLE I I . NUMBER OF COUNTRIES COVERED IN TABLES ON SELECTED
TOPICS IN THE INTERNATIONAL LABOUR REVIEW, 1 9 2 4 - 1 9 3 5 1

Number of countries for which statistics are shown
in the International Labour Review

Subject

1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935
Employment
(general level)
Unemployment 4
(general level)
19
Wages:
General level
Wages in October»
Cost of living and
retail prices:
Cost of living 22
Retail prices in
October»

1
Selected topics only as specified. The number of countries is that of the table which was the
first on the specified topic published during the year. s In some cases for a month other than
October.

YEAR BOOK OF LABOUR STATISTICS

the wide differences between the coverage of the different series:
most countries have cost of living figures, but a much smaller
number have good employment, unemployment and wages data.
Together these two tables thus give a rough picture of the
progress accomplished in the international standardisation of
labour statistics, as more and more countries develop and publish
statistics on the various topics which can be included in the comparative tables. The expanding scope of the materials available
in the different countries reflects also the growing interest in labour
statistics on the several topics.

CONCLUSION
In this short introduction to the resolutions which follow, it
has only been possible to touch briefly on the scope of the resolutions adopteid and on the work of the Office in publishing the various
statistics, in describing the methods adopted, and in analysing the
results; the reader is referred to the detailed studies on particular
subjects listed in the notes on the foregoing pages for fuller information o.a these questions. The introduction has, however,
been sufficient to show that the Office has not been inactive in
endeavouring to obtain, as a help to the solution of the problems
with which it is called upon to deal, full and comparable statistics
on all subjects which affect the conditions of labour. For many
countries, statistics on many branches are still not available at
all, or are available only at very irregular intervals, and even
the most highly industrialised and best-equipped countries have
serious gaps in certain of their statistics. The resolutions given in
the following pages may, in specific fields, fall short of completeness
and desirable detail—gaps which future conferences may be expected to fill. As a whole, however, they set high standards as a
goal to be reached by the labour statistics of the several countries.
Considerable progress towards the application of these resolutions
has been made, and together they form a scheme which will lead,
it is hoped, to still further improvements in the future.

PART II
Documentary Texts

I
THE CLASSIFICATION OF INDUSTRIES AND OCCUPATIONS FOR PURPOSES OF LABOUR STATISTICS
1. Resolutions of the First International Conference of
Labour Statisticians, Geneva, October 1923
[f] (1) Occupied persons should be classified in the first
instance according to the industry in which they are employed,
and within each industry they may be further classified according
to their individual occupations. When it is not possible to give
this double classification in sufficient detail to show the total number
of workers in each individual typical occupation, it is necessary
to make a second classification of all occupied persons according
to their individual occupations, so that for comparative purposes
two separate classifications will be available (a) by industry and
(b) by individual occupation.
\_2\ (2) Industries should be classified into the following main
divisions :
(a) Primary Production
Agriculture, pasturing, forestry, hunting, fishing, etc.; mining,
quarrying, etc.; i.e. extraction of minerals.
(6)

Secondary Production
Manufacturing industries, etc. ; i.e. the transformation or modification of materials, together with the construction of buildings, roads, etc., and the repair of finished products.

(c) Services
Transport and communication;
Commerce, finance, and trade;
Public administration and defence;
Professional services;
Domestic service and personal services for which remuneration
is paid; supply of board and lodging.
[3] (3) In classifying manufacturing industries the establishment considered as a technical unit should be taken as basis.

INTERNATIONAL STANDARDISATION OP LABOUR STATISTICS

M (4) In the absence of an agreed classification of industries
and as a preparation for such a classification, and in order to facilitate international comparison, the groupings of industries used in
the different: countries should be so subdivided that it would always
be possible to secure separate information concerning an adequately
complete number of industries included in a provisional list drawn
up in alphabetical order, which might be prepared by the International Labour Office after consultation with any statistical
services or organisations which might usefully assist in this work.
K l (5) It is desirable for purposes of international comparison
that each country should publish definitions of the occupational,
industrial and other terms most commonly used in that country
in connection with its labour statistics.
2. Resolution of the Second International Conference of
Labour Statisticians, Geneva, April 1925
jjf] The Conference, realising, as it did in 1923, that it is not
possible at the present time to reach an agreement as to the scientific bases of an international classification of industries, but recognising that it is necessary to ensure in the immediate future some
form of international comparison, at least for the principal industries,
Recommends that pending the time when a complete classification becomes possible there should be drawn up, after consultation
with the various statistical services and international organisations
concerned, a. provisional list of the most important industries.
In order that the question of drawing up a more complete
list of industries which would serve as a base for international
statistical comparisons might be thoroughly investigated,
The Conference recommends the International Labour Office
to invite the Governments to examine the draft classification
which has been submitted to them, and to inform the Office of the
difficulties which they would experience in adopting it, calling
attention to cases where the draft list might be reduced or expanded.
3.

Resolutions of t h e Third International Conference of
Labour Statisticians, Geneva, October 1926
DG (1) This Conference recommends the International Labour
Office to elaborate the list of the most important industries submitted to it so as to show in greater detail the contents of each
industrial group in the different countries, obtaining from the
Governments any necessary information on the subject. This
amplified list should be submitted in due course to the different
Governments for their information and comment.

CLASSIFICATION OF INDUSTRIES AND OCCUPATIONS

[<?] (2) The Conference also recommends the International
Labour Office to pursue its endeavours in collaboration with the
League of Nations and the International Statistical Institute to
compare the existing classifications and to promote their gradual
readjustment in the direction of greater uniformity and better
comparability.

4. [_9] International Scheme for the Presentation of Data by
Industry (or Occupation), framed1 by the
International Labour Office
Primary Production:
I.
Agriculture, etc.
A = Agriculture
fishing).
II.

(including forestry, hunting, and

Mines.
B = Mines (including quarries and oil wells).

Secondary Production:
I I I . Industries (manufacturing, construction, etc.).
C = Metal (metallurgy, mechanical and electrical
engineering, construction and repair of vehicles, ships, etc.).
D = Brick and stone, pottery and glass.
E = Building and construction.
F = Wood (felling, floating, saw-milling, and woodwork, including furniture).
G = Paper and printing.
H = Textiles.
I = Clothing (including boots and shoes).
J = Skin, leather, and rubber.
K = Chemicals.
L = Food, drink and tobacco.
M = Water, gas and electricity (production and distribution).
Services:
IV. Transport and communications.
N = Land transport (road and rail).
O = Navigation (inland, maritime and air).
P = Post, telegraph and telephone.
1
First published in the International Labour Review, Vol. X X V I I I , No. 3,
Sept. 1933, pp. 414-415; the list was adopted provisionally for the purpose of
presenting the labour statistics published by the Office in a uniform order. The
list given is the latest revision.

INTERNATIONAL STANDARDISATION OF LABOUR STATISTICS

Public administration.
Q
R

VI.

= Street cleaning, police, fire and other public
services not mentioned above.
= Public administration (properly so-called).

Commerce, personal and professional services, etc.
S
T

= Commerce, banking and insurance.
= Hotels and restaurants, domestic service, laundries, hair-dressing, etc.
U = Entertainments, sports, etc.
V = Professional services, education, etc.
Not Classifiable:
X
Y

= Miscellaneous.
= Occupations covering several branches, e.g.,
clerks, general labourers, etc.

5. Excerptsi from the Report of the C o m m i t t e e of Statistical
Experts of t h e League of N a t i o n s o n Statistics of t h e
Gainfully Occupied Population 1
Definitions and Classifications Recommended
A.

DEFINITION OF THE GAINFULLY-OCCUPIED POPULATION

Definition

£l02 For the purpose of international classification, any occupation for which the person engaged therein is remunerated,
directly or indirectly, in cash or in kind—i.e., any principal remunerated occupation or any secondary occupation which is the sole
remunerated occupation of the person concerned—is to be considered as a gainful occupation. Housework done by members of
a family in their own homes is not included under that description,
but work done by members of a family in helping the head of the
family in his occupation is so included, even though only indirectly
remunerated. The occupation of persons working in labour camps
or other similar institutions, or engaged in unemployment relief
work, is to be considered as a gainful occupation.
[ i í ] The particulars given should be based, generally speaking, on the occupation at the moment of the census. A person who
1

LEAGUE OF NATIONS: Studies and Reports on Statistical Methods, No. 1:

Statistics of the Gainfully-Occupied Population. Definitions and Classifications
recommended by the Committee of Statistical Experts (Official No. C. 226 M. 128.
1938 (C.E.S. 127) Appendix I; Ser. L.O.N., Pub. II, Economic and Financial,
1938, II. A. 12).

CLASSIFICATION OK INDUSTRIES AND OCCUPATIONS

has recently exercised a gainful occupation is to be considered as
still engaged in that occupation even though, by reason of sickness, injury, vacation or inability to obtain work, he may, at the
time of the census, be temporarily not working.
\_12~] Young persons of working age and not at school, who
have never actually exercised a gainful occupation, are not to be
treated as part of the gainfully-occupied population, even though
they may be seeking work and consequently included in statistics
of unemployment. It is, however, desirable that censuses should
be so taken that the number of young persons so situated can
be ascertained.
//.

Distinctions in the Gainfully-Occupied
Population
Required for International Comparison

\_13~] For reasons stated above . . . [not reproduced here],
it is essential, in order to bring within the range of international
comparability the totals of the gainfully-occupied population,
that the following groups and sub-groups be shown separately
among the persons conforming to the definition of Section I.
Gainfully-Occupied

Population.

\_14~] (1) All persons gainfully occupied in the above-defined
sense . . . with the exception of those specified in group 2 below.
[_15~\ (2) Members of families, whatever their degree of relationship, aiding the heads of their families in their occupations,
showing separately:
(a) Wives
(6) Wives
(c) Other
(d) Other
farmers.

of farmers;
of persons other than farmers;
members of farmers' families;
members of families of persons other than

[_16~] The term "farmers" shall be deemed to apply exclusively
to the heads of enterprises classified under Chapter I of the Nomenclature of Industries. It is to be understood, however, that a
person who works for the head of his family is to be classed under
group 1, if the information regarding his personal status, or other
available information, justifies the inference that he receives regular
wages for that work.
III.

Persons Not Considered as Gainfully

Occupied

\_17^\ For the same reasons which necessitate the classification
set forth [above] it is essential that among those not conforming

INTERNATIONAL STANDARDISATION OF LABOUR STATISTICS

to the definition of persons gainfully occupied, or, by reason of
the system current in their countries, not included in the record
of persons gainfully occupied, the following groups and sub-groups
be shown separately:
Population

Not Recorded as Gainfully

Occupied.

\_18~\ (1) Members of families of working age, capable of work
and not attending school, distinguishing:
(a) Wives
(6) Wives
(c) Other
(d) Other
farmers.

of farmers;
of persons other than farmers;
members of farmers' families;
members of families of persons other than

[_19~\ (2) All other persons, including persons living on private
means or maintained by their families or from public or private
resources—e.g., rentiers and pensioners, annuitants, landlords;
children, including those attending school, and students, whether
living with their parents or guardians or not; aged, sick or infirm
persons living with their families or in hospitals, sanatoria ór
asylums; inmates of prisons. Persons only temporarily unoccupied
should be included in the occupied population.
IV. Definition and Classification of the Unemployed
\_20\ The Committee examined the difficult problem of the
classification of persons who at the moment of the census are unoccupied owing to lack of work, and decided to recommend a definition of the gainfully-occupied population which should be inclusive of such persons.
\_21~\ Persons recorded at the census as unemployed should be
classified according to their last occupation. In national tables, the
Committee thinks it advisable that such unemployed persons should
be counted separately, as classification by their most recent occupation may give an erroneous idea of the importance of some
branches of economic activity which are on the decline. For the
reasons given below, the Committee has not made this distinction
in the international programme.
\_22~] In conformity with the general principle according to
which information provided should be based on the occupation at
the time of the census, the persons and unemployed men working
in labour camps and similar institutions, or on unemployed relief
work should be classified in a special category and not according

CLASSIFICATION OP INDUSTRIES AND OCCUPATIONS

to their particular task at the date of the census (a principle applying also to the classification of persons undergoing military service).
In view of the great differences existing between the systems for
relieving unemployment in the various countries (establishing
labour camps, organising public works, granting subsidies to
existing industries, etc.) the Committee does not believe that the
separation from other "gainfully-occupied persons" or individuals
receiving assistance of the kind indicated would increase the
accuracy of international comparisons.
[23] The Committee realises that the above recommendations
do not allow of a bird's-eye view of a country's unemployment
situation, or of the total number of persons potentially belonging
to a particular branch of activity or having a particular occupation. The picture provided by a census, however, can be only of
an instantaneous nature and cannot throw light on all aspects of
the problem of unemployment. For this purpose, a special census of
unemployed would be necessary.
B. T H E D I F F E R E N T SYSTEMS OF CLASSIFYING THE GAINFULLYOCCUPIED POPULATION

\_Z4\ By a recommendation of the International Conference
relating to Economic Statistics, the Committee was asked to study
the question of systems of classification of the occupied population
by industry and by personal occupation, and the distribution of
the occupied population according to occupational status (i.e.,
employers, employed, etc.). The Committee has accepted the
view, already expressed by other conferences of statisticians, that
a classification according to the three following principles:
(a)
(&)
(c)

By branches of economic activity;
By personal status;
By individual occupation

is necessary to obtain an adequate picture of the economic and
social structure of the population.
\j25~] As regards classification by branches of economic activity,
the Committee has endeavoured, for the reasons given in paragraph
C, to apply this principle as strictly as possible. In particular,
it has been anxious to avoid overlapping of this principle with
that of individual occupation ; and, for that reason, the Committee's
classification by branches of economic activity will be found not
to include a section for the "liberal professions", which was included
in the classifications drawn up by the International Conferences of

INTERNATIONAL STANDARDISATION OP LABOUR STATISTICS

Labour Statisticians convened by the International Labour Office
in 1923 and by the International Statistical Institute in 1925. l
[,26] The Committee wishes to emphasise that, from the international point of view, the first two principles should be applied
in the form of a cross-classification, in which persons belonging to
the same branch of economic activity are distinguished in accordance with their personal status within that branch.
C. CLASSIFICATION BY BRANCHES OF ECONOMIC ACTIVITY

I. Object and Nature of this

Classification

[27] The Committee's recommendations with regard to the
classification by branches of economic activity assume that the
census cards or schedules contain a question, or even several
questions, on the nature of the establishment in which the occupation is carried on. Information obtained by means of the question
regarding the individual occupation only does not enable the grouping recommended from the international point of view to be carried
out. On the other hand, a question regarding the individual occupation furnishes supplementary information which is very useful,
and sometimes even indispensable, for grouping persons by branches
of economic activity, especially in cases where different branches
of activity are carried on within the same establishment.
[28] The Committee is of the opinion that the main object of a
classification by branches of economic activity for the purposes of
statistics of the gainfully-occupied population is to show the absolute and relative importance of the various branches of industry. 2
It should be understood that industries are carried oh in establishments, and that a branch of industry should therefore comprise
all the persons occupied in a given kind of establishment, regardless
of what individual occupation they perform within that establishment. T h a t being so, it is the aim of the classification by individual
occupations to show how the personnel belonging to a given branch
of industry is composed from the point of view of individual occupations, and to supply, if desired, total figures for all persons exercising
the same individual occupations. The Committee thinks that any
procedure under which persons might, on account of their individual
occupation, be assigned to another branch of industry than that
in which they are actually employed would distort reality, and lead
to rather meaningless results from every point of view.
1
By recommendation IV of the International Conference relating to Economic
Statistics,
the Committee was asked to take these classifications into account.
2
For convenience, the term "industry" is frequently used throughout this
statement for a.ll branches of human economic activity.

CLASSIFICATION OP INDUSTRIES AND OCCUPATIONS

[29] The Committee wishes to make this point of principle
entirely clear, because it is of the utmost importance for the international comparability of the various branches of economic activity
and because its practical application involves, in certain cases,
great difficulties.
II.

Definition

"Establishment"

[30] For the reasons set out above, a classification of persons
by branches of economic activity requires a classification of establishments, which in turn presupposes a definition of the term "establishment".
[31] The term "establishment" is used hereafter not merely
in connection with production, transport and commerce, but as
covering also all units belonging to other branches of economic
activity. For instance, schools, lawyers' offices and dentists' consulting-rooms are covered by the term "establishment".
[32] There are three different conceptions of an establishment
to be distinguished.
[33] (1) The narrowest conception is that which is based on
the technique of the work and which considers an establishment
as comprising all persons working together under a certain direction and in closely related individual occupations. Under this
conception, the furniture repair shop of a public administration,
the painting shop of a motor manufacturing plant, the shops where
wooden models are made in engineering works, would become
separate establishments, and the persons occupied therein would
be assigned, e.g., to the furniture, building and woodworking industries respectively.
[34] (2) The second conception is based on the place at which
the work is performed and considers all persons working together
in the same factory or works as belonging to one and the same
establishment.
[35] Under this conception, the above "establishments"
(painting shop, furniture repair shop, etc.) would not be recognised
as separate establishments, and the persons occupied therein
would be assigned to the branch to which the whole factory belongs.
[36] (3) The third conception is based on the legal and financial
organisation of the enterprise and considers all persons working
together under the control of the same concern as belonging to
one and the same establishment—e.g., the various branch shops
belonging to the same entrepreneur. This conception, which has
considerable importance for censuses of industrial and commercial
establishments, is, in practice, not used for censuses of the gainfully-occupied population and can therefore be disregarded here.

INTERNATIONA!, STANDARDISATION OF LABOUR STATISTICS

[_37~] For the reasons outlined above under C, I, the recommendations of the Committee are based on the second conception
of an establishment. It is the success or the failure of the factory
as a whole which mainly governs the situation of an occupied
person; the "technical" establishments as defined under I have
no independent life of their own. The persons are thus dependent
on the development of the branch of industry to which the factory
as a whole belongs, and it appears therefore reasonable to show
these persons and their individual occupations in the framework
of an industrial classification which takes the factory as a whole
as the unit to be classified. In addition, this conception is the
easiest to follow with the methods of a census of the gainfullyoccupied population.
Q?£] There is no doubt that the practical application of the
principle stated above meets with serious difficulties. The principal
difficulty arises in connection with establishments combining a
number of different activities. The Committee does not believe
that indefinite or defective information as to the nature of an
establishment (or of that part of an establishment in which the
occupation is carried on) can be improved by the insertion in the
census cards or schedules of a definition of "establishment", however unchallengeable such a definition might be from a theoretical
standpoint. The following considerations are, therefore, intended
to serve rather as guiding principles for the preparation of the
material than for the collection of the basic data.
\J9] In order to ascertain the nature of any given establishment
and, if necessary, to allocate the individuals to the different branches
of an establishment, the statistical offices are frequently compelled
to take the individual occupation as a criterion. Information from
other sources (census of establishments, registers of establishments,
list of addresses, etc.) may also be useful in this connection.
\jf02 As international comparisons of the results are largely
dependent on this distribution, by branches of establishments,
being carried out according to uniform principles in the different
countries, the Committee believes it to be desirable to indicate
general guiding lines, even though a strict application of uniform
rules may sometimes be found very difficult.
\jtl2 The following conditions may lead to the combination of
different branches of activity in a single establishment—on the
assumption that all persons working together in the same place,
factory, workshop, etc., belong to such an establishment:
\i42~\ (1) An establishment may include branches of an accessory character working in connection with the main process of
production or forming a part of that process.

CLASSIFICATION OF INDUSTRIES AND OCCUPATIONS

Examples.—The manufacture of wooden models in an
engineering works, a printing press in a cigarette factory,
or painting and upholstering shops in motor manufacturing
works, a small repair shop in a commercial business.
[43] (2) A number of successive stages of production, transformation or distribution may be combined in the same establishment.
Examples.—Carding, spinning and weaving may be carried
on in one establishment; a mine combined with a blast furnace,
a rolling mill and an engineering works; an agricultural enterprise combined with a dairy manufacturing cheese, a factory
combined with a retail shop. (Such small establishments as
bakers', butchers', etc., shops, in which the retail sale is to a
large extent carried on by the members of the family of the
head of the establishment, who help sometimes in the manufacture and sometimes in the sale, are not considered here.
In the case of these establishments, the retail sale should be
treated as accessory, and consequently assimilated with the
cases under (1).)
[44] (3) Two branches of economic activity of a different
technical character may be carried on in the same building and
under the direction of the same proprietor without the existence
of an economic relationship of the kind mentioned under (1) and (2).
Examples.—A flour-mill and a sawmill depending on the
same motive power, a butcher's shop combined with a hotel,
a motor-car sales establishment combined with a repair shop
which cannot be considered as accessory.
[45] If combinations of type (2) or type (3) come under the
same group in the International Minimum Nomenclature of Industries, it will not affect the international comparison of results
whether the different branches are, or are not, shown separately.
Where, however, the different branches fall into different groups
of the Minimum Nomenclature, the Committee thinks that the
persons belonging to each branch should be grouped with that
branch. Persons working in a mine, for example, should be shown
separately from those working in associated metallurgical or engineering works: but persons working in combined carding, spinning
and weaving establishments may be shown together, since the
Minimum Nomenclature comprises these branches of industry
in a single group.

INTERNATIONAL STANDARDISATION OP LABOUR STATISTICS

[462 Combinations of type (1) above should not be split up,
and all persons should be classified in accordance with the nature
of the principal branch—e.g., persons engaged in the carpentering
or upholstering or painting shops of motor manufacturing works
should all be classified under the motor industry.
[473 It goes without saying that the splitting-up of different
branches of an undertaking creates almost insuperable difficulties
in dividing between them, in appropriate proportions, the administrative and operative staffs of the services that are common to
these different branches. But the effect of uncertainties thus arising
on the international comparison of results is small as compared
with the effect resulting from the allocation being made in one
country according to branches of establishments, while, in another,
persons occupied in all branches of an establishment are classified
under its principal branch.
[482 It should also be made clear here that the juridical form
of ownership does not influence the definition of an establishment.
For example, printing works, gasworks and shipyards belonging
to a public authority are not to be considered as a branch of "public
services"; they should be treated in the same way as private establishments of a corresponding character, but counted separately
in the respective groups.
III.

International Minimum Nomenclature of Branches
of Economic Activity1

1. Purpose and Length of the Nomenclature.
[.49] The main purpose of an international classification of
branches of economic activity is to facilitate the comparison of
national statistics. The Committee considers that, in practice,
the best means of attaining this purpose would be for every country
to give the necessary details for regrouping the items of its national
classification in accordance with an international plan of classification, and publishing the results of this regrouping as a supplement
to its own classification.
[502 The Committee has deemed it advisable to establish an
international nomenclature of medium length, enabling a reasonably clear picture to be obtained of the economic structure of each
country. This nomenclature comprises forty-five chapters grouped
in nine sections. The Committee recommends that, for each country,
in addition to the results of its census according to its own national
1

See below, pp. 65-66.

CLASSIFICATION OF INDUSTRIES AND OCCUPATIONS

nomenclature, there should be published a supplementary table
grouping the data according to these forty-five chapters as the
indispensable minimum for international comparison. A key,
indicating which items of the national nomenclature are included
in each chapter, should accompany this table.
{51} In the case of twenty of the forty-five chapters, the Committee has distinguished two or more items. The total number of
items in the nomenclature thus reaches ninety-four. 1 The indications
of the contents of certain items should not be regarded as complete ;
generally speaking, they are given only in cases in which doubts
may arise as to the allocation of a particular branch of economic
activity. The Committee considers it desirable that, in statements
drawn up for international comparison, each country should publish
separate data for every item which is of any considerable importance
in its own economic life.
2. Principles of Classification.

{,52} The main principle of grouping adopted by the Committee
for the broad framework of the nomenclature is that of the technical
organisation of the work in the various establishments and branches
of industry. This can best be illustrated by the following scheme:
I.

II.

III.

Primary production:
A. Agriculture, forestry and fishing;
B. Mining and quarrying.
Transforming industries:
A. Manufacturing industries;
B. Building and other construction.
Services:
A. Transport and communication;
B. Commerce;
C. Hotels and personal services;
D. Public services and other services of general
interest.

{53} In the nomenclature itself, the form of presentation . . .
has been selected for reasons of convenience and only the eight
following sections are distinguished separately:
A.
B.
C.
1

Agriculture, forestry and fishing;
Mining and quarrying;
Manufacturing industries;

Not reproduced here; see note 1, p. 65.

INTERNATIONAL STANDARDISATION OP LABOUR STATISTICS

D.
E.
F.
G.
H.

Building and other construction;
Transport and communication;
Commerce;
Hotels and personal services;
Public services and other services of general interest.

\_54~] To these, a ninth section is added for activities inadequately described.
[_55^\ The main principles which can be applied to the grouping
of manufacturing industries are:
(1) The nature of the raw materials used;
(2) The use of the products;
(3) The technique of the work done.

[_56~\ The Committee is therefore of the opinion that . . . the
grouping of manufacturing industries should be based primarily
on a combination of the raw material principle with the technique
of work—this combination indeed forms the basis of the organisation of many establishments. In the case of industries producing
articles ready for consumption to which in many cases the above
principles cannot reasonably be applied, the use of the products
should be taken into consideration.
[.572 In the sections relating to primary production and manufacturing industries, the order of chapters follows as far as possible
the order adopted in the "Minimum List of Commodities for International Trade Statistics".

D. CLASSIFICATION BY PERSONAL STATUS

[582 The Committee recommends the classification set forth below or any other classification in which the following groups are distinguished :
[592 (1) Employers: Persons working on their own account
with paid assistants in their occupation.
[602 (2) Persons working on their own account, either alone
or with the assistance of members of their families.
[6Í2 (3) Members of families aiding the heads of their families
in their occupations.
[622 (4) Persons in receipt of salaries or wages:
(a) Directors and managers;

CLASSIFICATION OP INDUSTRIES AND OCCUPATIONS

(6) Other persons in receipt of salaries or wages, distinguishing, if possible, between :
(Î) Persons in receipt of salaries;
(w) Wage-earners.
[_63] This distinction between the employers in group 1 and
the persons in group 2 makes it necessary, in the case of countries
which do not take a census of establishments simultaneously with
a census of population, for the cards or schedules used for the
latter purpose to contain a special question or note as to whether or
not a person employs paid assistants or only members of his family.
[64] The Committee realises that, generally speaking, there is
little reliable information regarding the separation of group 4 (a),
"Directors and managers", from other persons in receipt of salaries.
As some countries include directors and managers in the group of
"employers", the distinction of group 4 (a) from group 4 (ò) is
important for the compilation of comparable data. It may be
noted in this connection that all State and local officials, irrespective of their rank in the administration, should be grouped under
4(0).
\_65~] The Committee's proposed subdivision of group 4 (b)
into persons in receipt of salaries and wage-earners is only optional.
The Committee does not think it possible to lay down definite
criteria which would enable the distinction to be made uniformly
in all countries. It is aware that this distinction, which is a matter
of some difficulty even in the case of national statistics, cannot,
for a variety of reasons (differences of legislation, of systems of
social insurance, etc.) yield results that are altogether satisfactory
or comparable from the international point of view. But, as difficulties of classification arise only in the case of a relatively limited
proportion of the individuals concerned and as the informative
value of the distinction is considerable, the Committee reached
the conclusion that such an optional distinction should be recommended.
[_66~\ The proposed classification by personal status, which
corresponds in the first instance with the status of persons occupied
in agriculture, mining and manufacturing industries, transport and
commerce, should be applied to all branches of economic activity
in order to permit of the preparation of aggregate figures for the
various groups. Although such aggregates, without subdivisions
by branches of economic activity, have not very much significance
for groups 1, 2 and 4 (a), aggregates for groups 3 and 4 (6) appear
to possess real significance.

INTERNATIONAL STANDARDISATION OF LABOUR STATISTICS
E. CLASSIFICATION BY INDIVIDUAL OCCUPATIONS

[67] [The Committee did not believe it practical or desirable
to prepare a detailed draft nomenclature of occupations for purposes of international comparison, while realising the value and
importance of such a classification in national statistics of the
gainfully-occupied population. For national purposes, the best
solution, in the Committee's opinion, is a classification giving an
enumeration of the important occupations within each branch of
economic activity, supplemented by a restricted list of occupations
that are common to a number of different industries: aggregate
figures for all industries together can then be compiled for the
more important occupations.]
F. CLASSIFICATION BY S E X , A G E AND MARITAL STATUS

[68] The Committee recommends that, for each chapter of
the classification by branches of economic activity, and for each
group of the classification by personal status, the gainfully-occupied
population should be subdivided first of all by sex. The data for
each sex should then be subdivided according to age and marital
status.
[69] The grouping by age should include the following groups:
(a) Under 15 years of age;
(ò) 15-19 years of age;
(c) 20-64 years of age, divided, if possible, into two
sub-groups:
20-44 years of age (preferably subdivided into five-year
or ten-year divisions);
45-64 years of age;
(d) 65 years of age and over.
[70] The Committee is of the opinion that, for the purpose
of international comparisons, it is essential to have total figures
for the two groups (a) and (d) but that details as to branches of
economic activity and personal status, which are of great importance for groups (6) and (c) have less importance for groups (a)
and (d).
[71] If a complete grouping by marital status (unmarried,
married, divorced, widowed) cannot be achieved, at least the
married group should be shown separately.
[72] Group B, 1, of the non-gainfully-occupied population—
i.e. members of families of working age, capable of work and not
attending school—should, in like manner, be subdivided according
to sex and age.

CLASSIFICATION OP INDUSTRIES AND OCCUPATIONS

6. Minimum Nomenclature of Industries recommended by
the Committee of Statistical Experts
of the
League of Nations 1
[73]

MINIMUM NOMENCLATURE OP INDUSTRIES

Section A.—Agriculture,
I.
II.
III.

IV.
V.

Forestry and Fishing

Agriculture and stock-breeding.
Forestry, hunting and trapping.
Fishing.
Section B.—Mining and Quarrying
Mining and quarrying of coal, ores and other minerals,
n.e.s.
Quarrying of stone and other materials used principally for construction.
Section C.—Manufacturing

VI.
VII.
VIII.
IX.
X.
XI.
XII.
XIII.
XIV.
XV.
XVI.
XVII.
XVIII.
XIX.
XX.
XXI.
XXII.
XXIII.

Industries

Manufacture of food.
Beverage industries, distilleries.
Tobacco industries.
Manufacture of animal and vegetable oils and fats,
including margarine.
Chemical industries, n.e.s.
Manufacture of rubber.
Manufacture of wood and cork.
Pulp and paper mills; manufacture of paper articles.
Printing and bookbinding; photography.
Skin, hide and leather industries.
Textile industries.
Manufacture of wearing apparel and of made-up textile goods, n.e.s.
Production and supply of electricity, gas and heat;
water supply.
Manufacture of mineral fuels.
Manufacture of non-metallic minerals, n.e.s.
Smelting, casting, rolling, forging and drawing of
base metals.
Manufacture of articles of base metals, n.e.s.
Manufacture of machinery, including all electrical
appliances and means of transport by land, water
and air.

1
LEAGUE OP NATIONS: op. cit. (see above, p. 52), pp. 21-22. A detailed list
with illustrative examples is also given, pp. 22-32.

INTERNATIONAL STANDARDISATION OF LABOUR STATISTICS

XXIV.
XXV.
XXVI.

Manufacture of instruments, clocks and watches,
and jewellery.
Other manufacturing industries, n.e.s.
Manufacturing industries not adequately described.

Section D.—Building
XXVII.

Building and other constructional work, including
repair and maintenance of structures, n.e.s.

Section E.—Transport
XXVIII.
XXIX.
XXX.
XXXI.
XXXII.

and Other Construction

and

Communications

Rail transport.
Road transport.
Water transport.
Air transport.
Post, telegraph and telephones.
Section F.—Commerce

XXXIII.
XXXIV.
XXXV.

Wholesale and retail trade.
Finance, banks, insurance.
Agencies for other purposes, n.e.s.
Section G.—Hotel and Personal Services

XXXVI.
XXXVII.

Provision of lodging, food and drink.
Undertakings for cleaning and maintenance of goods
and premises, care of the person.
X X X V I I I . Household and personal services.
X X X I X . Recreation and amusement, sport.
Section H.—Public Services and Other Services of General Interest,
N.E.S.
XL.
XLI.
XLII.
XLIII.
XLIV.

Medical and hygienic services.
Education, religion, arts and sciences, n.e.s.
National defence.
Labour camps and similar organisations.
Public services, n.e.s., and other services of general
interest.

Section J.—Activities not Adequately Described
XLV.

Activities not adequately described. 1

1
The Report also gives details of the different industries and items to be
included in each group of industries. It gives also, in many cases, reasons on
the basis of which the Committee based its recommendations. The members
of the Committee are given on page 5.

II
STATISTICS OF EMPLOYMENT AND UNEMPLOYMENT
7. Resolutions of the Second International Conference of
Labour Statisticians, Geneva, April 1925
UNEMPLOYMENT STATISTICS

{74} (1) In countries in which a widespread system of unemployment insurance exists the information obtained from the working of such a system forms the best basis for unemployment
statistics.
{75} (2) These statistics should furnish the following information as a minimum :
(a) Annually, the total number of workpeople insured
against unemployment, which should be related so far as practicable to the total number of workers.
(6) Monthly, the total number of unemployed on a given
day in receipt of benefit, and the total number of insured workers
unemployed (whether on benefit or not) on the same day.
(c) Monthly, the percentage that on a given day the
total number of insured unemployed (whether on benefit or
not) forms of the estimated number of insured workpeople.
(d) Annually, the total amount paid during the year in
benefits.
{76} A method according to which, as in the Netherlands, a
percentage is calculated between the number of days of unemployment during one week and the total number of days of work which
might have been performed is one which is worthy of attention.
{77} (3) Where statistics based on unemployment insurance,
compulsory or voluntary, are not available, it is desirable to obtain
from workers' organisations the following information:
(a) Monthly, the total number of unemployed on a given
day and the percentage they form of the total membership
covered by the enquiry.
(6) Annually, the number of workers covered by the
enquiry as a percentage of the total number of workers in the
corresponding industries or occupations.

INTERNATIONAL STANDARDISATION OP LABOUR STATISTICS

Even when statistics based on unemployment insurance become
available it is desirable, for purposes of comparison, to continue
trade union statistics as long as they are reliable.
{78] (4) The statistics derived from public employment offices
should give:
(a) The number of workpeople registered on a given day
of the month as seeking work, with the total number of vacancies
remaining unfilled on the same day.
(è) The number of workers' applications registered, of
vacancies notified, and of vacancies filled, during the month.
{79] (5) Employment exchange statistics should be compiled
so that as far as possible unskilled workers are distinguished from
other classes of workers.
{80] (6) Information as to the state of employment should
also be published periodically, if possible monthly, based on returns
made by a representative number of employers.
{81] (7) In countries in which satisfactory information concerning unemployment cannot be obtained by the means indicated
above, it is desirable that an attempt should be made to obtain at
the general population census, or at an industrial or occupational
census, information as to the amount of unemployment, or that
special enquiries relating to the whole population or to an adequate
sample thereof should be made from time to time with a view to
ascertaining the number and condition of the unemployed.
{82] (8) The statistics indicated in the preceding resolutions
should distinguish males and females and should, so far as possible,
give figures for separate occupations in the case of statistics arising
out of the activities of employment exchanges, and for separate
occupations or industries as may be most convenient in the case
of trade union and insurance statistics and in the case of special
industrial or occupational enquiries.
The classifications used
should be based on the classifications adopted in the general population census in so far as these are applicable to employment exchange
operations.
{83] (9) It is desirable, for comparing statistics of unemployment internationally:
(a) That precise and detailed information should be published and kept up to date as to methods adopted in compiling
the various statistics of unemployment; in particular, each
country should indicate any legislative or administrative changes
affecting the value of its unemployment statistics;

STATISTICS OP EMPLOYMENT AND UNEMPLOYMENT

(£>) That copies of all forms and questionnaires used in
collecting the various statistics should be forwarded to the
International Labour Office;
(c) That in each country an enquiry should be made to
determine, as exactly as possible, the representative value of
the unemployment statistics in relation to the "ideal statistics"
which would give, at any date, the total number of unemployed
in relation to the total number of workers. For this purpose
it is agreed:
(t)
that the ideal population "field" to which the
statistics should relate should be all persons whose normal
means of livelihood is employment under contract of
service as well as those persons not hitherto wage earners
who seek to become so;
(it) that the unemployment measured should exclude that due to sickness, invalidity, participation in
trade disputes or voluntary absence from work, and
should be limited to unemployment due to lack of employment or to lack of work while in employment;
(Hi) that the necessary and sufficing condition for
being enumerated as unemployed is that the individual
must have been not at work for one day at least.
[_84~\ (10) It is desirable that the different statistics of unemployment (insurance, trade union, and employment exchanges)
should be presented together, preferably in graphic form, so that
they might be co-ordinated and checked one with another and as
clear and correct an idea as possible of the fluctuations in unemployment obtained.
[_85~] (11) Statistics of short-time employment should, if
possible, be given separately from those of whole-time unemployment

IK
STATISTICS OF WAGES AND HOURS OF WORK*
8. Resolutions of the First International Conference of
Labour Statisticians, Geneva, October 1923
[86] Detailed statistics of rates of wages, of actual earnings,
and of normal and actual hours of labour should be collected and
published in each country as frequently as possible, account being
taken of the special circumstances and conditions obtaining in
each case. With a view to facilitating international comparisons,
the responsible authorities in each country should, as far as practicable, observe the following principles:
[87] (1) At regular intervals, and a t least once a year, there
should be published:
(a) Statutory minimum rates';
(ò) Rates fixed in collective agreements;
(c) Rates accepted by organisations of employers and
workpeople for typical categories of workers.
[_88^ (2) In order to provide an indication of the general
course of wage movements, information should be published at
more frequent intervals as to the nature and amount of any changes
resulting from alterations in the statutory minimum rates or arranged between organisations of employers and workpeople. Particulars should be given of changes in the normal hours of labour
and of alterations in the level of piece work rates.
[89] (3) At regular intervals, not less than once a year, average
actual earnings and actual hours of labour during a year or a typical
period in a year should be given for each of the principal industries,
and based on data supplied by representative employers or establishments.
[90] (4) From the data indicated above, index numbers
should be computed to show the general course of changes in nominal
wage rates and in actual earnings. Index numbers of the purchasing
power of the wages should also be calculated by relating changes
in actual earnings to changes in the cost of living, the necessary
precautions being taken to ensure that the two series of data are
comparable.
1

See also V. International Comparison of Real Wages.

STATISTICS OP WAGES AND HOURS OF WORK

The nominal wages employed in computing the index numbers
should be given in every case.
[9f\ (5) At less frequent intervals general wage censuses
should be taken, information being obtained from the pay sheets
of establishments to show rates of wages and the actual earnings
in a typical week. The information should be given by industries,
districts, occupations, and sex, and a distinction should be made
between adults and young persons.
[92} Until the principles enunciated above have been applied
in the different countries, statistics of wages and hours of labour
should at least give:
(a) Current rates of wages (hourly or weekly) and normal
hours of work of typical categories of time workers; and at
regular intervals averages weighted according to the number
of workers to whom the data apply both for such categories
and for all categories combined;
(b) Actual and full-time earnings and hours of labour for
typical categories of workers, especially those paid on piece
work; such statistics should be available for sample periods,
at least once a year;
(c) Real wage index numbers based on nominal wage
and cost-of-living index numbers.
[93] In order that the International Labour Office may make
tentative comparisons of the level of real wages in the different
countries, the competent statistical authorities of each country
should furnish the International Labour Office at regular intervals
(if possible monthly) with statements, in a form to be agreed upon,
showing for the capital cities of their respective countries:
(i) the time-rates of wages and normal weekly hours of
labour current in a limited number of typical occupations; and
(M) information as to the prices of a limited number of
those items upon which the income of working-class families
in most industrially developed countries is largely spent.
9. Resolutions of the First International Conference of
Statisticians convoked by the Social Science
Research Council of the United States of
America, January 1929
INDEX NUMBERS OF W A G E S

[94] The Committee is fully aware that in the majority of
countries the available data are too limited for all the requisite
conditions for the compilation of satisfactory index numbers of

INTERNATIONAL STANDARDISATION OF LABOUR STATISTICS

wages to be fulfilled. However, it considered it desirable in the
circumstances to discuss the conditions to be observed in order
that index numbers of wages should, as far as possible, be adapted
to the purposes for which they may be used. It seemed to the
Committee that such an examination was useful not merely from
the point of view of statistical theory, but also from the practical
point of view, because many countries are at present passing.through
a period in which wage statistics are being reorganised on fresh
bases, and it may be very useful for these countries to have certain
guiding principles for the collection and tabulation of their figures.
Moreover, even if it appears that the programme outlined is incapable of realisation at the moment, it will, nevertheless, have a
certain practical utility because it will make clear the extent of
the reservations to be made when interpreting necessarily incomplete series of wage index numbers, whether existing or to be drawn
up in the future.
[95~\ The Committee unanimously recognised that the compilation of index numbers of wages must proceed along perceptibly
different lines as regards both the data to be taken into account
and the methods of collection and tabulation, according to the
purpose for which the index numbers are to be used.
\_96~] It would consequently serve no useful purpose to attempt
to establish an index number for wages for all purposes; on the
contrary, it is necessary to consider separately the establishment
of different indices corresponding to different purposes.
[_97~] There are three obvious and very distinct purposes for
which index numbers of wages may be required:
(1) to determine one of the elements for measuring fluctuations in the standard of living;
(2) to calculate the remuneration per hour of work;
(3) to calculate the labour cost per unit of production.
[98] Within each of these purposes, the object may be to
determine the fluctuations in wages in time, or to make comparisons
between different districts or different countries. In accordance
with the task entrusted to it, the Committee, generally speaking,
considered chiefly the first of these points of view, that is to say,
the variations in wages in time within a country.
[99] At the same time, international comparisons of the remuneration per hour of work seemed so much more valuable than
comparisons in time in a single country, that the Committee felt
impelled to examine certain specially difficult points raised by this
question.

STATISTICS OP WAGES AND HOURS OP WORK

[1002 For each of the purposes mentioned above, a distinction
must be made according to whether the index is considered as
applying to the working class as a whole, or as an indication of
the variations occurring in different industries or classes of workers,
independently of the influence which the relative importance of
these different industries or classes of workers may have on the
general wage averages. For example, if index numbers of wages
are considered as one of the factors to be taken into account in
estimating the standard of living of the workers, then the aim
may be to measure separately the variations in the standard of
living of skilled, semi-skilled, or unskilled workers; or the variations occurring among the workers as a whole, which may arise
to a considerable extent from the fact that skilled workers may be
at present much more numerous than formerly. The Committee
therefore considered it desirable to keep in mind these two points
of view. A similar distinction may be made, and was in fact taken
into account by the Committee, in considering the labour cost
per unit of production.
[.1012 In the same way a calculation of the remuneration per
hour of labour may be concerned with the remuneration for any
hour, or for an hour of the same nature (normal or supplementary
work, skilled, semi-skilled, or unskilled work) and of the same
efficiency in regard to either individual qualifications or technical
equipment. The Committee, in view of the limited time at its
disposal, restricted its examination to the latter point of view.
[1022 In addition to the three main purposes analysed above,
there are two others for which index numbers of wages have been
employed. There is, first, that of measuring the variations in the
total amount of wages as a percentage of the national income, and
secondly, that of using them as one of the factors in economic
forecasting. The Committee was of opinion that this latter purpose
was particularly important, and therefore gave it special attention.
[1032 It must be understood that when the Committee states
that index numbers of wages may be used for measuring the standard of living or the remuneration for an hour of work, or again
as a factor in economic forecasting, it does not thereby imply that
they represent the only factor to be taken into account for the
purpose in view. It is obvious that the index numbers of wages
can only be one factor, and that other measurements must also
be taken into consideration. In particular, the index number of
the cost of living must be considered when estimating the standard
of living or the real remuneration for an hour of work, and, similarly,
the level of prices when the index numbers of wages are to be used

INTERNATIONAL STANDARDISATION OP LABOUR STATISTICS

in calculating the labour cost per unit of production. In the same
order of ideas the Committee unanimously decided (as the second
Conference of Labour Statisticians had done) that the data concerning actual earnings should be taken for preference, save in
exceptional cases, in examining the question of the standard of
living.
£.104] The Committee was unanimously of the opinion that in
determining the standard of living no attempt should be made
to allow for the subjective cost of labour, which it is quite impossible
to measure; on the contrary, the term "standard of living" should
retain the connotation which it generally has in economic science,
namely, that of material well-being. In order to avoid any misunderstanding the Committee would even have adopted the suggestion of using the term "purchasing power" instead of "standard
of living", if it had not been that it considered it more practical to
retain the commoner expression which is more familiar to the
majority of workers, rather than employ a technical term whose
exact significance is probably fully grasped only by specialists.
£105] The Committee did not claim to deal with all the important points, but was content to consider for each type of index :
(1) the wages data (rate of wages, average wages, or actual
earnings) to be taken into account in each case;
(2) the industries or classes of workers to which these data
should refer;
(3) the areas from which returns should be collected;
(4) the period during which the data should be gathered ;
(5) the method of arriving at the index numbers, that is,
the weighting of the various classes or industries, the choice
of the basic period, etc.
£106"] Generally speaking, the Committee wishes to emphasise
the importance, whenever a series of index numbers of wages is
published, of giving a detailed explanation of the principles adopted
with reference to each of the five points enumerated above, and of all
the other factors which may be essential for a true interpretation
of the index numbers.
Q/073 Finally, while the Committee, in accordance with the task
entrusted to it, considered in particular the fluctuations of wages
within single countries, it also frequently emphasised the fact
that comparisons between different districts and different countries
were no less important, and might even be of greater value.

STATISTICS OF WAGES AND HOURS OP WORK

I.—Construction of Indices to Measure Changes in the Standard
of Living of the Working Class as a Whole
(1)

Choice of Type of Wage Data.

Hl08~} If the standard of living is used in its generally accepted
sense of material well-being, it is evident that for the purpose of
constructing wage index numbers to measure changes in the standard of living of the working class as a whole, the ideal type of data
to be used should be actual annual earnings, or earnings for a
shorter period representative of annual earnings.
[iOP] Experience in certain countries shows that in fact changes
in actual earnings may differ from changes in rates of wages and in
average wages per hour. But this does not exclude the fact that in
other countries changes in rates of wages may be considered as
approximately representative of changes in actual earnings in
certain industries and under certain conditions.
[_110~\ While recognising that, in normal conditions, the standard of living may be considered proportional to actual earnings,
it must be recognised that changes in hours worked and other
social conditions equally affect the standard of living, and the
accuracy of an index number based on actual earnings. In these
cases, it would be desirable to take hours of labour into consideration.
(2)

Industries and Categories of Workers to be Included.

\_llf\ For the construction of such index numbers it would be
desirable to consider changes in the wages of the wage-earning
class as a whole; but as in practice this would be rarely possible,
it is necessary to select certain industries and categories of workers
which would be representative of the wage-earning class. In this
case it will be necessary to indicate clearly the industries and categories of workers covered by the statistics.
[112} If it is desired to measure the standard of living of the
wage-earning class as a whole, it is indispensable to include agricultural workers, as well as workers in industry proper. But as
in most countries statistics of the cost of living of agricultural
workers are not available, it will be necessary, when one wishes to
compare index numbers of wages with index numbers of the cost
of living, to consider separately the workers in industry proper.
(3)

Choice of Districts.

\_113} Where it would not be possible to cover the whole country, districts should be chosen so as to be representative of the

INTERNATIONAL STANDARDISATION OP LABOUR STATISTICS

whole country. If adequate consideration is not given to this
very important point, there is the danger that wage index numbers
will be compiled which are not representative of the actual situation, either from the point of view of the levels of wages of the
wage-earning class, or from the point of view of their changes.
(4)

Frequency of Compilation.

{114] From the point of view of the practical use of these
statistics, it would appear reasonable that the frequency of compilation should be determined in relation to the stability of economic
conditions, compilation being less frequent when these conditions
are more stable.
l_115~\ From the point of view of facility in collecting the statistics and for the purpose of ensuring their accuracy, there may be
certain reasons, as is shown by the experience of a number of countries, for the compilation of the statistics at frequent intervals, even
when economic conditions are stable.
\_116~\ In any case, it is opportune to determine frequency of
compilation in relation to the systems of wage payment in use in
the different countries, selecting periods for which the wage payments may be considered complete. Generally a period of one
month (or four weeks) would appear to be the most suitable, and
to be satisfactory in relation to other general uses of the wage
index numbers. Where the supply of information depends on the
good will of the employers, it is necessary to take care that the
statistics for successive periods relate to the same establishments.
If this condition cannot be realised, it would become necessary to
obtain also statistics for the preceding period in order to enable the
figures for successive periods to be connected by the chain method.
[117~\ When a system of payment includes supplementary
items of remuneration that cover more than one period of normal
payment, it is necessary to allocate such items proportionately
among the normal periods that they cover or to average them over
the whole period intervening between one such supplementary
payment and the next.
(5)

Method of Compilation.

[_118} As to the method used in establishing these index numbers, the most precise consists in relating the total actual earnings
compiled or calculated for a given period to the corresponding
actual earnings for the period taken as base. These averages may
be compiled per worker, or better, per member of the family, if it is

STATISTICS OP WAGES AND HOURS OP WORK

possible to include the earnings of other members of the family.1
The last consideration, which may be important in international
comparisons, does not, however, have any appreciable effect on
time comparisons in a country.
\_119] The calculation of a simple or a weighted average index
number for the various industries and occupations, the weights
remaining constant during the various periods, may lead in certain
circumstances to false conclusions, for they do not take into account
the movement of workers from one industry to another and from one
occupation to another. These movements may in certain cases be
of great importance. In consequence, it may be necessary to use
separate indices for the different industries, and in combining them
to revise the weighting as often as possible so as to take account
of these changes. For such a series of index numbers, the choice
of base period does not influence the index number.
[_120~\ It would be desirable, however, that all countries should
adopt a common base period and for various reasons a post-war
period is favoured. This does not exclude, naturally, the possibility
for certain countries of constructing a series with pre-war bases.
The fact that in various countries statistics of wages are in their
infancy or in course of being organised, and also the fact that in
various countries economic conditions are not yet stable enough,
hinders the fixing at present of the most appropriate base period
common to all countries. In a general way it would be preferable
not to choose the first period for which the enquiries are made, as in
general the first enquiries are not the most exact.
II.—Construction of Index Numbers of Wages for Measuring, in each
Country, the Changes in the Fluctuation in the Standard of
Living in Different Industries or Occupations
\_12f\ It is of interest for certain purposes to measure fluctuations in the standard of living, not of the workers as a whole, but in
different industries or occupations, without taking into account
fluctuations in the standard of living of the workers as a whole,
which may arise from variations in the numerical importance of
different industries or occupations. In this case, the same methods
are to be recommended as to the choice of wage data, of localities,
and of periods.
1
These averages will usually be compiled per earner, b u t when it is possible
to obtain the information they may also be compiled per family. In this case
the average size of the family should be given, expressed, if possible, in terms
of "adult men".

INTERNATIONAL STANDARDISATION OF LABOUR STATISTICS

III.—Calculation of Index Numbers of Wages as an Aid in Forecasting Economic Conditions
(1)

Choice of Data.

\_122~] In countries which do not possess well-organised statistics
of unemployment, overtime, and short time, or in countries where
the wage rates are fixed for a determined and fairly lengthy period,
the statistics of earnings have advantages for following economic
fluctuations.
In countries for which the statistics of unemployment, etc., are well organised, and where rates of wages can be
changed according to economic conditions, it may be useful to consider rates of wages also as an index for economic forecasting, as,
at the beginning of a crisis, these rates may vary according to the
orders received by establishments, and consequently may constitute a more sensitive index than that of actual earnings.
(2)

Industries and Categories of Workers.

Q/2J] In every country it is necessary to take into consideration
the industries and categories of workers whose variations in wage
rates or actual earnings are the most sensitive ; the work of choosing
the industries and categories the most appropriate for these studies
must be left to the specialists of each country. Certain industries
even of less economic importance, like the luxury trades and printing, may give interesting information in this field, whilst industries
which receive orders long periods in advance, and may in certain
cases be the most important ones from the economic point of view,
may be less suitable for forecasting.
^_124\ In each branch of industry the best organised establishments which can estimate with the greatest certitude the future
development of the market should primarily be considered.
(3)

Choice of Localities.

\_125~^ In order to establish economic forecasts, it is advisable
to take into special consideration certain centres or localities which,
either because of the concentration of industry, or because of some
special development, or for other reasons, are, as experience has
shown, the first to show symptoms of crisis or of economic revival.
(4)

Frequency.
\J26~\ The interval chosen should be as short as possible.

(5)

Method of Compilation.

\_127~\ It would be advisable, even if it were desired to reduce
the various information into one index, to publish separately the

STATISTICS OP WAGES AND HOURS OP WORK

information for the various industries, occupations, and localities,
so that all necessary information would be available for the interpretation of these indices.
IV.—Compilation of Index Numbers of Wages to Measure Earnings
per Hour in respect of Work of Comparable
Character and Efficiency
(1)

Choice of Data.

[128} It may be assumed that the best method theoretically
would be to determine separately the average wages per hour of
normal work and the average wages per hour of overtime, and then
to eliminate the influence of varying amounts of overtime by combining these two averages in a constant ratio.
[129} As in many countries it is not possible in practice to
distinguish between normal and overtime earnings, the most practical method is to calculate a single average by dividing the aggregate amount of actual earnings — for normal and overtime work,
including special allowances, payments in kind, etc.—by the total
number of hours worked. The objection to this method of calculation is that changes in the payment per hour worked may result
from a mere change in the proportion of overtime worked to normal
hours; this objection being the more important, the more significant
the changes in this proportion.
(2)

Industries and Categories of Workers.

\_130~] The comparison aimed at is one between certain categories of workers belonging to selected occupations who work
with approximately similar equipment and under similar conditions
in different countries or in different periods. It must be recognised,
however, that the individual workers who perform the same occupations in different countries and in various periods may diverge in
varying degrees from the average worker in respect to individual
skill and efficiency. This objection will be the more important in
practice, the more restricted the number of occupations and the
number of workers considered.
[131] Even if the comparisons between different countries
may on this account be open to doubt, it may well be that the
variations in time shown by these index numbers may be comparable.
(3)

Localities.

[132'] It may be assumed that the localities should be representative of the conditions of the whole country. Consequently it may

INTERNATIONAL STANDARDISATION OP LABOUR STATISTICS

not be sufficient to limit the statistics to capital or principal cities,
because the differences, in this respect, between these and the rest
of the country may be more or less great according to the country
or the period considered.
(4)

Frequency.

\il33^\ If it were possible to determine separately the average
wages per normal hour and per hour of overtime, it would be desirable to secure data as frequently as possible. Since in general such
data are not available, it will be necessary to take rather long
intervals so that the different circumstances influencing the amount
of overtime work might arrange themselves out.
(5)

Method of Calculation.

H134] It goes without saying that the different occupations
chosen for international comparisons will comprise different numbers of workers in different countries, and that these differences
will not be similar in all countries and will even present considerable
variations from country to country.
\^135~J¡ Since the index numbers are to be used to measure the
changes in the earnings per hour of work of a given nature and
given efficiency, it will be necessary to calculate as many special
index numbers as there are from this point of view different categories, and then to compute an average of these index numbers.
\_136~} It is evident that the index numbers calculated for different categories of workers should be weighted according to the number of workers in each category. A practical difficulty in comparing
the index numbers in different countries arises from the fact that
the numbers of workers of each category would vary from country
to country, and that there is no reason for taking as basis the number of workers in one country rather than that in another country.
In order to avoid this difficulty, it might be advisable to repeat the
calculation as many times as there are countries, using in turn the
weight corresponding to the conditions prevailing in each country.
\_137~} The results of the calculation will show whether the
application of the different methods of weighting will produce
significant deviations.
V.—Compilation of Index Numbers of Wages for the Purpose
of Measuring the Labour Cost per Unit of Production
\^1382 If the unit of production is taken in the sense of a unit of
value, the determination of labour cost per unit of production
consists in calculating the percentage of the total value of produc-

STATISTICS OF WAGES AND HOURS OF WORK

tion represented by the aggregate wages paid. It is evident that
even if in some countries and in certain branches of productive
activity firms already furnish data of this kind, it is not possible to
expect at present an extension of these enquiries over all branches
of economic activity and all countries. In this respect the Committee, while hoping that this will be possible in respect to an increasing part of production and increasing number of countries,
considers that the collection of data on wages, cost of labour, and
cost of production should be organised as far as possible so as not
to give rise to fears or opposition on the part of the firms concerned.
Even if this opposition to supplying such detailed information is
sometimes due rather to habits or traditional ideas which are no
longer justifiable, it is necessary in any case to take account of them
because the sincere collaboration of the heads of firms constitutes
an exceedingly useful element in ensuring the accuracy of the data.
\_139~\ In any case the possibility may be considered of measuring the variations, between different periods or different localities,
in the labour cost as an element in the total cost of production.
There are, in fact, several countries which conduct censuses of the
value of production for the whole country or for industry as a
whole, or for industrial production, or for manufacturing, or for
large industries, or even for agricultural production. For these
countries it would be possible to establish index numbers showing
the variations in the value of the production as between different
periods. If it should be possible to establish, in addition to this
series of index numbers, another series of index numbers relating
to aggregate wages paid to the workers covered, it would be possible
to state whether the value of production has increased more or less
rapidly than the aggregate wages bill, and consequently whether
the percentage of wages in the total cost of production has increased
or decreased during this period.
[_14Cß Similarly, it is possible to proceed to comparisons between
different countries, provided that the data concerning the production of these countries are comparable between themselves; it is,
however, difficult to satisfy this condition.
\^14f\ It is evident that this method may be applied either to
production as a whole or to certain special branches of production,
or further to a particular industry. It is obvious that care should
be taken in interpreting the variations in the percentages showing
changes between different periods as well as between different
countries, because these variations or differences may be due to
extremely different circumstances, such as differences in machinery
and equipment, in natural conditions, in the efficiency of labour,
etc.

INTERNATIONAL STANDARDISATION OF LABOUR STATISTICS

[_142~] In compiling these index numbers, it is convenient to
proceed as follows:
\_143~] (1) To take as bases the actual earnings—that is to say,
the aggregate and not the average earnings.
[_144~\ (2) and (3) If it is not possible to take into account all
localities and industries concerned, to choose representative localities and industries.
[145~\ (4) To obtain records of wages for the same periods as
for the censuses or evaluations of production, which can only be
effected at rather long intervals.
[,1462 (5) To adopt for the index numbers of wages the same
base as for the index numbers of production, and to follow in respect
of weighting the principles specified under I and II, according as it
is proposed to make, or not to make, allowance for the effects of
differences i a the importance of the production of different industries.
VI.—Compilation of Index Numbers of Wages for
Measuring the Variations in the Proportion of
the National Income Formed by Wages
[_147~] The percentage of the value of production represented
by the aggregate wages of labour may have a very different significance according as it applies to primary production (extractive
industries, forestry or agriculture) in which the consumption of the
products of other industries is of small importance, or to manufacturing industries in which an important part of the value of the
product is due to raw materials or semi-manufactured products.
This inconvenience might be avoided if the total wages are related
not to the total value of production but to the value of the net
product of industry.
[_148~\ The relation between the aggregate earnings of labour
as a whole and the national income is of particular interest from
theoretical and practical points of view.
\_149\ As regards the choice of data concerning wages, industries, localities, the frequency of collection, and the methods of
calculating the index numbers, the principles indicated under V
may in general be applied.
\JL50\ However, it is to be observed that the comparisons between different countries may be in practice more frequent and less
difficult in this case. In fact, many countries in which comparable
data concerning the value of production are not available calculate
approximately the amount of national income.
\_15f] Per contra, the comparisons as between different periods
would probably be less frequent in the present case because the

STATISTICS OP WAGES AND HOURS OP WORK

variations in the income can be evaluated with sufficient accuracy
only at fairly long intervals.

10. Resolutions of the Second International Conference of
Statisticians convoked by the Social Science Research
Council of the United States of America, May 1930
INDEX NUMBERS OF W A G E S

\_152~] 1. The construction of index numbers to measure the
movement of wages, although designed primarily for use within a
country, is of importance also for the international comparison of
wages. Any divergence between the movement in two countries
is the reflection of a significant change, and may be a cause of
further change in the economic relations of the two countries. It is
important, therefore, that indices of wage movements in different
countries should be made available for purposes of comparison,
and the Committee welcomes the initiative of the International
Labour Office since the last Conference in publishing together all
the available indices of wage movements.
{_153} 2. The Committee are of opinion that the indices
published should take three forms.
{_154} (a) It is desirable, whether a single numerical index is
compiled or not, that as much as is practicable of the detailed
information on which such numerical indices may be based should
be published. The problems in the elucidation of which wage statistics are required are so numerous and varied that it is not possible
to meet all the needs that may arise. It is, therefore, important
that such material as exists should be made available for the use
of students, and not merely numerical indices based on this material. Full publication of details is necessary also to prevent misunderstanding and misuse of indices.
[_155~] (6) A general index, inclusive of as much of the available
material for each country as can be obtained regularly, should be
constructed, where the material is adequate. The publication of
such an index, however, would be dangerous unless accompanied
by an indication of the movement of wages in the chief component
industrial groups. Such more detailed indices should, where possible, give separately the movement of wages of skilled or unskilled
and male and female workers.
\_156~] (c) A special index for each country embodying the
movement of wages in certain selected industries for the special
purpose of international comparison. This index might be modelled
on the index set forth in the American report 1 , combining the
1

Report submitted to the Conference.

INTERNATIONAL STANDARDISATION OP LABOUR STATISTICS

indices for general manufacturing, railways, building and coal
mining. Representative rates for these groups can probably be
found among the wage statistics of most countries1 ; but whatever
combination is included, it should be restricted to industries in
which wages are determined by purely industrial conditions, industries in which wages are influenced by political considerations being
excluded.
[.757] 3. The Committee would call attention to certain desiderata in the presentation of statistics. Where possible actual
figures of wage rates or of averages of wage rates should be given,
as well as the index or indices of change in wage rates, since it is
possible for actual wage rates to remain unchanged, while the index
number in which they are combined changes on account of some
alteration in the proportion of different classes of workers, or the
rise to importance of some hitherto unimportant industry.
[ i 5 # ] Tables should be accompanied by notes warning the
reader of the limits within which and purposes for which the figures
may be used. References to the official publications of the different
countries from which the figures are taken or the indices compiled
should always be given.
[\í5P] The Committee wishes to stress the Recommendation
of the 1929 Conference that statistics of both rates of wages and
earnings should be compiled. Pending collection of such comprehensive statistics, they recommend that rates and earnings should
be given wherever they are available; that it should be made clear
whether the figure given represents wages or earnings where only
one is available'; and that where rates only are embodied in the
index, the relation between earnings and rates, in the years for
which figures of earnings are available, should be given. They suggest also that it may be advisable to omit from any index based
exclusively upon rates, occupations in which it is known that the
divergence between earnings and rates is considerable.
[160] Where possible the data on wages should be compiled
and published monthly, as is already done in certain countries.
This is necessary to make possible the study of seasonal fluctuations.
[J6Í] Both hourly and weekly rates and earnings should be
given when possible.
[_162~\ In any industry or occupation in which a large proportion of women is employed, it is desirable that a separate index
number of women's wages should be compiled. Where only a small
proportion of women are employed, this would be impracticable;
but in this case it should be stated whether the women's rates are
included or excluded from the index for the industry or occupation

STATISTICS OP WAGES AND HOURS OF WORK

which is given. Statistical authorities should print full and repeated
explanations of the principles on which, and the methods by which,
any indices which they publish are constructed.

11. Convention concerning Statistics of Wages and Hours
of Work adopted by the International Labour
Conference at its 24th Session, June 1938
Draft Convention [No. 63] concerning statistics of wages and hours of
work in the principal mining and manufacturing industries, including building and construction, and in agriculture.
[/tfj] The General Conference of the International Labour
Organisation,
Having been convened at Geneva by the Governing Body
of the International Labour Office, and having met in its
Twenty-fourth Session on 2 June 1938, and
Having decided upon the adoption of certain proposals
with regard to statistics of wages and hours of work in the
principal mining and manufacturing industries including building and construction, and in agriculture, which is the sixth item
on the agenda of the Session, and
Having determined that these proposals shall take the
form of a Draft International Convention, and
Having determined that, although it is desirable that all
Members of the Organisation should compile statistics of average
earnings and of hours actually worked which comply with the
requirements of Part II of this Convention, it is nevertheless
expedient that the Convention should be open to ratification
by Members which are not in a position to comply with the
requirements of that Part,
adopts, this twentieth day of June of the year one thousand nine
hundred and thirty-eight the following Draft Convention which
may be cited as the Convention concerning Statistics of Wages and
Hours of Work, 1938:
PART I.—GENERAL PROVISIONS.

Article 1.
\_164~] Each Member of the International Labour Organisation
which ratifies this Convention undertakes that:
(a) it will compile as required by this Convention statistics
relating to wages and hours of work;

INTERNATIONA!, STANDARDISATION OP LABOUR STATISTICS

(6) it will publish the data compiled in pursuance of this
Convention as promptly as possible and will endeavour to
publish data collected at quarterly or more frequent intervals
during the succeeding quarter and to publish data collected
at intervals of six or twelve months during the succeeding six
or twelve months respectively; and
(c) it will communicate the data compiled in pursuance
of this Convention to the International Labour Office at the
earliest possible date.
Article 2.
\_165~\ 1. Any Member which ratifies this Convention may, by
a declaration appended to its ratification, exclude from its acceptance of the Convention :
(a)
(6)
(c)

any one of Parts II, III, or IV; or
Parts II and IV; or
Parts III and IV.

2. Any Member which has made such a declaration may at
any time cancel that declaration by a subsequent declaration.
3. Every Member for which a declaration made under paragraph 1 of this Article is in force shall indicate each year in its
annual report upon the application of this Convention the extent
to which any progress has been made with a view to the application of the Part or Parts of the Convention excluded from its
acceptance.
Article 3.
[.166"] Nothing in this Convention imposes any obligation to
publish or to reveal particulars which would result in the disclosure
of information relating to any individual undertaking or establishment.
Article 4.
\_167~\ 1. Each Member which ratifies this Convention undertakes that its competent statistical authority shall, unless it has
already obtained the information in some other way, make enquiries
relating either to all, or to a representative part, of the wage earners
concerned, in order to obtain the information required for the
purpose of the statistics which it has undertaken to compile in
accordance with this Convention.
2. Nothing in this Convention shall be interpreted as requiring

STATISTICS OF WAGES AND HOURS OF WORK

any Member to compile statistics in cases in which, after enquiries
made in the manner required by paragraph 1 of this Article, it
is found impracticable to obtain the necessary information without
the exercise of compulsory powers.
PART II.'—STATISTICS OF AVERAGE EARNINGS AND OF HOURS
ACTUALLY WORKED IN MINING AND MANUFACTURING
INDUSTRIES.

Article 5.
\_168^\ 1. Statistics of average earnings and of hours actually
worked shall be compiled for wage earners employed in each of the
principal mining and manufacturing industries includirig building
and construction.
\_169] 2. The statistics of average earnings and of hours actually worked shall be compiled on the basis of data relating either to
all establishments and wage earners or to a representative sample
of establishments and wage earners.
[_170] 3. The statistics of average earnings and of hours actually worked shall:
(a) give separate figures for each of the principal industries;
and
(ò) indicate briefly the scope of the industries or branches
of industry for which figures are given.
Article 6.
\171~\ The statistics of average earnings shall include:
(a) all cash payments and bonuses received from the
employer by the persons employed ;
(6) contributions such as social insurance contributions
payable by the employed persons and deducted by the employer ;
and
(c) taxes payable by the employed persons to a public
authority and deducted by the employer.
Article 7.
\_172~] In the case of countries and industries in which allowances in kind, for example in the form of free or cheap housing, food
or fuel, form a substantial part of the total remuneration of the
wage earners employed, the statistics of average earnings shall
be supplemented by particulars of such allowances, together with
estimates, so far as practicable, of their money value.

INTERNATIONAL STANDARDISATION OP LABOUR STATISTICS

Article 8.
[_173] The statistics of average earnings shall be supplemented,
so far as practicable, by indications as to the average amount of
any family allowances per person employed in the period to which
the statistics relate.
Article 9.
\_174\ 1. The statistics of average earnings shall relate to average earnings per hour, day, week or other customary period.
[ i 7 5 ] 2. Where the statistics of average earnings relate to
average earnings per day, week or other customary period, the
statistics of actual hours shall relate to the same period.
Article 10.
\il76~] 1. The statistics of average earnings and of hours actually worked, referred to in Article 9, shall be compiled once every
year and where possible at shorter intervals.
[_177~\ 2. Once every three years and where possible at shorter
intervals the statistics of average earnings and, so far as practicable, the statistics of hours actually worked shall be supplemented
by separate figures for each sex and for adults and juveniles; provided that it shall not be necessary to compile these separate figures
in the case of industries in which all but .an insignificant number
of the wage earners belong to the same sex or age group, or to
compile the separate figures of hours actually worked for males and
females, or for adults and juveniles, in the case of industries in
which the normal hours of work do not vary by sex or age.
Article 11.
[_178] Where the statistics of average earnings and of hours
actually worked relate not to the whole country but to certain districts, towns or industrial centres, these districts, towns or centres
shall, so far as practicable, be indicated.
Article 12.
\_179~\ 1. Index numbers showing the general movement of
earnings per hour and where possible per day, week or other customary period shall be compiled at as frequent and as regular
intervals as possible on the basis of the statistics compiled in pursuance of this Part of this Convention.

STATISTICS OF WAGES AND HOURS OP WORK

[_180~] 2. In compiling such index numbers due account shall be
taken, inter alia, of the relative importance of the different industries.
[I81] 3. In publishing such index numbers indications shall be
given as to the methods employed in their construction.
PART III.—STATISTICS OF TIME RATES OP WAGES AND OF NORMAL
HOURS OP WORK IN MINING AND MANUFACTURING INDUSTRIES.

Article 13.
[_182] Statistics of time rates of wages and of normal hours of
work of wage earners shall be compiled for a representative selection of the principal mining and manufacturing industries, including building and construction.
Article 14.
[1832 1- The statistics of time rates of wages and of normal
hours of work shall show the rates and hours:
(a) fixed by or in pursuance of laws or regulations, collective
agreements or arbitral awards;
(ò) ascertained from organisations of employers and
workers, from joint bodies, or from other appropriate sources
of information, in cases where rates and hours are not fixed
by or in pursuance of laws or regulations, collective agreements
or arbitral awards.
[184] 2. The statistics of time rates of wages and of normal
hours of work shall indicate the nature and source of the information from which they have been compiled and whether it relates to
rates or hours fixed by or in pursuance of laws or regulations,
collective agreements or arbitral awards, or to rates or hours fixed
by arrangements between employers and wage earners individually.
[185] 3. When rates of wages are described as minimum (other
than statutory minimum) rates, standard rates, typical rates, or
prevailing rates, or by similar terms, the terms used shall be explained.
[1862 4. "Normal hours of work", where not fixed by
or in pursuance of laws or regulations, collective agreements or
arbitral awards, shall be taken as meaning the number of hours,
per day, week or other period, in excess of which any time worked
is remunerated at overtime rates or forms an exception to the
rules or custom of the establishment relating to the classes of wage
earners concerned.

INTERNATIONAL STANDARDISATION OP LABOUR STATISTICS

Article 15.
\_187~\ 1. The statistics of time rates of wages and of normal
hours of work shall give1:
(a) at intervals of not more than three years, separate
figures for the principal occupations in a wide and representative
selection of the different industries; and
(6) at least once a year, and if possible at shorter intervals,
separate figures for the main occupations in the most important
of these industries.
£1882 2- The data relating to time rates of wages and of normal
hours of work shall be presented, so far as practicable, on the
basis of the same occupational classification.
\_189~\ 3. Where the sources of information from which the
statistics are compiled do not indicate the separate occupations to
which the rates or hours apply, but fix varying rates of wages or
hours of work for other categories of workers (such as skilled workers, semi-skilled workers and unskilled workers) or fix normal
hours of work by classes of undertakings or branches of undertakings, the separate figures shall be given according to these
distinctions.
\_190"} 4. Where the categories of workers for which figures are
given are not separate occupations, the scope of each category
shall, in so far as the necessary particulars are given in the sources
of information from which the statistics are compiled, be indicated.
Article 16.
{.19Q Where the statistics of time rates do not give the rates
per hour but give rates per day, week, or other customary period:
(a) the statistics of normal hours of work shall relate to
the same period; and
(6) the Member shall communicate to the International
Labour Office any information appropriate for the purpose of
calculating the rates per hour.
Article 17.
[.192"] Where the sources of information from which the statistics are compiled give separate particulars classified by sex and
age, the statistics of time rates of wages and of normal hours of
work shall give separate figures for each sex and for adults and
juveniles.

STATISTICS OP WAGES AND HOURS OF WORK

Article 18.
{193} Where the statistics of time rates of wages and of normal
hours of work relate not to the whole country but to certain districts, towns or industrial centres, these districts, towns or centres
shall, so far as practicable, be indicated.
Article 19.
{194} Where the sources of information from which the statistics of time rates and of normal hours of work are compiled contain
such particulars, the statistics shall at intervals not exceeding
three years indicate:
(a) the scale of any payment for holidays;
(b) the scale of any family allowances';
(c) the rates or percentage additions to normal rates paid
for overtime; and
(d) the amount of overtime permitted.
Article 20.
{.195} In the case of countries and industries in which allowances in kind, for example in the form of free and cheap housing,
food or fuel, form a substantial part of the total remuneration of
the wage earners employed, the statistics of time rates of wages
shall be supplemented by particulars of such allowances, together
with estimates, so far as practicable, of their money value.
Article 21.
{196} 1. Annual index numbers showing the general movement
of rates of wages per hour or per week shall be compiled on the
basis of the statistics compiled in pursuance of this Part of this
Convention, supplemented, where necessary, by any other relevant
information which may be available (for example, particulars as
to changes in piece-work rates of wages).
{197} 2. Where only an index number of rates of wages per
hour or only an index number of rates of wages per week is compiled, there shall be compiled an index number of changes in normal
hours of work constructed on the same basis.
{198} 3. In compiling such index numbers due account shall
be taken, inter alia, of the relative importance of the different
industries.
{199} 4. In publishing such index numbers indications shall
be given as to the methods employed in their construction.

INTERNATIONAL STANDARDISATION OF LABOUR STATISTICS
PART IV.—STATISTICS OP WAGES AND HOURS
OF WORK IN AGRICULTURE.

Article 22.
{200} 1. Statistics of wages shall be compiled in respect of
wage earners engaged in agriculture.
[20f\ 2. The statistics of wages in agriculture shall:
(a)
(6)

be compiled at intervals not exceeding two years;
give separate figures for each of the principal districts ;

and
(c) indicate the nature of the allowances in kind (including housing), if any, by which money wages are supplemented,
and, if possible, an estimate of the money value of such allowances.
[2023 3. The statistics of wages in agriculture shall be supplemented by indications as to:
(a) the categories of agricultural wage earners to which the
statistics relate;
(6) the nature and source of the information from which
they have been compiled;
(c) the methods employed in their compilation; and
(d) so far as practicable, the normal hours of work of the
wage earners concerned.
PART V.—MISCELLANEOUS PROVISIONS.

Article 23.
[_203~] 1. Any Member the territory of which includes large
areas in respect of which, by reason of the difficulty of creating the
necessary administrative organisation and the sparseness of the
population or the stage of economic development of the area, it
is impracticable to compile statistics complying with the requirements of this Convention may exclude such areas from the application of this Convention in whole or in part.
£204] 2. Each Member shall indicate in its first annual report
upon the application of this Convention submitted under Article
22 of the Constitution of the International Labour Organisation
any areas in respect of which it proposes to have recourse to the
provisions of this Article and no Member shall, after the date
of its first annual report, have recourse to the provisions of this
Article except in respect of areas so indicated.
\_205~] 3. Each Member having recourse to the provisions of

STATISTICS OF WAGES AND HOURS OF WORK

the present Article shall indicate in subsequent annual reports
any areas in respect of which it renounces the right to have recourse
to the provisions of this Article.
Article 24.
[206] 1. The Governing Body of the International Labour
Office may, after taking such technical advice as it may deem
appropriate, communicate to the Members of the Organisation
proposals for improving and amplifying the statistics compiled in
pursuance of this Convention or for promoting their comparability.
[207"] 2. Each Member ratifying this Convention undertakes
that it will:
(a) submit for the consideration of its competent statistical
authority any such proposals communicated to it by the Governing Body;
(ò) indicate in its annual report upon the application of
the Convention the extent to which it has given effect to such
proposals.
PART V I . — F I N A L PROVISIONS.

Article 25.
[208"] The formal ratifications of this Convention shall be
communicated to the Secretary-General of the League of Nations
for registration.
Article 26.
[209] 1. This Convention shall be binding only upon Members
of the International Labour Organisation whose ratifications have
been registered with the Secretary-General.
[210] 2. It shall come into force twelve months after the date
on which the ratifications of two Members have been registered
with the Secretary-General.
[211] 3. Thereafter, this Convention shall come into force
for any Member twelve months after the date on which its ratification has been registered.
Article 27.
[212] As soon as the ratifications of two Members of the International Labour Organisation have been registered, the SecretaryGeneral of the League of Nations shall so notify all the Members
of the International Labour Organisation. He shall likewise notify
them of the registration of ratifications which may be communicated
subsequently by other Members of the Organisation.

INTERNATIONAL STANDARDISATION OP LABOUR STATISTICS

Article 28.
[.213] 1. A Member which has ratified this Convention may
denounce it, after the expiration of ten years from the date on
which the Convention first comes into force, by an act communicated to the Secretary-General of the League of Nations for registration. Such denunciation shall not take effect until one year after
the date on which it is registered.
[214] 2. Each Member, which has ratified this Convention
and which does not, within the year following the expiration of
the period of ten years mentioned in the preceding paragraph,
exercise the right of denunciation provided for in this Article,
will be bound for another period of ten years and, thereafter, may
denounce this Convention at the expiration of each period of ten
years under the terms provided for in this Article.
Article 29.
[215] At the expiration of each period of ten years after the
coming into force of this Convention, the Governing Body of the
International Labour Office shall present to the General Conference a report on the working of this Convention and shall consider the desirability of placing on the agenda of the Conference
the question of its revision in whole or in part.
Article 30.
[216] 1. Should the Conference adopt a new Convention
revising this Convention in whole or in part, then, unless the new
Convention otherwise provides:
(a) the ratification by a Member of the new revising Convention shall ipso jure involve the immediate denunciation
of this Convention, notwithstanding the provisions of Article
28 above, if and when the new revising Convention shall have
come into force;
(6) as from the date when the new revising Convention
comes into force this Convention shall cease to be open to
ratification by the Members.
[217] 2. This Convention shall in any case remain in force
in its actual form and content for those Members which have
ratified it but have not ratified the revising Convention.
Article 31.
[218] The French and English texts of this Convention shall
both be authentic.

STATISTICS OF WAGES AND HOURS OF WORK

12. Proposed Recommendation suggested by the Fifth
International Conference of Labour 1Statisticians,
Geneva, October 1937
\_219~] Whereas the Convention on Statistics of Wages and
Hours is necessarily limited in scope to statistics which are sufficiently developed to permit of a substantial number of Members
entering into definite obligations in respect thereof; and
Whereas it is of special importance that every effort should be
made further to improve statistics of wages and hours of work;
The Conference, with a view to completing the provisions of
the Statistics of Wages and Hours Convention, recommends that
each Member should give effect to the following provisions:
[220] 1. (1) Statistics of time rates of wages and of normal
hours of work and/or statistics of average earnings and of hours
actually worked should be compiled in respect of wage-earners
engaged in transport, commercial or trading establishments, and
administrative services.
[221] (2) Statistics of salaries and hours should be compiled
for salaried employees, excluding persons holding positions of
management.
[222] (3) These statistics should be compiled as far as possible
at the intervals specified in the Statistics of Wages and Hours
Convention in respect of statistics relating to persons to whom
that Convention applies.
[223] 2. (1) At least every ten years, statistics of average
earnings and of hours actually worked should be compiled for each
of the principal occupations in each industry.
[224] (2) These statistics1:
(a) should show the number of workpeople grouped according to the amount of their earnings; and
(6) should be given by sex and by age groups.
[225] 3. (1) Statistics showing the aggregate amount of wages
and salaries paid per annum in each of the principal industries in
mining and manufacture, including building and construction, and
as far as possible for other branches of economic activity, should
be compiled regularly and where possible at annual intervals, and
distinguishing as far as possible between wages paid to wageearners and salaries paid to salaried employees.
1
Paragraph 1 of this text and certain provisions of the other paragraphs
were outside the scope of the subject included in the Agenda of the 24th Session
of the International Labour Conference, June 1938, and could therefore not be
considered by it.

INTERNATIONAL STANDARDISATION OP LABOUR STATISTICS

(2) An approximate indication should be given of the extent
to which the returns are representative of all employed persons.
[_226~\ 4. Statistics showing average annual earnings of workers
in each of the principal industries in mining and manufacture,
including building and construction, and as far as possible for
other branches of economic activity should be compiled at intervals
of three years, distinguishing as far as possible between wages paid
to wage earners and salaries paid to salaried employees.
[227~\ 5. The statistics of time rates of wages should so far as
possible be accompanied by indications as to the number of workers
covered, as well as by any information which may be available
as to the relation between the time rates of wages and actual
earnings.
\_228] 6. The statistics compiled in pursuance of this Recommendation should be published and communicated to the International Labour Office as prescribed by the Statistics of Wages
and Hours Convention for the statistics compiled in pursuance
thereof.
13. Proposals Calling for Further Action 1
{229} 1. (1) At least once every ten years, the statistics of
average earnings and of hours actually worked relating to wage
earners employed in the principal mining and manufacturing
industries, including building and construction, should give separate
figures for each of the principal occupations in each industry.
[230~] (2) These statistics:
(a) should show the number of workpeople grouped according to the amount of their earnings ; and
(b) should be given by sex and by age groups.
[23f\ 2. (1) Statistics showing the aggregate amount of
wages paid per annum in each of the principal industries in mining
and manufacture, including building and construction, should be
compiled regularly and where possible at annual intervals.
1
Based on the text of the proposed Recommendation suggested by the Fifth
International Conference of Labour Statisticians, Geneva, 1937. The International Labour Conference at its 24th Session, June 1938, adopted a resolution
requesting the Governing Body of the International Labour Office to arrange for
these proposals to be examined in accordance with the procedure laid down in
Article 24 of the Convention, and meanwhile to communicate them to Governments. On 2 February 1939 the Governing Body instructed the International
Labour Office to communicate the proposals to Governments for their information
and noted that action should be taken in accordance with Article 24 of the Convention in due course after the coming into force of the Convention.

STATISTICS OF WAGES AND HOURS OF WORK

[_232~\ (2) An approximate indication should be given of the
extent to which the returns are representative of all employed
persons.
[233] 3. Statistics showing average annual earnings of workers
employed in each of the principal industries in mining and manufacture, including building and construction, should be compiled
at intervals not exceeding three years.
[234] 4. The statistics of time rates of wages should so far as
possible be accompanied by:
(a) indications as to the number of workers covered, and
(è) any information which may be available as to the
relation between the time rates of wages and actual earnings.

IV
COST OF LIVING INDEX NUMBERS!
14. Resolutions of t h e Second International Conference of
Labour Statisticians, Geneva, April 1925
COST-OF-LIVING I N D E X NUMBERS

(intended to measure changes in the cost of living in a given country at
different periods)
[23f] (1) In each country statistics should be published
showing changes in the cost of living. They should cover the food,
clothing, heating and lighting, rent, and miscellaneous groups.
Each group should include the most important items of consumption, and especially those articles the price changes of which may
be taken as representative of changes in the prices of articles not
included.
[236~] Separate index numbers should be published for each
of the groups and general index numbers for all the groups combined. In calculating these index numbers weights should be used
proportionate to the relative importance in consumption of the
items in each group and of the different groups. Changes in the
cost of living in a given period should be calculated by using
weights which remain unchanged throughout the period. When
changes in consumption render the weights inappropriate a new
series of index numbers based on new weights should be calculated.
[_237~\ It would be desirable that as far as possible the information published should enable the differences which would result
in the index numbers from the inclusion or exclusion of direct
taxes to be determined.
[238~\ (2) A series of national index numbers, calculated by
using weights based on the consumption of industrial workers
generally, should be published. In cases where desirable, index
numbers may also be calculated for other classes of workers.
[239] In countries in which there are marked differences in
economic conditions in different districts or among different sections
See also V. International Comparison of Real Wages.

COST OF LIVING INDEX NUMBERS

of the working classes, it is desirable that separate index numbers
should be published for such areas and classes. In these cases the
weights used should, as far as possible, be based on working-class
consumption in the districts or among the sections of workers
to which the index numbers relate.
[240] It may be convenient that the national index should be
computed by combining the separate indices, account being taken
of the relative populations in the different groups.
[_241~] (3) In view of the existing inadequacy of statistics of
consumption, the most satisfactory method of determining the
weights of the different items and groups appears to be that based
on a standard family budget obtained as the result of an enquiry
into the average expenditure of a number of families during a
given period.
[242] Where statistics of family budgets are not available,
statistics of aggregate consumption may be of some value in the
determination of the relative importance of certain articles of
consumption in the country as a whole, account being taken of the
fact that certain articles are used for purposes other than human
consumption.
[243] Theoretical budgets computed from all information
available may also be used.
[244] (4) Care should be taken to ensure that the quality
of each item for which prices are obtained at different dates and
the methods of collection and calculation remain unchanged during
the period covered.
[2452 (5) Index numbers should be computed as often as
may be necessary having regard to the extent to which prices are
changing. In present conditions they may with advantage be
computed at monthly intervals, but under more stable conditions
their publication at less frequent intervals may be adequate.
[246'] (6) 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.
[247"] (7) It is desirable that the Governments, in collaboration with the International Labour Office, should agree to adopt
the same year as base for the calculation of a new series of index
numbers. I t is suggested that the year 1930 would be suitable for
this purpose.
[248] (8) It is of great importance that the principles adopted
in the selection of articles and of the different qualities of such
articles, in the collection and computation of the price data and

100

INTERNATIONAL STANDARDISATION OP LABOUR STATISTICS

in the fixing of the weights, should be indicated in full detail when
such index numbers are first published, also the precise data to
which the index numbers refer.
15. International Convention relating to Economic Statistics, of 14 December 1928, adopted by t h e International
Conference relating to Economic Statistics
called by the L e a g u e of N a t i o n s 1
VI.

INDEX NUMBERS OP P R I C E S

[249] Index numbers:
(6) Showing the general movement of the cost of living,
to be compiled and published at least quarterly.
\_250] The cost of living indices may be based on data relating
to a single town or to several towns selected as being representative
and taken either separately or collectively.
[251] Each statement of index numbers mentioned above shall
contain a reference to a short official publication showing the items
the prices of which have been used, and the methods employed in
the calculation of the indices.
16. R e c o m m e n d a t i o n of t h e International Conference
relating to Economic S t a t i s t i c s called by t h e
League of Nations 2
S_252~\ IX. That, with a view to ensuring the comparability
of index numbers referred to in Article 2-VI of this Convention,
the same year or period should be taken as a basis for such index
numbers in all countries, and that the Committee of Experts,
taking into consideration the work of the International Conferences
of Labour Statisticians convened by the International Labour
Office and the work of the International Institute of Statistics,
should investigate this matter and prepare a report thereon, which
should be circulated to the Governments of all the High Contracting Parties.
1
LEAGUE OP NATIONS: International Conference relating to Economic Statistics
(Geneva, 1928), International Convention, Article 2, p. 4 (Ser. L.O.N. Pub. 1928.
II. 52).
* Ibid., Final Act, Recommendation IX, p. 76.

V
INTERNATIONAL COMPARISONS OF REAL WAGES AND
COSTS OF LIVING
17. Resolutions of the Second International Conference of
Labour Statisticians, April 1925
£255] The Conference recognises the great theoretical and
practical importance of the international comparisons of real
wages which were initiated by the British Ministry of Labour and
the interest of the researches which have been subsequently carried
on by the International Labour Office and which have certainly
led to progress in this sphere.
\_254~] The Conference, however, expresses the opinion that,
since these figures cannot be considered as representative of the
real differences in the workers' standard of living which exist from
country to country, it is desirable that a second series of index
numbers should be published along with the first, measuring the
relative standards of living of the working classes in the different
countries. In this case it would be necessary to base the calculations on the actual earnings of working-class families and to take
account of the differences in the physiological needs of the population due chiefly to climate and race.
[255'] The International Labour Office, in publishing these
index numbers, should give in the greatest detail the original figures,
the methods of calculation adopted, and any reservations necessary.
The different countries should send to the International Labour
Office the fullest information, together with a detailed description
as to the nature and value of their figures.
18. Resolutions of the First International Conference of
Statisticians convoked by t h e Social Science Research
Council of t h e United States of America,
January 1929
INTERNATIONAL COMPARISONS OP W A G E S

I
[256} (1) The Committee is impressed by the fact that a
great variety of wage statistics is collected in the various countries.

102

INTERNATIONAL STANDARDISATION OP LABOUR STATISTICS

In general these statistics may be described as statistics of earnings
or of wage rates. Some countries are strong in statistics of rates,
others of earnings. It is desirable that records of the movement
both of rates of wages and of earnings should be available. The
Committee therefore recommends t h a t :
(a) the countries which are weak in statistics of earnings
take steps to add to those statistics by collecting statistics of
earnings, for the principal occupations as well as for the chief
industries, and of total hours worked from which it would be
possible to compute earnings per hour, and that
(6) the countries which have concentrated on statistics of
earnings improve their figures of wage rates and of changes in
wage rates.
\ZS7"\ (2) I t is evidently of the greatest importance that the
expressions "earnings" and "rates of wages" should have the same
meaning in the statistics of the different countries. With this object
the Committee recommends t h a t :
(a) the statistics of the different countries indicate clearly
what is understood by the expressions "rates of wages" and
"earnings", and particularly what are the payments in money
and in kind which are included in "earnings";
(6) at intervals, special enquiries should be undertaken to
show what are the differences between the rates of wages and
average wages (earnings) of different branches of industry and
categories of workers.
\ZSB~\ (3) The Committee is of the opinion that attention
might be particularly directed to defects in the information at
present available in the various countries if the International
Labour Office would compile an annual statement, if possible in
tabular form, by industries, either of wage rates or earnings or
both. The statement should, if possible, also include for the various
industries in each country the full-time and actual hours together
with their distribution.
It is considered that the publication of such information would
encourage the various countries to supplement their statistics as
to rates or earnings or both, and that in course of time the annual
statement would gradually improve until it would provide material
for more adequate comparisons.
II
\_Z59~\ The Committee is impressed by the limitations of the
present quarterly statistics of wages published by the International

INTERNATIONAL

COMPARISONS

103

Labour Office, due to the fact that few industries and occupations
in, generally, only one city in each country are included.
[_26(f\ The Committee believes that it would be possible to
improve the statistics by adding to the number of occupations
and of localities. It therefore recommends that the Conference
invite the International Labour Office to consider whether these
defects could possibly be remedied by asking the national authorities concerned to supply them with additional information necessary for this purpose.
[26 f\ The Committee has selected a few additional occupations
which appear not unsuitable for immediate inclusion. These are
named in the annex to these Recommendations. The Committee
also considers that the number might be further increased.
\_262~] The Committee also ventures to suggest that the International Labour Office might be asked to consider the possibility
of consulting with competent bodies in the several countries as to
whether other localities could be added to the present list, and
the further question whether it might be possible to indicate more
definitely the divisions of industry to which some of the occupations
included belong and to omit any localities which are not really
representative of any country as a whole.
\_263~] The occupations proposed for addition are confined to
non-localised industries. It might be possible also to include occupations from other non-localised industries, such as railway transportation, and also commercial and municipal employees. Localised
industries such as iron and steel, textiles, mining, and shipbuilding
can only be dealt with by consultation with the representatives of
the several countries, and certain occupations which come under
this head and the towns in which they are found might also be
added to the list.
\_264] If fuller information on the lines indicated is obtainable,
the collection and publication of statistics for localised and nonlocalised industries on a half-yearly or yearly basis might be
adequate.
[265^} ANNEX

Building Industry:
TransportStructural iron workers.
Tram and bus drivers.
Concrete workers.
Tram and bus conductors.
Electric Power Distribution:
Motor drivers (van and lorry).
Electrical engineers.
Horse drivers.
Electrical Installation (Buildings):
Electrical fitters.

104

INTERNATIONAL STANDARDISATION OF LABOUR STATISTICS

19. Resolutions of the Fourth International Conference of
Labour Statisticians, Geneva, May 1931
INTERNATIONAL COMPARISONS OP REAL WAGES

\_266~\ This Conference, recognising the value of the information as to wages and prices which has in the past been collected
by the International Labour Office for the purpose of calculations
relating to the purchasing power of wages in different countries,
and taking account of the resolutions adopted by the Governing
Body at its Fifty-first Session, with the object of improving the
bases and methods of compilation of such statistics, makes the
following recommendations:
[2671, (1) The International Labour Office should continue
to collect and publish particulars of wages, in a representative
selection of towns, industries and occupations in those countries
for which information is obtainable; the statistics hitherto published
should be extended, as far as possible, and in publishing
the information the Office should give detailed particulars as to the
sources, nature and scope of the data included.
\_268^\ (2) Information as to earnings per unit of time are preferable for purposes of international comparison but can generally
be furnished only as a result of special enquiries. Governments
should make these enquiries on uniform lines on the basis of the
resolutions adopted by the first International Conference of Labour
Statisticians and should supply the information to the Office.
Comparisons based on this information should be made and published by the Office.
\_269~\ (3) Time rates of wages and piece-work basis time
rates as established by collective agreements or other arrangements
can be furnished by many countries without special enquiry. These
are of value for purposes of international comparison and should
be collected and published.
\_270\ (4) In principle, it is undesirable to compare rates of
wages in one country with actual earnings in another; but where,
owing to incomplete information, rates of wages in certain countries
have to be compared with earnings in other countries, any available
information as-to the relation between the level of rates of wages
and earnings should be used to correct the figures of wage rates.
\_271), (5) In addition to the data as to money wages referred
to in the foregoing paragraphs, particulars should be supplied with
regard to the amounts of family allowances, if any, paid to the
various classes of workpeople.
\272~\ (6) The International Labour Office should continue to
collect and publish statistics of the retail prices of the articles of

INTERNATIONAL COMPARISONS

105

food consumed by working-class families in the towns for which
particulars as to wages are collected; particulars should also be
collected of the prices of fuel and light in these towns. The statistics
hitherto published should be extended as far as possible and in
publishing the information the Office should give such indications
as can be obtained as to the scope of the data.
[2732 (7) In view of the desirability of including, in the data
relating to cost of living, information as to the relative level of rents
in each country, the International Labour Office should invite the
statistical authorities in each country to furnish regular information with regard to the average rents of the predominant types
of working-class dwellings in the towns for which statistics of
wages and prices are supplied, together with a description of the
accommodation provided and of the general character of the dwellings covered. The International Labour Office should study the
possibility of obtaining one or more standard units of housing for
the various purposes of international comparison.
[2742 (8) In those countries in which no family budget enquiries have been made during the last ten years, such enquiries
should be undertaken at the earliest possible date, on the lines
laid down in the resolutions adopted at the Third Conference of
Labour Statisticians in 1926.
[275^ (9) The wages and prices information referred to in the
preceding paragraphs should be collected and published by the
International Labour Office at annual intervals. The Office should
address each year to the statistical offices of each contributing
Government a questionnaire soliciting the fullest possible particulars of wages and retail prices and rents relating as nearly as possible
to the month of October. In view of the fact that there are important seasonal variations in the prices of certain foodstuffs, such
prices should be obtained several times a year.
[2762 (10) The information so obtained should be tabulated
in a form facilitating ready comparison, and should be published,
at the earliest possible date, in a special annual volume. This
volume would be, in essentials, a development and amplification
of the statistical tables published by the Office at regular intervals
in the International Labour Review. It should include, as regards
wages, tables both of rates of wages and of earnings. It should
include explanatory notes regarding the source, nature and scope
of all the figures given, and should set out and illustrate the various
possible methods of combining the data so as to yield any desired
indices of relative real wages. This Conference believes that such
a volume would increase in scope and in importance with each
addition to the range of information obtained ; and would form from

106

INTERNATIONAL STANDARDISATION OF LABOUR STATISTICS

the outset the nucleus of what must become in due course an authoritative source-book of comparable international information regarding wages and retail prices in all industrially developed countries.
\_277~] (11) Following upon the publication of this volume,
there should be prepared and published in the International Labour
Review an article in which the data contained in the published
volume should be discussed from the standpoint of determining the
relative levels of purchasing power of wages in the various countries. In this article the difficulties of the procedure should be
fully and frankly disclosed. Indices on various bases should be
given, with the purpose of indicating the various points of view
from which comparisons can be made and of preventing a n y one
figure being regarded as authoritative for all purposes.
' [278~\ In view of the fact that the index numbers of purchasing
power of wages serve different national purposes, it is impracticable
for the International Labour Office to compute all the comparisons
which are possible between a series of countries and between different occupations and industries. The series of index numbers compiled by the Office should be illustrative of the methods by which
the data may be used for computing further series.
[27PJ (12) In any statistics which may in future be compiled
by the International Labour Office as to the relative purchasing
power of wages in different countries, account should be taken of
the diverse circumstances and conditions of the workers in different
countries resulting, for example, from differences: (a) in systems
of wage payment; (b) in the nature and amount of payments supplementary to wages, such as family allowances and paid holidays;
and (c) in systems of social insurance. Alternative series of index
numbers should be so far as possible computed to show the effect
of the inclusion or exclusion of such items.
[280~\ (13) The calculation of these index numbers by the
use of information as to working-class consumption furnished by
family budget enquiries should be made not only as hitherto on a
fixed international budget but also on the basis of various national
or regional budgets.
\_281~\ (14) The Conference realises that comparisons cannot
usefully be made between countries of widely differing habits and
customs; and that the closer are the consumption habits in different
countries, the more trustworthy are the comparisons likely to be.
\_282~\ (15) The Conference is unable to recommend the proposal to compare countries of dissimilar conditions by progression
through countries with intermediate conditions owing to the lack
of information on such conditions at the present time and to the
absence of any criterion for measuring the degree of dissimilarity
among the intermediate countries.

INTERNATIONAL

COMPARISONS

107

[283~\ (16) in all comparisons between two or more countries,
however, allowance should be made for the difference in articles
consumed in each of the countries by basing the calculations successively on the list of important articles consumed in one country
and the list of important articles (but not necessarily the same
articles) consumed in the others.
{.284} (17) It is also desirable that as regards certain articles
of food for which direct comparison of prices as between one country and another is impracticable, the International Labour Office
should explore the question of supplementing the present method
of calculation by a method in which account would be taken of
the possibilities of comparisons of prices on a basis of nutritive
value.
R E COMMENDATION

£2852 The Conference recommends that the Governing Body
place the question of the supply of the data required for the purpose of these international comparisons of wages and cost of living
on the agenda of a future International Labour Conference with
a view to the framing of a Convention binding the Governments
which ratify it to collect and supply the information at regular
intervals.
£286~] It recommends that a small committee of experts representing the competent national statistical authorities might
be set up. The function of this committee would be to assist the
Office in its work of developing and publishing wage and cost-ofliving statistics and in preparing for any future international conference which might appear desirable in the near future.
A.

STATISTICS

OF ELEMENTS
OTHER
THAN

OF
WAGES

REMUNERATION

20. Resolution of t h e C o m m i t t e e oí S t a t i s t i c a l Experts of t h e
International L a b o u r OSice, First Session, Geneva,
12-15 December Í933
[287~\ The Committee,
Having examined the results of the latest enquiry into elements
of remuneration other than wages (family allowances, paid holidays,
employers' and workers' contributions to compulsory social insurances) carried out by the Office in conformity with the resolutions of the Fourth International Conference of Labour Statisticians,
as a complement to its enquiry into wages, makes the following
recommendations :

108

INTERNATIONAL STANDARDISATION OF LABOUR STATISTICS

\_288~] 1. In order that the results of such enquiries may serve
really useful purposes, particularly in contributing towards a complete explanation of the data relating to wages, it is desirable that
the national statistical services of all the countries which collaborate
in the annual enquiry into wages should be able to supply more
complete information on these elements of remuneration, and
should furnish more exact information as to the sources, character
and scope of the figures, so as to enable the portion which these
different factors form of the total remuneration of the workers to
be evaluated;
[_289~] 2. The Office should continue its work with a view to
elaborating methods appropriate for arriving at such evaluations,
taking into consideration the different hypotheses which may
be adopted in this respect;
\_290~] 3. In view of the nature of these different elements and
the lack of adequate information, it is not essential that the simultaneous calculation of several index numbers of wages, showing
the effect of including or excluding these factors, should be repeated each year. It would be sufficient to recall, in the analysis
dealing with the wages tabulations, the general character of the
supplementary remuneration of this kind existing in each country,
and periodically to revise the available information and these
calculations by special enquiries at intervals of some years.
B.

STA TISTICS

OF THE COST OF

LIVING

21. Resolutions of the Second International Conference of
Statisticians convoked by the Social Science Research
Council of the United States of America, May 1930
[20 JQ I. In view of the fact that international comparisons of
cost of living which are the bases of all international real wages
comparisons are defective because of the lack in various countries
of adequate recent comparable budget studies, the Committee
recommends that the various countries which have not made
family budget enquiries within the last five years be requested to
make new and comprehensive studies of the budgets of workingclass families; to conduct these enquiries as soon as possible (and
not later than the end of 1932) ; and to take account of the studies
on these subjects issued by the International Labour Office and
recommendations made by the several Conferences of Labour
Statisticians.
[2P2J II. In order to supplement international real wage
comparisons and to throw further light on actual living conditions
among workers in various countries, the Committee recommends

INTERNATIONAL

COMPARISONS

109

that a descriptive detailed study be made (carefully planned and
very limited in scope) of the mode of life (work, wages, lodging,
dress, dietary, amusements, etc.) of typical high- medium- and
low-paid workers in various countries, such as railway workers,
miners, builders, etc.
[,2932 H I . In order to see how far different methods of comparing real wages agree with one another, the Committee recommends as desirable:
[294] (a) while awaiting new budget enquiries, to consider the compilation of these comparisons on the basis of the
budgets of the principal countries included in the comparison,
instead of grouping the budgets, or using an international
budget; such comparisons might be made once a year;
[2P5J (ò) when results of family budgets are available, to
make allowances as far as possible in indices based on food,
fuel and lighting, for rents and other items of expenditure; the
results of these allowances should be indicated separately;
[2962 (c) in the case of countries where there are outstanding differences in the mode of living in different districts, to
use as far as possible, where comparisons relate to wages in
different towns, the budgets of the towns considered, and,
where comparisons relate to wages in the country as a whole,
the national budget;
£297'] (d) that in view of the fact that cost-of-living index
numbers depend more on the prices than on the composition of
the budgets, and that retail prices of the same article vary
considerably according to quality, precautions be taken with
a view to ensure that, as far as possible, the comparability of
prices is maintained. It would be desirable before making comparisons that the statistical offices of different countries be
requested to furnish detailed information as to the quality of
the articles indicated in the budgets, and for any observations
to which these budgets may give rise from the point of view of
international comparisons.
C. STATISTICS
OF THE COST OF FOOD
22. Resolution of the Committee of Statistical Experts of the
International Labour Office, First Session, Geneva,
12-15 December 1933
[298~} 1- As regards the cost of food, the Committee recommends that:
The International Labour Office continue;
(a)

Its theoretical research, not only on the lines indicated

110

INTERNATIONAL STANDARDISATION OF LABOUR STATISTICS

in its memorandum (CES 1/3/1933), but also in other directions and particularly from the point of view of physiological
standards;
(b) Its practical studies, and the periodical publication of
international indices of the cost of food.
[_299\ 2. Before indicating some of the principles on which it
is possible to express an opinion at present, the Committee recognises that, on the one hand, the absence, in very many cases, of
recent enquiries into family budgets classified according to incomes
and localities, and showing both the expenditure and the quantities
consumed, and, on the other hand, the defects in the statistics of
retail prices, present serious obstacles to the application of any
method of comparison.
[300] In order, however, to enable the International Labour
Office to utilise all the data of sufficient value available in the
different countries, the Committee recommends that in each case
the procedure applied should be that which is best adapted to the
character and the value of these data. It is nevertheless obvious
that relatively summary methods can alone be used when the data
necessary for more refined calculations are at present lacking. The
following principles therefore apply only to comparisons relating
to the countries in which the statistical information available
enables more elaborate methods to be applied.
[301] 3. In accordance with the recommendations of the
Fourth International Conference of Labour Statisticians, the comparisons should be made on the basis of national or regional budgets
of family expenditure ; further, indices obtained by different methods
should be calculated and published simultaneously.
[302] For the purpose of these index numbers, there should be
included within the same "region" only those countries, parts of
countries, or towns in which the habits of consumption and the
price systems are sufficiently comparable; in the first place, the
percentage of the total food expenditure represented by the expenditure on articles of common consumption should be sufficiently
high ; the further criteria suggested in the above-mentioned memorandum may then be applied, and account taken of variations in
comparability according to the level of incomes.
[303] In principle, "regions" should be formed by towns and
not by countries; consequently, they should not necessarily be
constituted in such a manner as to cover the whole of a country;
they may, for example, include certain towns and exclude others
belonging to the same country. When, however, the available
data do not admit of the separata consideration of several towns

INTERNATIONAL

COMPARISONS

111

within a country, a single town or an average covering several
towns may be taken, and in this case, in default of appropriate data,
it will be admissible to utilise, for the purpose of weighting, average
quantities relating to the country as a whole. Geographical proximity and comparability from points of view other than that of
food consumption may be taken into account in allocating a town
to one region or another in those cases in which it is almost or
wholly a matter of indifference to which region a town is allotted.
It will be expedient to restrict the number of regions as far as
possible.
[304] In order to reduce the calculations involved, the number
of towns should be reduced so far as possible to the capital and,
where necessary, some other representative towns in each country.
[305] 4. In each "region" a town should be chosen as "regional
centre", well comparable with the other towns in the "region".
The Office is invited to prepare a primary system of "regions"
which will serve as a basis for its calculations. The Office might,
in addition, study other systems of "regions" with a view to the
ultimate adoption of the system which would yield the best results.
\_306~] 5. For the comparisons within each region, it will be
sufficient to calculate one index between each town and its regional
centre.
[307] 6. For comparisons between the regional centres a
formula should be adopted which takes account of the consumption
in each of them. Indices of this kind should be accompanied by an
indication as to the degree of confidence which can be placed in
them so as to avoid any misinterpretations.
[305] 7. For the calculation of an index relating to two towns
belonging to different regions the Committee is of opinion that it
is impossible to formulate a general rule applicable to all cases.
If questions of this nature arise, the Office should select in each
particular case, the method of calculation which appears the most
appropriate.
[30P] 8. The publication should include any necessary reservations, and the Committee expresses the opinion that it should contain not only the computed indices, but also the original data used
in the calculations.
D. STATISTICS

RENTS

23. Resolution of the Committee of Statistical Experts of the
International Labour Office, First Session, Geneva,
12-15 December 1933
\_310} As regards statistics of rents, the Committee:

112

INTERNATIONAL STANDARDISATION OF LABOUR STATISTICS

Considering that in international comparisons of cost of living
it is of prime importance to be able to include expenditure on rent,
since this may appreciably modify the relative positions established
on the basis of the cost of food alone;
Considering that the problem needs to be explored more exhaustively both from the theoretical and from the practical point of view,
before the publication of index numbers of rent is undertaken ;
Having examined with the greatest interest the memorandum
(CES 1/4/1933) submitted by the International Labour Office, in
which, after a very complete discussion of the principal methods of
calculation which might be considered applicable in this hitherto
little explored field, the results of various application of these
methods are analysed and compared ;
Considering that this preliminary work of exploration and the
results of these experiments were essential for the purpose of determining the difficulties of the problem, and that the evidence at present
available does not appearto admit of the use of very refined methods;
Recommends
That the Office should continue its work on this question, that
it should especially endeavour to find methods which, even though
approximative, can be readily applied to a large number of countries,
and in particular, that it should examine:
\_311~] (a) The correction to be made to the figures of rents per
room in order to allow for the differences in the prevailing types of
dwellings in different towns, especially as regards the size of the
rooms, and for this purpose should seek the collaboration of the
statistical offices in a number of countries.
\_312~\ (b) The relation which exists between the amount of rent
per head in the predominant types of working class dwellings in
each town.
24. Resolutions of the C o m m i t t e e of Statistical Experts of the
International Labour Office, Second Session, Geneva,
16-18 December 1935
[_3132 In the first place, the Committee recommends that, in
the collection of the data designed to serve as the basis for such
comparisons, information should be obtained which will enable the
predominant types of dwellings in each town to be classified, on the
one hand according to their general standard of accommodation,
and, on the other hand, according to the number of habitable rooms.
[314] As regards the first of these points, the Committee
recommends that the following approximate classification should
be adopted:

INTERNATIONA!, COMPARISONS

113

(a) dwellings with bathroom, private w . c , and water laid
on in the dwelling;
(b) dwellings with private w.c, and water laid on in the
dwelling (but without bathroom) ;
(c) dwellings in buildings containing a number of tenements, with w.c. common to several dwellings on the same floor,
and water laid on either in the dwelling or on the same floor;
(d) dwellings in houses containing one or two families;
with closet in common outside the house, and with water laid
on, either inside each dwelling or common to both families;
(e) dwellings in buildings containing a number of tenements, with exterior closets, and water supply in common.
\_315] As regards the second point, dwellings of one, two, three,
four, five and six rooms should be separately distinguished. The
rooms taken into account should be habitable rooms generally
(dining-room, sitting-room and bedroom) including the kitchen,
but not corridors, bathrooms, sculleries, etc.
[_316] It should be understood that, in each town covered by
the statistics, data relating to rents need only be collected for the
predominant types of working-class dwellings in that town.
£317] In the second place, having regard to the marked variations, between different towns, in the size of the rooms, the Cornmittee considers that rents distinguished only according to the
numbers of rooms do not form a sufficient basis for comparisons.
The Committee accordingly recommends that particulars of the
rents of each category of dwelling should be accompanied by an
indication of the total floor-space of each type of dwelling (exclusive
of cellars and attics unsuitable for use as living-rooms), and that
on the basis of these particulars, index numbers should be compiled,
for each type of dwelling enumerated above, showing the relative
level of rents expressed (a) per room and (ò) per square metre of
total floor-space.
£318] The data collected regarding rents, and the two series
of index numbers, should be accompanied, when published, by a
full explanation of their significance and interpretation.
£319] It should be observed that it would not be necessary to
repeat the enquiries relating to floor-space on each occasion when
particulars of rents are collected. When representative particulars,
on this subject, have once been obtained, they will generally be
valid for a period of several years.
[320] The Committee recommends that the International
Labour Office should now take the necessary steps, in collaboration
with the national statistical authorities, to collect data, on the

114

INTERNATIONAL STANDARDISATION OF LABOUR STATISTICS

basis of these proposals, as to the rents of working-class dwellings
in different countries. In particular, the Office should compile a
list of the particulars which should, as a minimum, be obtained
when special enquiries as to rents are made, and also on any occasion when a general housing census is taken. The data necessary
for the calculation of the index numbers should be transmitted to
the International Labour Office at regular intervals and, if possible,
at least every two or three years.
\_32f] The Committee also recommends that the Office should
continue its theoretical studies, not only along the lines indicated
in its memoranda, but with the further object of making international comparisons of expenditure on rent; in particular that it
should study the criteria which are necessary for ensuring comparability, and the question of classifying the various towns in
regions of good comparability; and also recommends that the
results of these studies should be submitted to a further meeting
of the Committee.
25. Resolutions of the C o m m i t t e e of Statistical Experts of the
International Labour Office, T h i r d Session, Geneva,
6-7 October 1937
[322] 1. The Committee have examined the results of the
further enquiry into rents of working-class dwellings in different
towns carried out by the Office in accordance with suggestions
made by the Committee at its previous session. Data have been
obtained for 20 countries covering about 60 towns. An examination of these data and of the problems which have arisen concerning their compilation in certain countries represented on the Committee has shown that it is desirable to revise slightly certain
of the "standards" which were drawn up for the purpose of distinguishing dwellings according to their principal characteristics. Thus,
the Committee recommends that a distinction be made in Standard
A (dwellings with bathroom, private lavatory and running water)
between dwellings consisting of a house and those consisting of a
flat. Further, a special category might eventually be adopted for
dwellings in Standard A (flats) in which the cost of heating supplied
by the landlord is included in the rent without the possibility of
deducting it. On the other hand, the Committee suggests that the
Office should reconsider the definition of Standard C 1, for which
a very small number of data have been received, and, in particular,
should examine whether it would not be possible to cover a category
of dwelling more prevalent in certain countries by slightly modifying this definition, especially as regards the expression "shared
lavatory outside the house".

INTERNATIONAL COMPARISONS

115

[_32f\ As regards the general definition of a dwelling considered
at present strictly as a separate architectural unit, the Committee
thinks that this definition should be made more flexible so as to
cover parts of a house originally intended for one family but now
occupied by two families without being separated architecturally.
Finally, the Office should examine in further detail the procedure
to be adopted as regards the various services (water, hot and cold;
maintenance of common passages; heating; lighting; kitchen
equipment; etc.) which may be included in the rents, or for which
supplementary charges may be made.
\_324~\ 2. As regards the floor space of the dwellings, the Committee, after having re-examined the question and in particular the
character of the data available in different countries, recommends
that account should be taken only of the area of living rooms
(including kitchen) excluding subsidiary accommodation (such as
passages, corridors, lavatories, bathrooms, verandahs) as well as
cellars, attics, lofts, etc. When the only particulars available refer
to the floor space of the main parts of the dwelling (that is to
living rooms and adjoining subsidiary accommodation but excluding
attics, separate lofts, cellars, etc.) these data might be used after
deducting a certain fixed percentage corresponding to the proportion generally borne by this subsidiary accommodation as a whole;
this percentage, necessarily arbitrary, would be fixed by the Office
in consultation with experts on the subject. This correction, however, should be made only for flats and not for dwellings consisting
of separate houses, in which cases the percentage deduction would
be much more variable.
[325] 3. The Committee endorses the recommendation made
at its previous session concerning the further development of the
work of the Office on the question of international comparisons
of rents as one of the elements of international cost of living comparisons. For such studies it is necessary that comparable information should be available on housing conditions in the different
countries. It would therefore appear very desirable that housing
censuses, undertaken in many countries at the time of the general
population censuses, which form the most important source of
information on this subject, should be standardised as completely
as possible. The Committee having been informed of a decision
of the Assembly of the League of Nations concerning studies by
the Secretariat on the subject of housing, and of the work which
is being undertaken on this subject by other international organisations, decided to await the results of these studies before proceeding further with the examination of this question.

VI
METHODS OF FAMILY LIVING ENQUIRIES
26. Resolutions of the Third International Conference of
Labour Statisticians, Geneva, October 1926
METHODS OF FAMILY BUDGET ENQUIRIES 1

\_326~] (1) In order to provide adequate information with
regard to actual standards of living, enquiries should be conducted
at intervals generally of not more than ten years into the income,
expenditure, and conditions of living of families representative of
large homogeneous sections of the population.
[ 3 2 / ] The families should be selected to include a sufficient
number from different income classes within the sections of the
population under consideration, and also from different industries
and districts. It is preferable to ask for less detailed information
rather than to reduce the number of families covered by the enquiry.
[328] (2) In order that the results of the enquiries may be as
complete and comparable as possible, information should be based
on daily records of income and expenditure kept by a member of
the family for a period of twelve months.
[329] Where it would be impracticable to obtain annual records,
every effort should be made to secure from as large a number of
families as possible budgets covering at least four periods of not
less than a week, one in each quarter, or two periods of at least a
fortnight in different seasons of the year. These records would
supply adequate information regarding items of expenditure which
recur daily. In the case of items of income and expenditure (clothing, furniture, etc.) which recur only at considerable intervals,
annual records kept by a smaller number of families or information on which annual estimates could be based should be sufficient.
[330] Special forms or account books should be distributed to
the families for their use in making these records of income and
expenditure. It is desirable that competent persons should be ap1
For a discussion of the change of terminology from "family budget enquiries"
to "family living enquiries" see Robert Morse WOODBURY: Methods of Family
Living Studies, International Labour Office, Studies and Reports, Series N,
No. 23 (Geneva, 1940), pp. 2-3.

METHODS OP EAMILY LIVING ENQUIRIES

117

pointed to visit the families during the course of the enquiry and
advise them regarding the keeping of the records.
£3312 (3) The records should show the district in which the
family resides, the industry and occupation to which members of
the family who are gainfully occupied belong, and the composition
of the household, including the sex of each member and the age of
juvenile members. They should show in sufficient detail the nature
of the housing accommodation.
\_332~] Information should be given for each important item
of income and expenditure. By income should be understood earnings in money and kind, sums received from boarders and lodgers,
from insurance funds, pensions, and investments, together with
income from allotments or in the form of gifts. The chief items
of income and expenditure should be specified on the forms, or
account books, distributed. In the case of items of expenditure,
the quantity purchased, as well as the cost, should be recorded
wherever practicable. Where a family includes boarders, lodgers,
or domestic servants, information should be given to show the
proportion they represent in the family consumption. In the calculation of expenditure, the cost of maintenance of domestic servants
should be added to their money wages.
£3332 (4) In compiling the results of an enquiry, if the number
of budgets secured is adequate, separate averages should be given
for important districts and industries. Families of manual and
non-manual workers should be shown separately. Averages calculated for different income groups should be shown.
£334] In addition to averages per family and for families
classified according to size, the data expressed in terms of some
more precise unit of consumption, for example the consumption of
an adult male, should be given. For reducing data for families of
different size to terms of a common unit, it is desirable, where
suitable scales showing the relative consumption of persons of
different age and sex are available, to apply one scale to food
commodities and a second scale to other items.
£335'] The tabulated results should show the average income
from different sources, and the quantity of and expenditure on
each of the chief commodities consumed. Separate information
should be given for each item of income and of expenditure which
constitutes on the average not less than 1 per cent, of total income
or expenditure. Other items of interest may be shown separately as
desired. The items òf expenditure should be classified into the
following groups: food, clothing, housing accommodation, fuel and
light, furniture and furnishings, and miscellaneous. It is desirable
that payments of direct taxation should be given as a separate

118

INTERNATIONAL STANDARDISATION OP LABOUR STATISTICS

item. It is also desirable to determine the nutritive value of the
principal articles of food together with the cost of a thousand
calories of each commodity.
£33<f\ Where the sole object of an enquiry is to provide weights
for the calculation of cost-of-living index numbers, satisfactory
results may be obtained from a less detailed investigation than
that indicated above. Information regarding district, industry,
composition of family, and expenditure upon each of the several
items to be included in the index numbers only is necessary. It is
desirable, however, to have, in addition, information as to quantities.
rjJ373 The results will be tabulated to meet the requirements
of the series of cost-of-living index numbers which it is proposed
to publish in each country. Thus if separate series of cost-of-living
index numbers are to be published for a number of districts or
categories of workers, the results of the family budget enquiries
will be tabulated separately for each district or category; if a single
national index is to be calculated, the results may be shown in the
form of general averages only.
27. R e c o m m e n d a t i o n of t h e C o m m i t t e e of S t a t i s t i c a l
Experts of t h e I n t e r n a t i o n a l L a b o u r Office,
F i r s t Session, 12-15 D e c e m b e r 1933
[338] The Committee of Statistical Experts emphasises the
fact that statistical enquiries into international comparisons of
the cost of living and the purchasing power of wages cannot be
undertaken without information as to the family budgets of the
working classes. In view of the fact that in many countries no
family budget enquiries have been undertaken since a period now
more or less remote, which fact substantially impedes the Office
in its investigations, the Committee draws special attention to the
resolutions adopted by the Second and Third International Conferences of Labour Statisticians, which recommended that all
countries should conduct enquiries into family budgets at regular
intervals and on a uniform, or in any case comparable, basis. In
order to increase the value of these enquiries, it is desirable that
they should be carried out simultaneously, and at periods as near
to each other as possible.

VII
STATISTICS OF INDUSTRIAL ACCIDENTS
28. Resolutions of the First International Conference of
Labour Statisticians, Geneva, October 1923
(1) CLASSIFICATION OF INDUSTRIAL ACCIDENTS

[339] Industrial accidents should be classified according to the
industry of the injured worker, the cause of accident, the extent
and degree of disability, the location of the injury, and the nature
thereof.
[.340] (a) The classification of industrial accidents according
to the industry of the injured worker should conform to the list
indicated in paragraph 4 of the Resolution concerning the Classification of Industries 1 , with such subdivisions as will allow special consideration to be given to industries with a relatively high accident
rate.
[.3412 (b) The classification of accidents according to the
cause of accident should as far as possible be in accordance with
the table given below, with such subdivisions as may be considered
necessary.
(i)

Machinery:
(a) Prime-movers;
(fi) Transmission machinery;
(c) Lifting machinery;
(d) Working machinery;

(ii)

Transport:
(a)
(b)
(c)

(iii)

Railways ;
Ships;
Vehicles;

Explosions; fire;

i See \4J.

120

INTERNATIONAL STANDARDISATION OF LABOUR STATISTICS

(iv)
(v)
(vi)
(vii)
(viii)
(ix)
(x)
(xi)
(xii)
(xiii)

Poisonous, hot, or corrosive substances;
Electricity;
Falls of persons;
Stepping on or striking against objects;
Falling objects;
Falls of ground;
Handling without machinery;
Hand tools;
Animals;
Miscellaneous.

[342] (c) In the classification of accidents according to the
extent and degree of disability a distinction should be made between
fatal and non-fatal accidents and between temporary and permanent disabilities.
[34J] Temporary disabilities should be classified according to
duration, and uniformity should be obtained by using the following groups: (i) 2 weeks or less; (ii) over 2 and up to 4 weeks; (iii)
over 4 and up to 13 weeks; (iv) over 13 weeks and up to 6 months;
(v) over 6 months and up to 1 year; (vi) over 1 year and up to
2 years; (vii) over 2 years and up to 3 years.
[344] Permanent disabilities should be classified by degree,
and uniformity should be obtained by using the following groups:
(i) under 20 per cent, disability; (ii) 20 and under 40 per cent.;
(iii) 40 and under 60 per cent. ; (iv) 60 and under 80 per cent. ; (v)
80 and under 100 per cent. ; (vi) 100 per cent.
[345] Permanent disabilities should be classified at the time
they are recognised as such.
[346] (d) The location of injury should be clearly distinguished
from the nature of the injury. The most suitable classification is
that of the common anatomical divisions of the body, namely:
(i) the head ; (ii) trunk ; (iii) upper extremities ; (iv) lower extremities; (v) general.
Each of :hese groups should be subdivided if necessary.
Q?47J (e) The nature of the injury should be classified as
follows: (i) contusions and abrasions; (ii) burns and scalds; (iii)
concussions; (iv) cuts and lacerations; (v) punctured wounds; (vi)
amputations; (vii) dislocations; (viii) fractures; (ix) sprains and
strains; (x) asphyxiation; (xi) drowning; (xii) other injuries.
[348] Note. In publishing the above statistics a note should
be added on the following points:
(a)
(ò)

The scope of the legislation;
The system of insurance (compulsory or optional);

STATISTICS OF INDUSTRIAL ACCIDENTS

121

(c) The nature of the accidents included;
(d) The methods of reporting the accidents and of compiling the statistics;
(e) A summary of the benefits given to the injured or to
their dependants.
[349] In countries in which industrial diseases are compensated
as accidents they should, wherever possible, be distinguished
separately in the tables.
(2) ACCIDENT R A T E S
r

[3502 F ° industrial and international comparison it is essential
to calculate frequency rates and severity rates.
[351"] (a) The frequency rate should if possible be calculated
by dividing the number of accidents (multiplied by 100,000) by
the number of hours of working time.
[352} (b) The severity rate should similarly be calculated by
dividing the number of working hours lost (multiplied by 100,000)
by the number of hours of working time.
[353~] Where practical difficulties prevent the calculation of
the number of hours of working time, this number should be replaced by the number of full-time workers (i.e. the number of
working days divided by 300) or the average number of workers,
as may be best suited to the economic and social needs of the country or industry concerned.

Vili
HOUSING STATISTICS
29. Recommendations of the C o m m i t t e e of Statistical
Experts convoked by the International Union of
Towns (May 1928)
I.

GENERAL DEFINITIONS

{354} Dwelling.—The dwelling comprises a separate collection
of rooms which, for whatever purpose they may be in use, are at
the. date of the census intended for occupation by one family and
have separate access to a street or to a common passage or stairway. Detached rooms which are clearly intended to form a part
of the dwelling should be counted as part of the dwelling.
\_355} Premises not intended for habitation but in use as such
should be separately counted and should be reckoned as as many
units as there are families in occupation.
{356} Habitable buildings.—Habitable building comprises every
property standing by itself or completely separated from other
properties by party walls, intended as a whole or in part for habitation or in use for this purpose. Huts, tents, carts, caravans, boats
and barges, which are in use as dwellings but which cannot be included in this definition, should be classified separately.
{357} Rooms.—A room, defined as a space entirely enclosed,
should be distinguished as follows:
[358} (1) Habitable rooms, i.e. rooms intended to be actually
inhabited, such as bedrooms, rooms in which meals are habitually
eaten and rooms reserved for the common life of the family. The
kitchen (unless by reason of its limited size it should be reckoned
as a subsidiary room) should be reckoned as a habitable room.
Attics and servants' quarters should be reckoned as habitable
rooms when they are in fact habitable.
{.359} (2) Subsidiary rooms (sanitary accommodation, bathrooms, corridors, verandas, attics, lofts, wash-houses).

HOUSING STATISTICS

123

[360~] Occupants.—An occupant comprises every person who
is habitually in occupation of the dwelling at the date of the census.
Persons who although absent at the date of the census are ordinarily
in occupation should be included, and persons who are only in
temporary occupation should be excluded.
\J361~y Household.—The household consists of an independent
group of persons enjoying a common domestic life.
[_362} A distinction should be drawn between the family household and the non-family household. It would appear most easy
to draw the line of demarcation at the point where the number of
boarders exceeds the number of the permanent members of the
household (including domestic servants). In this connection it is
more important to have regard to the actual participation of any
person in the life of the family than to the form of contract between
the inhabitants.

II.

CENSUS OF D W B L U N G S

[363"} 1. Classification of dwellings. — Dwellings should be
classified as follows :
£.364} A. According to the purpose for which they were
intended and the purpose for which they are in use :
(1) Dwellings intended for the occupation of family
households;
(a) in use as such ;
(6) not actually in use, i.e. vacant and available for occupation.
(2) Dwellings intended for the occupation of nonfamily households;
(a) in use as such ;
(ò) not actually in use, i.e. vacant and available for occupation.
(3) Premises not intended for habitation but in
use for this purpose.
[[JÖ5] B. According to their size measured by the number of
habitable rooms and classified as follows :

124

INTERNATIONAL STANDARDISATION OP LABOUR STATISTICS

Dwellings with 1 habitable room
u
2
rooms
ii
It
3
ti
a
4
ti
u
5
II
ti
6
it
U
11
7
and more

(1)
(2)
(3)
(4)
(5)
(6)
(7)

[JtfiT] It is also desirable that, in the above categories, the existence of kitchen, sanitary accommodation or bathroom should be
indicated wherever possible.
[.3672 2. Classification of habitable buildings.—A. Habitable
buildings should be classified according to the proportion of the
building occupied by dwellings :
(1) habitable buildings primarily utilised as dwellings;
(2) habitable buildings primarily utilised for purposes other
than dwellings but which contain a certain number of dwellings.
[368} B. Habitable buildings primarily intended for the occupation of family households or utilised for this purpose should be
classified as follows:
(1)
(2)
(3)
(4)
(5)
C6)

Habitable buildings containing 1 dwelling
"
"
"
2 dwellings
"
"
"
3-4
"
"
"
5-10
"
"

"
"

11-20
21

a n d over.

[36&2 3. Classification of occupants according to the characteristics of the dwellings and habitable buildings which they
occupy.—The

occupants of dwellings intended for the occupation of family households should be classified as follows :
A. According to the size of the dwellings which they occupy,
arranged in the same categories as in II 1 B.
B. According to the character of the habitable buildings
arranged in the same categories as in II 2 B.
ß?70] 4. Combined classification according to the number of
rooms and the number of occupants of a dwelling.—The occupants of
dwellings intended for family habitation should be classified both
according to the number of habitable rooms contained in the dwellings which they occupy and according to the number of persons
occupying the dwelling as in the following table:

HOUSING STATISTICS

125

Number of occupants occupying dwellings of :
In dwellings
inhabited by

1 habitable
room

2 habitable
rooms

3 hab- 4 habitable itable
rooms rooms

7 hab6 hab- itable
itable rooms
rooms and over

Total

1 person
2 persons
3
4
5
6
7
8
9
10
11
and over
T otal

[_37lJ Dwellings should be classified according to the same
principle in a similar table:
Number of dwellings
Dwellings
occupied by

1 hab- 2 habitable itable
room rooms

3 habitable
rooms

4 habitable
rooms

7 hab6 hab- itable
itable rooms
rooms and over

Total

1 pen on
2 pers ons
i
3
i
4
i
5
i
6
7
i
8
i
9
10
t
11
and over
T otal

[_372~] 5. Classification according to density of occupation.—
The density of occupation should be calculated in the case of each
dwelling by dividing the number of occupants by the number of
habitable rooms.
[37'3j Dwellings in each of the size categories II 1 B should
then be classified according to their degree of density of occupation
as follows:

126

INTERNATIONAL STANDARDISATION OF LABOUR STATISTICS

Dwellings occupied by 4 or more persons per room
"
" 3 persons but less than 4 per room
it

-2

"
" less than 1 person per room.
[374} Occupants should also be classified according to the
density of occupation of the dwellings which they occupy, using
similar categories.
C-?75] 6. Classification of dwellings according to the number of
households in occupation.—Dwellings in each of the size categories
II § 1 B should be classified according to the number of households
in occupation.
[376} 7. Definition of rent.—The term "rent" should be taken
as signifying only those charges which strictly relate to the letting
of the premises, including any charges which may be made for the
services of caretaker, etc., where such charges are not included in
the net rent. All additional charges for lighting and heating the
dwelling should, on the other hand, be systematically excluded.
Rates and taxes payable by the tenant should be given separately.
In cases where it is not found possible to observe these distinctions,
indications should be given which would show the margin of error
involved.
[377} When a dwelling is occupied bv two or more households,
each household paying rent directly to the landlord, the total of
any rents so paid should be recorded. Rents paid for furnished
apartments and rooms, and rents paid for portions of dwellings
which are sub-let should be separately recorded.
£37<?J 8. Classification of dwellings according to the amount of
rent paid.—Dwellings which are rented by family households should
be classified according to the amount of rent paid, making use of a
certain number of groups (between 10 and 20) which should be
drawn up in accordance with a uniform plan in each country, having
regard to prevailing conditions. This classification should be combined with a classification of dwellings according to their size in the
categories suggested in II 1 B.
\_379~] The number of occupants of dwellings should also be
classified in a double classification, according to the amount of rent
paid and to the size of the dwelling occupied, making use of the same
categories.
[380} 9. Distinction between dwellings subject and not subject
to rent restrictions.—In countries where the restrictions as to the
amount of rent which may be charged for dwellings, imposed during,
and immediately after, the war are still in force, or still exert a
marked influence on the level of rents, it is important that in the

HOUSING STATISTICS

127

classification of dwellings, as also in the classification of occupants,
a distinction should be made between dwellings which are subject
to and those exempt from these restrictions.
III.

PERIODICAL STATISTICS RELATING TO VARIATIONS IN THE
TOTAL NUMBER OP EXISTING DWELLINGS

[381} Particulars of the variations in the total number of dwellings should be regularly obtained, based on information supplied
by the public authorities responsible for the supervision of new
buildings (registration of completed works; inspection of completed
works in order to ascertain whether they comply with the approved
plans; issue of habitation permits).
[382} The following particulars should be obtained at intervals
not exceeding one year:
[383} (a) The number of habitable buildings erected (distinguishing those erected to replace buildings demolished) ; the
number of such buildings structurally altered (by enlargement,
by the superposition of another storey and by interior alterations) ; the number of such buildings demolished or destroyed.
[384} (b) The number of dwellings erected; the number of
new dwellings brought into existence as a result of structural
alterations; the number of dwellings ceasing to exist as a result
of structural alterations; the number of dwellings demolished
or destroyed. In each case the dwellings should be classified
according to their size.
[385} It is also desirable to denote, in the case of buildings
erected, dwellings erected and dwellings brought into existence as a
result of structural alterations, whether the works have been undertaken by (a) public authorities, (b) institutions of public utility, or
(c) private enterprise.
IV.

PERIODICAL STATISTICS OF VACANT DWELLINGS

[386} It is desirable to obtain annual returns of vacant dwellings, i.e. habitable dwellings which are unoccupied and immediately
available for occupation.
[387} The dwellings should be classified according to their size
and in the case of dwellings intended for letting according to the
rent to be charged. These particulars could be supplied by the
Public Housing Offices in cases where the notification of vacancies
is obligatory.

128

INTERNATIONAL STANDARDISATION OP LABOUR STATISTICS

30. Excerpts from Recommendations of the Committee
of
Statistical Experts of the League of Nations1
HOUSING STATISTICS

A. Definitions of Basic Units
[_388~\ Rooms: As "rooms" should be regarded bedrooms, diningrooms, living-rooms, habitable attics, servants' rooms, kitchens and
other habitable spaces, separated from one another by walls reaching
from the floor to the ceiling and of a size large enough to hold a bed
for an adult (approximately 4 square metres at least).
\_38f\ Kitchenettes, corridors, verandas, lobbies, etc., not
falling under the above definition, as well as bathrooms and toilets,
should not be counted as rooms.
\_39(f] Rooms regardless of size, if used only for professional or
business purposes, should not be counted as parts of the dwelling
and should not be taken into account for calculating the number of
persons per room.
[_391~\ Dwelling: The dwelling should comprise any structurally
separated room or suite of rooms in permanent buildings used or
intended for habitation by family households and having a separate
access to a street or to a common passage or stairway. Detached

rooms for habitation which are clearly intended to form a part of
the dwelling should be counted as part of the dwelling. Rooms used
only for business purposes should not be counted as parts of the
dwelling.
[]JP2] The family household should comprise the members
of a family forming the nucleus of the household, including
resident domestic servants. Other persons sharing the rooms
occupied by the family or having separate rooms in the dwelling,
but taking their meals with the family, should be counted as members of the family household.
[3P3] Persons living alone and lodgers occupying rooms of
their own and not sharing in the meals of the family should be
counted as separate family households.
QJP4J It should be noted that, under the above definition, one
dwelling may be occupied by several family households, and that
the family household is not necessarily identical with the total
number of persons living in the same dwelling.
1
LEAGUE OF N A T I O N S : Housing Statistics: A Minimum Programme of Housing
Statistics draw n up by the Committee of Statistical Experts. Studies and Reports
on Statistical Methods, No. 5 (Official number: C. 133. M. 85. 1939 (C.E.S.
145) Appendix I; Ser. L.O.N. Pub. 1939. I I . A. 6).

HOUSING STATISTICS

129

[JP53 In the present minimum programme of housing statistics,
only the family household is used as a basic unit. For reasons of
precision, the Committee [gives also the definition of non-family
household].
[3P(i] Non-family households should comprise all persons
living in groups which are not to be considered as family households—e.g., in hotels, institutions, military barracks, etc.
C?07] In doubtful cases, which may particularly arise in respect
of pensions and boarding-houses, the border-line between the
family household and the non-family household should be drawn
at the point where the number of boarders, etc., exceeds the number
of the family members, including servants, etc.
[398~\ It should be noted that directors or the personnel of
institutions, hotels, etc., occupying a separate dwelling should be
counted as family households.
[3992 Occupants: If the methods adopted for enumerating the
population in the general population census enable the usual
residents of a dwelling to be identified as distinct from the persons
occupying the dwelling on the day of the census, only the usual
residents should be counted as occupants. Each physical person,
regardless of age, should be counted as one unit.
B. Statistics Relating to the Characteristics of Dwellings
[400] The Committee was of the opinion that the most important information bearing on the characteristics of dwellings is,
from the point of view of international comparisons, that dealing
with the number and size of the dwellings and the existence of
certain types of amenities.
[4012 The size of dwellings should be measured only by the
number of rooms, as any other method requiring reference to the
dimensions of the premises appears to be unsuitable for adoption
in general censuses.
[4022 As regards the amenities available' in a dwelling, the
Committee recommends that they should be ascertained for each
dwelling to which they structurally belong. There is no doubt that
it would be equally important to show the amenities available for
each household; the necessary basic information on this point is,
however, more difficult to obtain in practice and has therefore been
disregarded.
[403] The Committee recommends that the total number of
dwellings should be shown, subdivided by size into dwellings with
one, two, three, four, five, six, seven, eight and more rooms in
accordance with the definition of rooms given above.

130

INTERNATIONAL STANDARDISATION OF LABOUR STATISTICS

\_404~] For each of the above groups, the tabular analysis should
distinguish the number of dwellings possessing the following
amenities:
(1) Kitchenettes {i.e., kitchens not corresponding with the
definition of rooms) ;
(2) Bathrooms:
(a)
(ò)

Private ;
Common ;

(3) Toilets:
(a)
(b)

Private ;
Common ;

(4) Running-water supply:
(a)
(b)

Inside the dwelling;
Outside the dwelling;

(5) Electricity supply;
(6) Gas supply.
[405] In many countries, the heating system, and especially
central heating, is a matter of importance, but as international
comparisons of this kind have less significance owing to differences
in climatic conditions, the Committee did not include this topic
in its programme.
C. Statistics Relating to the Occupation of Dwellings and Rooms
I. Occupation of Dwellings by More than One Family

Household.

[406] One of the main objects of housing censuses in several
countries is to throw light on the occupation of dwellings by more
than one family household, as this information is of importance in the
determination of existing housing needs. The Committee therefore
recommends that the number of dwellings occupied by:
(a)
(ò)
(c)

One family household;
Two family households;
Three and more family households

should be given, at least separately for dwellings with one, two,
three, etc., eight and more rooms.
[407"] In addition to the total number of dwellings occupied
by family households, the number of vacant dwellings should be
given for the same room categories, and it should be exactly stated
which kinds of dwellings have been included as vacant dwellings.

131

HOUSING STATISTICS

II. Statistics Showing the Density of Occupation.
[408"] The occupation of dwellings and rooms should be shown
for international comparison on the basis of physical persons,
adults and children being counted as equal units. Although information as to sex, age, marital status and the situation in the
household is frequently available for all persons of the household,
the Committee abstained from including such distinctions in the
minimum programme.
'•' '• [409] Statistics showing the density of occupation—i.e., the
number of persons in relation to the number of rooms—are computed in different ways in various countries. They are generally
based on one of the following methods:
[410] (a) By reference to the dwellings as the unit, the
total number of persons being related to the total number of rooms
in the dwelling.
[411] (ò) By reference to the household as the unit, the
number of persons in a family being related to the number of rooms
occupied by that family.
[412] In those cases—the great majority—in which a dwelling
is occupied by one household, the two methods yield the same
result. In cases where more than one household occupies a dwelling,
the Committee recommends that the second method be adopted,
in view of the fact that the number of persons per room may vary
considerably for each of the households in the same dwelling. It is
understood that the kitchen, or any other room, if shared by more
than one household, should be allocated to the main household
in the dwelling.
.. [413] The Committee recommends that the tabulation displaying family households in relation to the rooms they occupy
should be in the following form:
NUMBER OF FAMILY HOUSEHOLDS OF DIFFERENT SIZES OCCUPYING
DIFFERENT NUMBERS OF ROOMS

Number of persons
per
family household^

One
Three
etc.
Ten and m o r e . . . .
Total number of
family households

Number of family households occupying the
following number of rooms
1

Total
number
of
10 family
and housemore holds

132

INTERNATIONAL STANDARDISATION OF LABOUR STATISTICS

[414} The Committee has adopted this form of presentation
as it permits of calculating the total number of persons associated
with each room category and the total number of rooms associated
with each size of family household. At the same time, the several
items can be aggregated in any desired density group, in terms
either of persons per room or rooms per person; an arrangement
which facilitates the application of varying national crowding or
overcrowding standards.
D.

Other Statistics Relating to Households and Dwellings

[4152 The Committee has considered various other statistics
which are frequently obtained from national housing censuses,
without making recommendations in respect of them for its first
minimum programme.
[4162 If the housing census is undertaken in connection with
a census of population in which occupations are recorded, the
necessary basic data for a classification of households by the occupation of the head of the household are available. The Committee
is aware that such a grouping is sometimes undertaken as a means
of throwing light on the housing conditions of various social groups,
and emphasises the importance of such grouping. At this stage,
however, the Committee does not suggest a uniform classification
for that purpose. The classifications prepared by the Committee
of Statistical Experts for the purpose of grouping the gainfullyoccupied population will offer certain groups which might be useful
also for the purpose of housing censuses.
[4172 Statistics of rents, although of importance for statistics
of housing conditions, have been excluded from consideration,
owing to the difficulty of defining what is included in rents in
different countries and the difficulty of obtaining such information
by means of a general housing census. In this connection, the SubCommittee draws attention to the compilation of rents on a uniform basis for certain selected types of dwellings which is being
undertaken by the International Labour Office and its Committee
of Statistical Experts. 1
[418] Statistics of the tenure of dwellings are generally required
if statistics of rents are prepared. They are, however, also of considerable importance for other national purposes, especially in
relation to differences between urban and rural districts and the
changes in the various forms of tenure. The definitions and classi1
See " R e n t s of Workers' Dwellings in Various Towns, 1936-37", International Labour Review, Vol. X X X V I I I , No. 6, Dec. 1938; also Robert G U Y E :
"A Further Contribution to the International Comparison of Rents", International Labour Review, Vol. X X X I V , No. 5, Nov. 1936.

HOUSING STATISTICS

133

fications needed depend largely on legal practice and national
customs; they have therefore not been included in the minimum
programme.
[419] Statistics of family income and their combination with
housing statistics are also of great interest. As, however, only a
very small number of countries compile such statistics, it was
impossible to include them in the minimum programme.

Geographical Scope of General Housing Censuses

[4202 In accordance with its terms of reference, the recommendations of the Committee apply to both urban and rural
housing. Rural housing conditions, however, differ substantially
in certain respects from urban conditions, and in view of the particular interest which, in many countries, attaches to the problem
of rural housing and of the special aspects of that problem 1 , the
Committee devoted particular attention to the information on
this subject which can be readily obtained on the occasion of a
population census. The Committee did. not consider it necessary
to define the border-line between urban and rural districts, as the
validity of the comparisons which can be made from the statistics
drawn up in accordance with the minimum programme is hardly
affected by the different solutions of the problem.
Note: In the report of the Committee of Statistical Experts the
section here given under E precedes the other sections. In addition
to the quotations given here the report contains a section
devoted to statistics of buildings; it includes also a general statement on the reasons for its recommendations; the personnel of the
Committee and its advisers are given on p. 5 of the report.

1
See, for example, the studies on rural housing being made by the Health
Organisation of the League and its Conferences on rural hygiene.

IX
STATISTICS OF COLLECTIVE AGREEMENTS
31. Resolutions of the Third International Conference of
Labour Statisticians, Geneva, October 1926
\_42l} In each country information concerning collective agreements and their principal contents should be collected and published
in a summary form at appropriate intervals.
[4222 It is desirable that any statistics compiled on the basis
of this information should be compiled in accordance with the
following general principles:
{42f\ (1) The collective agreement should be defined, for the
purposes of the statistics, as a written agreement concluded between
one or more employers or an employers' organisation on the one
hand, and one or more workers' organisations of any kind on the
other, with a view to determining the conditions of individual
employment, and in certain cases to the regulation of other questions relative to employment.
[424~] (2) The number of collective agreements should be
recorded at annual intervals according to the following scheme:
(a) number of agreements in force at the beginning of the
period of registration;
(b) number of agreements concluded during the period of
registration ;
(c) number of agreements expired within the period of
registration ;
(d) number of agreements in force at the end of the period
of registration.
[_425~] (3) The importance of each collective agreement should
be measured by ascertaining the numerical strength of the contracting parties, i.e. the number of establishments covered, the total
number of workers employed in these establishments, and the
number of workers covered by each agreement.

STATISTICS OE COLLECTIVE AGREEMENTS

135

[4262 The extent to which workers are covered by collective
agreements should be indicated by calculating the number of
workers covered by agreements as a percentage of the total number
of workers in the various industries.
[4272 (4) The collective agreements, together with the number
of establishments and of workers covered, should be classified
according to their principal legal and social characteristics on the
following lines:
[4282 A. Nature of Contracting Parties. The agreements should
be classified according to the nature of the contracting parties
as follows:
(a) agreements concluded between an employer and his
workers ;
(6) agreements concluded between one or more employers,
and one or more workers' organisations;
(c) agreements concluded between employers' organisations and workers' organisations.
[4292 B. Scope of Application.
The agreements should be
classified according to the extent of the area in which they are
applicable, as follows:
(a) shop agreements: i.e. agreements applicable to a single
establishment;
(b) local agreements: i.e. agreements applicable to several
or all establishments of similar kind situated in the same locality;
(c) district agreements: i.e. agreements applicable to several
or all establishments of similar kind situated in several or all
localities belonging to a district forming an economic or an
administrative unit;
(d) national agreements: i.e. agreements applicable to several
or all establishments of similar kind in several districts or in
the country as a whole.
[43Ö2 C. Subjects Regulated. The agreements should be classified in the following two principal groups:
(a) agreements regulating individual conditions of employment only;
(6) agreements regulating—in addition to individual conditions of employment—general matters relative to employment.
In group (b) the number of agreements providing for special
procedure for the enforcement of the agreement may be shown
separately.

136

INTERNATIONAL STANDARDISATION OF LABOUR STATISTICS

{431} The statistics should also indicate the number of agreements which regulate each subject of importance, e.g. wages, hours
of work, holidays, conditions of apprenticeship, labour exchanges,
works councils, conciliation and arbitration.
{432} D. Duration of Validity. The agreements should be classified according to the period for which they are concluded, as
follows :
(a)
(Ò)
(c)
(d)
(e)
(/)
(g)

3 months or less;
3 to 6 months;
6 months to one year;
1 to 2 years;
2 to 3 years;
more than 3 years;
indefinite period.

{433} E. Method of Conclusion. The agreements should be
classified according to the method of the conclusion of the agreement
and according to the method of negotiation, as follows:
(a) Collective agreements concluded as a consequence of
an industrial dispute:
(i)
(ii)

by direct negotiations,
through the intervention of a third party;

(6) Collective agreements concluded as a consequence of
peaceful discussion:
(i)
(ii)

by direct negotiations,
through the intervention of a third party.

{434} F. Industries Covered. The agreements should be classified according to the principal industrial groups. The classification
may be left to the domain of national statistics.
{435} G. Industrial Importance. The agreements should be
classified according to their industrial importance, as defined
under (3).
{436"] (a) Classification of agreements by the number of
establishments covered :
(i)
(ii)
(iii)
(iv)

agreements covering one establishment,
"
"
2 to 20 establishments,
"
"
21 to 100 establishments,
"
" more than 100 establishments.

[437"} (6) Classification of agreements according to the
number of workers covered:

STATISTICS OP COLLECTIVE AGREEMENTS

137

(i) agreements covering less than 100 workers,
(ii)
"
"
100 to 1,000 workers,
(iii)
"
"
1,001 to 10,000 workers,
(iv)
"
"
10,001 to 100,000 workers,
(v)
"
"
more than 100,000 workers.
[438] A distinction should also be made between workers who
are members of the organisation which is a party to the agreement
and other workers to whom the agreement applies in practice.

STATISTICS OF INDUSTRIAL DISPUTES
32. Resolutions of the Third International Conference of
Labour Statisticians, Geneva, October 1926
[439] In each country statistics of industrial disputes should
be compiled according to the following general principles:
\j440] (1) The basic unit—the case of dispute—should be defined as a temporary stoppage of work wilfully effected by a group
of workers or by one or more employers with a view to enforcing
a demand. Disputes affecting several establishments should be
considered as one case if they are organised or directed by one
person or organisation.
[44f\ It is desirable to make, as far as possible, a distinction
between strikes and lockouts.
\_442~\ (2) The statistics should relate to disputes beginning in
the period under review, and also, but separately, to those continuing from the previous period. The total of these two groups of
disputes represents the number of disputes in existence during
the period under review.
\_443~\ (3) The importance of the dispute should be measured
by ascertaining the number of establishments and of workers
affected by the dispute, the duration of the dispute, and the number
of man-days lost on account of the dispute.
[444] The number of establishments affected should be based
on the technical unit as defined in censuses of industries.
[445"2 The number of workers affected should be based on the
vacancies caused in the establishments affected by the dispute, and
calculated by taking an average of the number of vacancies each
day during which the dispute lasted, or, where this is not practicable,
by taking an average of the number of vacancies recorded at weekly
intervals.
\jf4<f\ The duration of the dispute should be expressed as the
number of working days from the date on which the dispute began
in the first establishment affected to the date on which it terminated
in the last one.
\J44f] The number of man-days lost should be based on the

STATISTICS OF INDUSTRIAL DISPUTES

139

number of vacancies caused by the dispute during each day of
the dispute, or, where this is not practicable, by multiplying the
number of days for which the dispute lasted by the average number
of vacancies recorded at weekly intervals.
[448'] (4) The disputes, together with the number of establishments and workers affected, the duration of disputes, and the
number of man-days lost, should be classified according to their
principal characteristics, namely:
[449] A. The Matter in Dispute. The principal criterion should
be the relation of the dispute to collective bargaining as follows:
(a)

Disputes related to collective bargaining:
(i) concerning trade unionism or refusal to
conclude a collective agreement;
(ii) concerning conditions of employment:
(1)
(2)
(3)
(4)

(b)

wages,
hours of labour,
engagement or dismissal of workers,
others.

Disputes not related to collective bargaining:
(i)
(ii)
(iii)

sympathetic disputes;
political disputes;
others.

\^450] Groups (a) (ii) may also be classified into disputes concerning general terms of future employment, and disputes concerning the interpretation of existing terms of employment.
[451] B. The Result of the Disputes. Disputes should be classified according to their general result from the point of view of
workers involved, as follows:
(a) disputes where the workers' demands have been entirely
accepted ;
(ò) disputes where the workers' demands have been
partially accepted ;
(c) disputes where the workers' demands have been rejected ;
(d) disputes where the employers' demands have been
entirely accepted;
(e) disputes where the employers' demands have been
partially accepted;
(J) disputes where the employers' demands have been
rejected;
(g) disputes with indeterminate or unknown result.

140

INTERNATIONAL STANDARDISATION OF LABOUR STATISTICS

[452] C. The Method of Settlement of the Disputes.
Disputes
should be classified according to the method of their settlement on
the following lines:
(a) disputes settled by direct negotiations between the
two parties;
(ò) disputes settled by the medium of a third party:
(i) through voluntary conciliation accepted
by the parties to the dispute,
(ii) through compulsory conciliation imposed
by the law,
(iii) through voluntary arbitration,
(iv) through compulsory arbitration;
(c)

Disputes terminated without successful negotiations.

[453] D. The Industries Affected. Disputes should be classified according to the principal industrial groups. The classification
may be left to the domain of national statistics.
[454] E. The Extent of the Disputes.
Disputes should be
classified according to the number of establishments affected as
follows :
(a)
(6)
(c)
(d)
{e)

disputes affecting one establishment,
"
"
2 to 10 establishments,
"
"
11 to 20 establishments,
"
"
21 to 100 establishments,
"
"
more than 100 establishments.

[455] F. The Importance of Disputes. Disputes should be
classified according to their importance, taking as criteria the
number of workers affected, the duration of the disputes, and the
number of man-days lost.
[456] (a) The classification of^disputes by the number
of workers affected should be as follows:
(i)
(ii)
(iii)
(iv)
(v)
(vi)

disputes affecting less than 10 workers,
"
"
10 to 100 workers,
"
"
101 to 1,000 workers,
"
"
1,001 to 10,000
workers,
"
"
10,001 to 100,000
workers,
"
"
over 100,000 workers.

STATISTICS OF INDUSTRIAL DISPUTES

141

[457} (b) The classification of disputes by their duration
should be as follows:
(i)

GO
(iii)
(iv)
(v)
(vi)

disputes lasting less than 2 working days,
from 2 to 10 working days,
11 to 50 working
days,
u
<<
51 to 100 working
days,
101 to 300 working
days,
over 300 working
days.
t(

[458} (c) Disputes should be classified by the number of
man-days lost as follows:

CO
(ii)
(iii)
(iv)

(v)

(vi)

disputes involving the loss of less than
20 man-days,
a
of 20 to 1,000
man-days,
tt
tt
tt
of 1,001 to 50,000
man-days,
tt
tt
tt
of
50,001
to
1,000,000 mandays,
tt
tt
a
of 1,000,001 to
30,000,000 mandays,
tt
tt
of over 30,000,000
man-days.
it

«f

[459} G. Amount of Wages Lost. The wage loss caused by
each dispute may be estimated by multiplying the number of mandays lost by the average daily wages of the workers affected. The
classification of disputes according to the amount of wages lost may
be left to the domain of national statistics.
[460} (5) For the purpose of comparing the risk of industrial
disputes between different industries and countries, frequency and
severity rates of disputes should be calculated.
• [461} The exposure to risk should be expressed in terms of the
number of full-time workers, obtained by dividing the total number
of days worked during the period by the total number of normal
working days in the same period.
[462} The frequency rate should show the number of disputes

142

INTERNATIONAL STANDARDISATION OF LABOUR STATISTICS

per 100,000 full-time workers. The severity rate should show the
number of man-days lost per 10,000 full-time workers.
{463} (6) The above-mentioned data should be published at
least yearly. It is desirable, however, that preliminary figures
dealing more particularly with the number of disputes should be
published monthly.

XI
STATISTICS OF MIGRATION
33. Resolutions of the International Conference of Migration
Statisticians, Geneva, 3-7 October 1932
I
[464] The Conference considers that in view of the fact that
migration statistics in general are, from the international point of
view, still in a very imperfect state, efforts should be made in the
various countries to arrive gradually at the following desiderata:
[465'] 1. In principle, every act of removal from one country
to another for a certain length of time should be included in the
statistics of migration, with the exception of tourist traffic.
[466] 2. When the removal is for one year or more the migration should be regarded as permanent migration.
[467] 3. When the removal is for less than a year the migration should be regarded as temporary, frontier traffic being excluded.
[468] 4. It is desirable to distinguish statistics of permanent
migration from those of temporary migration. When the distinction cannot be made on the basis of definite information it should
be made by presumption.
[469] 5. When it is possible to distinguish between temporary
migration according to its duration it is desirable that this should
be done.
[470] 6. Seasonal migration should be included in the statistics
of temporary migration. It is desirable that it should also be shown
in separate tables.
[471] 7. It is desirable that each country should compile
statistics of emigration and statistics of immigration. Both should,
if possible, include nationals as well as aliens.
[472] 8. I t is desirable that countries of transit should, if
possible, draw up separate statistics of transmigrants.
[473] 9. Although migration movements between territories
situated in different continents but coming under the same sover-

144

INTERNATIONA!, STANDARDISATION OF LABOUR STATISTICS

eignty are from the legal point of view of an internal character,
it is nevertheless desirable that they should be given in the statistics,
by way of addition, side by side with international migration
movements.
[474] 10. It is desirable that each country should state as
precisely as possible the definitions on which its statistics are based,
so that the International Labour Office may be able to indicate
to what extent its definitions differ from those recommended by
the Conference.
[475] 11. In cases where national methods are not such as to
make it possible to realise the desiderata stated above, certain
omissions might be remedied by collaboration between the statistical services of two or more different countries.
II
[476] The methods by which migration statistics are compiled
in different countries depend on the regulations to which migrants
are subject. These methods accordingly vary very considerably
from one country to another.
[477] The Conference notes that none of the methods are
entirely satisfactory or make it possible to observe the phenomenon
of migration completely in all its aspects. The Conference does
not think it desirable to make a choice between these various
methods and to recommend one rather than another. Some, however, are clearly inadequate and should only be regarded as substitutes to which recourse is had in the absence of anything better.
These include the statistics of information offices for emigrants,
statistics of passport visas, statistics of steamship passenger contracts, statistics of the recruiting and placing of migrant workers,
and statistics of employment permits issued to foreign workers.
The same applies in a less degree to passport statistics, which
were of a certain importance some years ago but which are now
not so much used.
[478] The Conference gave its attention in particular to port
statistics, statistics of declarations of residence, statistics of frontier
control, anc statistics of coupons detached from certain documents,
and makes the following observations on these classes of statistics.
A.—Port Statistics
[479] 1. In order to make it possible to collect migration
statistics, it is necessary that the passenger lists of seagoing vessels
should mention the reason why each passenger makes the journey
or whether or not he is a migrant.

STATISTICS OF MIGRATION

145

[480] 2. Shipping companies or the master of the ship should
be required to supply the competent statistical authorities with a
copy of the passenger lists.
[481'} 3. Before landing the master should prepare lists of
immigrants in the forms prescribed by the countries of immigration.
B.—Statistics of Declarations of Residence
[482~] 1. In countries which record the changes of residence of
their inhabitants, the declarations which are required to be made
on this subject and which are recorded by the local authorities
make it possible to establish statistics of migration, particularly of
permanent migration.
[483"] 2. It would be desirable that measures should be taken
if possible to make a distinction under this system between temporary and permanent migration.
C.—Statistics of Frontier Control
[484~\ In those countries where a developed system for the
control and supervision of passenger traffic by ports or across
frontiers is in existence, migration statistics may be based on the
returns obtained from the authorities responsible for this supervision.
[485] It is desirable that in the case of migrants individual
questionnaires stating the reason for the journey should be filled up.
D.—Statistics of Coupons Detachable from Certain Documents
[486] 1. In the case of countries whose nationals are required
to present special documents when crossing the frontier, such
documents may serve as a basis for migration statistics.
[487] 2. Such documents should be issued to all migrants, if
possible free of charge, and contain detachable coupons. The
coupons, duly filled up, should be detached when the migrants
leave or enter the country. Nationals of the country of emigration
residing in a foreign country should be able to obtain such documents from the consular authorities.
[488] 3. A distinction should be made between permanent
and temporary migration either when these documents are issued
or when the coupons are detached.
Ill
[489] The Conference also discussed the possibility of establishing a uniform international system. I t makes the following
observations on this point.

146

INTERNATIONAL STANDARDISATION OP LABOUR STATISTICS

[4902 1. With a view to obtaining complete and uniform
statistics of international migration, it would be desirable that all
countries should employ the same system based on the issue of an
identical individual document.
[4912 2. The realisation of such a system would obviously
involve serious difficulties and could not be effected immediately.
Further study appears necessary and the Conference suggests
scientific collaboration between the competent departments of
the International Labour Office and those of the International
Statistical Institute, which has made proposals on this subject.
IV
£492^ With a view to the classification of the statistical data,
it is desirable that so far as possible information should be supplied
by the various countries in such a way that the international tables
can include the following subdivisions:
[4932 1. Distinction between continental and inter-continental
migration. For this purpose it is desirable that emigrants should
be classified as completely as possible according to the country of
destination and immigrants according to the country from which
they originally started. This distinction might be based in principle
on the geographical list of countries adopted by the Statistical
Year-Book of the League of Nations in the table "Area and population".
[4942 2. Distinction between nationals and aliens. 1
[4952 3. As an additional or subsidiary distinction to that
given above, a distinction may be made according to language,
ethnic origin or the country of birth of the migrant.
[4962 4. Distinction according to sex and by quinquennial age
groups (under five, five to under ten, etc.).
[4972 5. It is also desirable that migrants should be classified
according to industrial or occupational groups.
[4982 It is suggested that this classification should be made
only in the case of migrants of over fifteen years of age and that the
following groups should be adopted, a distinction being made
between the sexes:
I.
II.
III.
IV.
V.

Agriculture.
Mines and quarries.
Manufacturing industries.
Transport and communications.
Other industries.

1
One member of the Conference asked that naturalised persons should be
distinguished from other nationals.

STATISTICS OP MIGRATION

VI.
VII.
VIII.
IX.

147

Commerce and finance.
Domestic and personal service.
Liberal professions.
Persons without occupation or of unknown occupation.

[_499~\ I t might also be desirable to include children under
fifteen in this classification under a separate heading, especially
in those cases where such information cannot be derived from the
classification mentioned in point 4.
[500} 6. I t is desirable that migrants should also be classified
according to their industrial or social status:
(a) Independent persons (employers and persons working
on their own account).
(£>) Employed persons.

(1)
(2)
(3)
Others.
Members

Directors, managers, etc.
Manual workers.
Salaried employees.
of the migrant's family without occupation.

[501} It is also desirable that this classification should be combined with the classification mentioned under 5, in order to show
so far as possible the status of migrants in the various occupational
groups.
[,502} 7. It would be desirable that an attempt should be made
to distinguish the following classes among migrants, a distinction
being also made according to sex:
(a) Married, widowed or divorced, unmarried.
(6) Persons emigrating alone, persons emigrating with
their families.
[503} The following distinctions might also be made for certain
countries specially concerned:
[504} (c) Persons travelling at their own expense, persons
travelling with financial assistance of private individuals,
persons travelling with financial assistance of public authorities
or private institutions.
£505} (d) Persons possessing or not possessing a contract
of employment.
[506} (e) Persons able or unable to write (it being understood that any language may be used and that the distinction
should only apply to migrants aged ten years or more).

148

INTERNATIONAL STANDARDISATION OF LABOUR STATISTICS

rj5073 8. In the case of internal migration between territories
coming under the same sovereignty but situated in different continents, it is desirable to make a distinction according to the continent of origin of the migrant.
[50c?] 9. The statistics should be established periodically, not
less than once a year and, if possible, for the calendar year. In
countries where some other period is used, it would be desirable
from the international point of view also to follow the calendar
year, or some shorter period which makes it possible to arrive at
the calendar year, and to make the geographical distribution
mentioned above (point 1).

APPENDIX
K E Y TO T O P I C S C O N S I D E R E D

AT E A C H

CONFERENCE OF LABOUR

International Labour

INTERNATIONAL

STATISTICIANS

Organisation.

F i r s t I n t e r n a t i o n a l Conference of L a b o u r Statisticians
( I n t e r n a t i o n a l Conference on L a b o u r Statistics), G e n e v a ,
October 1923.
A. Classification of Industries and Occupations
B. Statistics of Wages and Hours of Work
C. Statistics of Industrial Accidents

49
70
119

Second I n t e r n a t i o n a l Conference of L a b o u r
G e n e v a , April 1925.
A. Cost-of-Living Index Numbers
B. Unemployment Statistics
C. International Comparisons of Real Wages
D. Classification of Industries

Statisticians,
98
67
101
SO

T h i r d I n t e r n a t i o n a l Conference of L a b o u r
G e n e v a , October 1926.
A. Methods of Family Budget Enquiries
B. Statistics of Collective Agreements
C. Statistics of Industrial Disputes
D. Classification of Industries

Statisticians,

F o u r t h I n t e r n a t i o n a l Conference of L a b o u r
G e n e v a , M a y 1931.
International Comparisons of Real Wages

Statisticians,

116
134
138
SO

Fifth I n t e r n a t i o n a l Conference of L a b o u r Statisticians,
G e n e v a , Sept.-Oct. 1937.
Convention concerning Statistics of Wages and Hours of W o r k . . . .
(The text given is that of the Convention adopted at the 24th
Session of the International Labour Conference, June 1938.)
Proposed Recommendation

104

85
95

I.L.O. C o m m i t t e e of Statistical E x p e r t s .
F i r s t Session, 12-15 D e c e m b e r 1933.
International Comparisons of Food Costs
International Comparisons of Rents
Family Budgets
Elements of Remuneration other than Wage6

109
Ill
118
107

150

INTERNATIONAL STANDARDISATION OF LABOUR STATISTICS

Second Session, 16-18 December 1935.
International Comparisons of Rents

112

Third Session, 3—7 October 1937.
International Comparisons of Rents

114

24th Session of the International Labour Conference, June
1938.
Convention concerning Statistics of Wages and Hours of Work,
1938
Proposals calling for further action

85
96

L e a g u e of Nations.
League of Nations Committee of Statistical Experts.
Excerpts from Report on Statistics of the Gainfully Occupied
Population
Excerpts from Report on Statistics of Housing

52
128

International Conference relating to Economic Statistics
convened by the League of Nations.
Extract from International Convention relating to Economic
Statistics, 14 Dec. 1928
Recommendation relating to Index Numbers

100
100

Other International Conferences.
International Conferences of Statisticians, convoked by the
Social Science Research Council of the United States
of America.
First International Conference, January 1929.
Index Numbers of Wages
International Comparisons of Wages

71
101

Second International Conference, May 1930.
Index Numbers of Wages
International Comparisons of Cost of Living

83
108

Committee of Statistical Experts convoked by the International Union of Towns, May 1928.
Housing Statistics

122

International Conference on Migration Statistics, October
1932.
Statistics of Migration

143

INDEX

INDEX
Accident insurance, notes for industrial
accident statistics, 348
Accidents: see Industrial accidents
Adult male unit: see also Consumption
unit and Average size of family
earnings per family, 118
Age classification:
gainfully occupied, 11 ; 68-70, 72
migrants, 496, 498-499, 506
Agricultural workers:
' family living studies, 32
and wage index numbers, 112
Agriculture: see also Economic Activity,
Gainfully occupied, List of industries
wages, 18; convention, 200-202
Allowances: see also Family allowances,
Remuneration
agriculture, 26; convention, 201
in kind, 25, 26; convention, 172
special, taken into account in earnings, 129
supplementary to wages, 287-290
and wage rates, convention, 194-195
Amenities, of dwelling: see Conveniences, Housing, Standard dwelling units
Base period for index numbers: see also
Index numbers
Committee of Statistical Experts,
L. of N., 252
cost of living, 23; uniform, 247, 252
economic series, International Convention relating to Economic
Statistics, 252
International Institute of Statistics,
252
I.L.O., 252
wage index, 119:
same as for index of production,
146
uniform for all countries, 120 ^ Î P !
Bathrooms: see also
Conveniences,
Housing, Rooms, Standard dwelling
units, 315
classification of dwellings, 30; 366,
404
and definition of room, 359, 389
Benefits, total paid in year, 75
Branches of economic activity: see
Economic activity

British Ministry of Labour: see Great
Britain, Ministry of Labour
Buildings: see also Housing censuses,
Dwellings
habitable:
classification, 367-369
definition, 356
yearly data needed, 382-383
new, statistics, 381-385
Cause, classification of industrial accidents, 341
Censuses: see also Housing census
gainfully occupied, 11, 14; 10-12
production, and wage date, 145
unemployment, 12-13; 81; needed,
23; by sex, 82
wages, 19; 91
Chain method, wage indices, 116
Child labour: see Age classification,
Employment, Young persons
Classification of industries and occupations: see also Industries, List of
industries
Committee of Statistical Experts,
L. of N., 10; 10-73; occupations,
9-11
International Conferences of Labour
Statisticians, 9; 1-8
Climate, and international comparisons
of real wages, 254
Coal mining:
industrial accidents, report, 33-34
report of Committee of Experts in
Safety in Coal Mines, 34
wages, 18, 26; in general index, 156
Coats, R. H., 10
Collective agreements: 37; 421-438
agenda of Third Conference of Labour
Statisticians, 3
classification:
duration, 432
industrial importance, 435-437
industries, 434
method of conclusion, 433
nature of contracting parties, 428
scope of application, 429, 438
subjects, 430, 431
definition, 423
frequency of compilation, 424
importance, 425-426

Numerical references in italic type are to the numbers of the paragraphs of the resolutions as
given in Part I I ; other references are to page numbers.

154

INTERNATIONAL STANDARDISATION OP LABOUR STATISTICS

Collective agreements (cant.):
International Conference of Labour
Statisticians, Third, 37; 421-438
statistics, report on methods, 37
Collective bargaining, industrial disputes, 449
Committee o:: Experts in Housing
Statistics: see International
Union
of Towns
Committee of Experts in Safety in
Coal Mining, Report, 34
Committee of Statistical Experts of
I.L.O.: see also Appendix, p. 151; 6
appointment, 2, 4
duties, 2, 4
functions, 286
meetings, agenda, 5
recommendations :
cost of living, 27
family living studies, 338
food costs, comparisons, 29; 298309
real wages, comparisons, 24
remuneration, other than wages,
26; 287-290
rents, 30; 310-325
recommended, Fourth Conference,
286
wages, convention, 16
Committee of Statistical Experts of the
League of Nations: see also Appendix, p. 151; 1,4, 10
age classification of gainfully occupied, 11
base period:
economic statistics, 23
index numbers, 252
gainfully occupied, 14; 10-73
housing statistics, report, 35; 388420
housing studies, mentioned, 325
Commodities, minimum list for international trade statistics, 57
Conferences of American States Members of the I.L.O.:
Havana, 1940, 7
Santiago de Chile, 1936, 7
Conferences of Labour Statisticians:
see International Conference of Labour
Statisticians
Conferences of Statisticians called by
the Social Science Research Council:
see Social Science Research Council
Conferences on Rural Hygiene of the
Health Organisation of the League of
Nations, 133
Consumption: see also Cost of living,
Family living studies, Retail prices;
aggregate, use of weights for cost of
living, index numbers, 242

Consumption {coni.):
family budget enquiries, use in cost
of living indices, 241
habits, comparable, regional groups,
302; and international comparisons, 280-282
industrial workers, in cost of living
index, 238
theoretical budgets, in index numbers, 243
working class, retail prices, 93
Consumption unit:
average size of family, earnings per
family, 118
family living studies, 334
Consumption unit scales:
food, 334
general, 334
Contract of employment, and migration, 505
Conveniencies: see also Housing, Standard dwelling units
in dwelling, 366, 402, 404-405
in standard dwelling unit, 30
Convention concerning Statistics of
Wages and Hours of Work: see also
Hours of Work, Wages, 2, 5, 16-17,
20, 21, 26; 163-218
on agenda, Fifth Conference of
Labour Statisticians, 3
authentic texts, 218
declaration, excluding parts, 165
denunciation, 213-214, 216
exclusion of certain areas, by Members ratifying, 203-205
in force, 209-211, 213-214, 216-217
obligations assumed, 164
proposals for improving statistics,
206-207
ratifications, 208
recommended by Fourth Conference
of Labour Statisticians, 285
reports t o I.L.O., 164, 191, 204
reservations, 166-167
resolution of Havana Conference,
1940, 7
revision, 215-217
Convention relating to Economic Statistics (of the League of Nations),
1,4, 10:
cost of living, 249-251
index number, base period, 252
Cost of labour, subjective, rejected as
basis for index numbers, 104
Cost of living: see also Index numbers,
International
comparisons,
Retail
prices
agenda, 2nd Conference of Labour
Statisticians, 2
elements, 235
index numbers:
and direct taxes, 237

Numerical references in italic type are to the numbers of the paragraphs of the resolutions as
given in Part I I ; Dther references are to page numbers.

INDEX
Cost of living (cont.):
index numbers (cont.)
frequency, 245, 249
groups, separate, 236
methods, report, I.L.O., 22
national, by combining and weighting separate indices, 240
national, industrial workers, 238
principles, etc., stated, 248
reference to sources, and methods,
251
separate for classes and areas with
marked differences in economic
conditions, 238, 239
towns covered, 250
use in index of real wages, 103
weights: aggregate consumption,
242
based on consumption of class
or area, 238-239
for combining groups, 236
familv living studies, 241, 336,
337
use of theoretical budgets, 243
international comparisons:
agenda of International Labour
Conference, 285
Committee of Statistical Experts,
I.L.O., 27; 298-325
Detroit and 14 European towns,
28-29
International
Conference
of
Labour Statisticians, Fourth,
27* 272-286
methods, 280-284, 293-297
Social Science Research Council,
Conferences, 27; 291-297
International Conferences of Labour
Statisticians, 22-23, 27; 235-248,
272-286
related to actual earnings, 20
representative items, 235
statistics, 22-25, 27-30:
and Committee of Statistical
Experts (I.L.O.), 286
expanding scope, 41-45
Countries, list, Statistical Year Book
of League of Nations, in classifying
migrants, 493
Country of birth: see also Migration,
migrants, 495
Country of destination : see also Migration, migrants, 493
Country of origin: see also Migration,
migrants, 493
Density of occupancy: see also Dwellings, Housing
classification, 370-374, 408-414
definitions, 372, 408, 414
Detroit, U.S.A., comparison of cost of
living in, and 14 European towns,
28-29

155

Dietary surveys, report, 32
Disability, industrial accidents, classification, 342-345
Dwellings: see also Buildings, Density
of occupancy, Housing, Housing census, Rents
classification, 363-366, 371, 375, 378,
400-418:
conveniences, 404
number of households occupying,
375
purpose, 364
recommended, Committee of Stat.
Exp. I.L.O., 314-315
rent paid, 378
size, 365
by size and number of occupants,
371
conveniences, 30; 366, 402, 404-405
definition, 323, 324, 354-355, 391
density of occupancy, 370-374, 408414
number, in classification of habitable
buildings, 368-369
occupation by family households,
classification, 375, 394, 406
and rent restrictions, 380
size, 315:
classification of dwelling, 365, 371,
403
in classification of occupants, 369370
density of occupancy, 370-374,
410-414
in housing censuses, 401
vacant dwellings, 387
standard units, 273, 314-315, 322
statistics, 35-36
tenure, housing census, 418
vacant, 386; classification and definition, 387, 407
variations in total number, 381-385
working class, predominant types,
312-313, 316; rents, 320
Earnings: see also Income, Index numbers, International comparisons, Salaries, Wage rates, Wages
actual, 89:
in economic forecasting, 122
index numbers, 19; 118
industries, 89
representative employers, 89
typical categories of workers, 92
annual, industries, mining and manufacturing, and other branches, 226:
by industries, mining and manufacturing, 233
measuring changes in standard of
living, 108
definition, convention, 171, 174; uniformity needed,257
distribution, 224 ; occupations in each
industry, 230

Numerical references in italic type are to the numbers of the paragraphs of the resolutions as
given in Part I I ; other references are to page numbers.

156

INTERNATIONAL STANDARDISATION OF LABOUR STATISTICS

Earnings (cont.):
per family:
index numbers, 118
and international comparisons of
real wages, 254
per family member, index number,
118
frequency of compilation, convention, 176-177, 179, 223
per hour:
comparable character of work and
efficiency, 128-137
categories of workers, 130-131
choice of data, 128-129
frequency, 133
industries covered, 130-131
localities, 132
hourly, 161
index numbers, 90, 179
industries, 89; mining and manufacturing, convention,
168-170,
176-178
normal, typical categories of workers, 92
occupations in each industry, 223,
229
overtime, 128-129
from payrolls, in wage census, 91
piece work, typical categories of
workers, 92
sex, 177; and age groups, 224
statement, annual, by I.L.O., on
defects in information, 258
per time unit, for international comparison, 268
for transport, commerce, administration, recommended, 220
and wage data, special enquiries
t o show differences, 257
and wage rates, 234:
comparisons, 270
divergencies, 159
both needed, 159, 256
representative of, 109
weekly, 161
per worker, index numbers, 118
young persons, convention, 177
Economic activity, branches of: see
also Industry,
Gainfully
occupied,
List of industries, Occupations
in classification of gainfully occupied,
24-57:
age, sex, marital condition, 68-72
industrial status, 66
occupations, 67
minimum list, 73
principles of grouping, 52-57
Economic, Financial and
Transit
Department of the League of
Nations, 41
Economic forecasting :
index numbers of wages, 102, 103,
122-127:

Economic forecasting {cont.):
index numbers of wages (cont.)
choice of data, 122
choice of establishments, 124
choice of localities, 125
frequency of compilation, 126
industries and categories of workers, 123
Economic s t a t u s : see Industrial status
Emigration: see Migration
Employers: see also Gainfully occupied
classification, gainfully
occupied,
59, 63
Employment: see also Gainfully occupied, Hours, Unemployment
agenda, 2nd Conference, Labour
Statisticians, 2
and collective agreements, 426
estimate of insured workpeople, 75
full time workers, for accident rates,
353
International Conference of Labour
Statisticians, 14; 80, 85
international index, 15
short-time, statistics, separate from
whole-time unemployment, 85
statistics, 14-15:
expanding scope, 41-45
representative employers, published periodically, 80
Employment exchanges: see also Employment,
Unemployment
registration of unemployed, 78
statistics, 78, 79:
industries, 82
occupations, 82
sex, 82
unskilled,
distinguished
from
others, 79
vacancies, 78
Equipment, household: see Conveniences
Establishment: see also Gainfully occupied, Industries
accessory branches, 42
belonging t o public authorities, classification, 48
classification:
difficulties, 38-48
guiding principles, 39-48
manufacturing industries, 3
covered, collective agreements, 427,
436
definition, for classification of gainfully occupied, 30-48
divided between two or more branches of economic activity, 44-47
industrial disputes, 444, 448, 454
legal and financial organisation,
concept, 36, 48
manufacturing and retail sale, 43

Numerical references in italic type are to the numbers of the paragraphs of the resolutions as
given in Part I I ; other references are to page numbers-

INDEX
Establishment (core/.):
place of work concept, 34-35, 37-47
adopted b y Committee of Statistical Experts, L. of N., 37
technique of work concept, 33
Expenditure: see also Family living
studies
family living studies, 330, 332, 335:
grouping, 335
Exposure to risk: see also Employment,
Industrial accidents, Industrial disputes
for accident rates, 14; 350-353
industrial disputes, 461
Family allowances: see also Remuneration, 25, 26
convention, 194
and earnings, convention, 173
and international comparisons of
real wages, 271
real wage comparisons, 279
statistics, 287-290:
frequency of compilation, convention, 197
supplementing average earnings,
convention, 173
Family budget studies: see Family
living studies
Family earnings: see Earnings, family
Family household: see also Density of
occupation,
Dwellings,
Household,
Housing
definition, 362:
Committee of Statistical Experts,
L. of N., 392-395
density of occupation, 413
dwellings, 364, 406
Family income: see also
Earnings,
Income
and housing censuses, 419
Family living studies: see also Cost of
living, Dietary surveys.
Housing,
and names of items of income and
expenditure, 27; 326-338
agenda, 3rd Conference, Labour
Statisticians, 3
agricultural workers, 32
classifications, 332-334
Committee of Statistical Experts,
I.L.O., 338
different districts, etc., 296
food consumption, 334-335
frequency, 326
International Conference of Labour
Statisticians, Third, 3 1 ; 326-337
methods, 31-32; 327-337:
report, 31-32
new studies desired, 246; recommended, 291
period covered, 328, 329
quantities consumed, 299

157

Family living studies (cont.):
recommended, 274
sampling methods, 327
simultaneous, recommended, 338
statistics, expanding scope, 42-43
of unemployed, 32
uniform basis, recommended, 338
weights for cost-of-living indexes,
336-337
Year Book of Labour Statistics, 32,
42-43
Farmers: see also Agricultural workers,
Agriculture, Gainfully occupied, List
of industries
gainfully occupied, 16
Ferenczi, Imre, 40
Figuerola, José, 7
Filene, Edward A, 28
Fisher (Irving), ideal formula, 29
Floor-space: see also Rent
defined, 317, 323-324
rents, 311, 317, 319, 324.
Food: see also Cost of living, Dietary
surveys, Family living studies, Retail
prices
consumption unit scale, 334
nutritive value, family living studies, 335
Food cost:
international comparisons, 27-30;
298-309:
Committee of Statistical Experts,
I.L.O., 29; 298-309
International
Conference
of
Labour Statisticians, Fourth,
29; 272, 275, 280-284
methods, 300-309
Report, 29
Food habits, surveys, 32
Food prices: see Retail prices
Ford Motor Company, Ltd., 28-29
Frequency of compilation: see also
names of topics
collective agreements, 424
cost of living, index numbers, 245249:
and prices changes, 245
dwellings, changes, 382-385:
vacant, 386-387
earnings, 86, 89, 92, 93:
annual, by industries, 226-233
convention, 176-177, 179
occupations in each industry, 223,
229
per hour, 133
employment, 80
family allowances, convention, 194
family living studies, 326
holidays with pay, convention, 194
hours, 86, 89, 92, 93; salaried employees, 222

Numerical references in italic type are to the numbers of the paragraphs of the resolutions as
given in Part I I ; other references are to page numbers.

158

INTERNATIONAL STANDARDISATION OF LABOUR STATISTICS

Frequency of compilation (cont.):
hours, actual:
convention, 176, 177
occupations, 223
occupations in each industry, 229
hours, normal, convention, 187
index numbsrs, see under names of
topics, e.g., Cost of living, Wages
industrial accident rates, 350-351,353
industrial disputes, 463
migration statistics, 508
occupations in each industry, 223,229
overtime rates, convention, 194
real wages, 93
remuneration other than wages, 290:
allocation in indexes, 117
convention, 194
rents, 275, 320; floor space, 319
retail prices, 275
salaries, 222
standard o: living (real wages),
114-117
total wages and salaries, 149; industries, 225, 231
unemployment, 75, 77, 78
wage censuses, 91
wage index:
allocation for supplementary items,
117
and economic stability, 114
facility in collecting, 115
period: one month, 116
wage rates, 87, 88, 92,93; convention,
187
wages, 160:
agricultura, convention, 201
in chief towns, 275
economic forecasting, 126
index numbers for economic forecasting, 126
monthly, 160
occupations, 264
transport, commerce, administration, 222
Frequency rate, industrial accidents,
350-351, 353; industrial disputes, 462
Fuel prices: see Retail prices
Full time workers, for accident rates,
353; for industrial disputes, 461
Fuss, Henri, 13
Gainfully occupied: see also Unemployed, 9-11
categories for international comparisons, 13-19
census, 10-12
classification :
age, 68-70, 72
branches of economic activity,
24-57, 73; and industrial status,
jointly, 26:
minimum list, 73
purpose, 28
principle, 28-29

Gainfully occupied {cont.):
classification {cont.)
industrial status, 24, 58-66
industry, 1-9, 24-57, 73
marital status, 68, 71
occupation, individual, 1, 24, 67
occupation, not attempted, 67
principles, 24
sex, 68-72
Committee of Statistical Experts,
L. of N., Report, 4, 11; 10-73
definition, 10-12
directors and managers, 62, 64
employers, 59, 63
exclusions, 17-19
International Conference of Labour
Statisticians, 9-10; 1-8
members of families, 15
military service, and last occupation, 22
salaried employees, 62, 65
scope, expansion of statistics, 42-43
unpaid family workers, 60, 61
wage earners, 62, 65
Giusti, Ugo, 22
Great Britain, Ministry of Labour:
enquiry into real wages, 253
real wage comparisons, 24,25
Group basket method, food cost comparisons, 29
Guye, R., 30, 36, 132
Habitable buildings: see also Buildings,
Dwellings, Housing
classification, 367-369
definition, 356
yearly data needed, 382-383
Habitable rooms: see Rooms
Health Organisation of League of
Nations, 6, 133
Heating: see also Cost of living, Family
living studies, Rents, Retail prices
classification of dwellings, 405
Holidays with pay: see also Remuneration, 25, 26; 287-290
coal mines, 27
frequency of compilation, convention, 194
and real wage comparisons, 279
Hours: see also Employment
actual, 86:
annual statement by I.L.O., 258
categories of workers, 92
frequency of compilation, convention, 176, 177
industries, 89
mining and manufacturing industries, convention, 168-170, 176178
occupations, 223, 229
occupations in each industry, 229

Numerical references in italic type are to the lumbers of the paragraphs of the resolutions as
given in Part II ; other references are to page nun jers.

INDEX
Hours (coni.):
actual (cont.)
representative employers, 89
same unit as for average earnings,
convention, 175
transport, commerce, administration, 219-220
agenda at First Conference of Labour
Statisticians, 2
coal mining, 18
convention, 2 1 ; 163-2IS
full time, annual statement by I.L.O.,
258
typical categories of workers, 92
International Conference of Labour
Statisticians, 20-21 ; 86-93, 219-228
normal, 86:
agriculture, convention, 202
changes, 88
classification, occupation and skill,
convention, 188-190
definition, convention, 186
frequency of compilation, convention, 187
index number, convention, 197
industries, mining and manufacturing, convention, 182-199
industries, representative selection, convention, 182
occupational classification, convention, 188
occupations and industries, convention, 187
salaried employees, recommendation, 221
and standard of living, 110
time period, same as for wage
rates, convention, 191
for transport, commerce administration, recommendation, 219220
types, convention, 183, 186
typical occupations, capital cities,
93
typical categories of time workers,
92
young persons, convention, 192
postal services, report, 21
statistics, 20-21:
expanding scope 42-45
supplementary resolutions, outside
agenda of International Labour
Conference, 1938, 95
textile industry, 18
Households: see also Density of occupancy, Dwellings, Family household,
Housing
classification, occupation of head of
household, 416
composition, family living studies,
331-332
definition, International Union of
Towns, 361-362:

159

Households (cont.):
definition (cont.)
Committee of Statistical Experts,
L. of N., 392-398
family, definition, 362, 392-395
non-family, definition, 362, 396-398
occupation of dwellings, 364
occupying a dwelling, 375
Housework, and gainfully occupied,
definition, 10
Housing: see also Density of occupancy,
Dwellings, Habitable buildings, Housing census, Rents, Vacant dwellings
accommodation, family living studies, 331
Committee of Statistical Experts,
L. of N., 4, 35; 388-420
International Union of Towns, Committee of Experts in Housing Statistics, 35-36; 354-387
standard units, 30; defined, 314-315:
recommended, 273
revised, 322
statistics, 35-36
Conference, International Union
of Towns, 1, 3
Housing census: see also Dwellings,
Housing, 35-36:
classifications, 416
geographical scope, 420
particulars to be covered, 320
recommendations of Committee of
Statistical Experts, L. of N., 388420
standardised, 325
at time of general population census,
325
Huber, Michel, 4, 5
Hudson, Manley O., 10
Identical establishments, wage indices,
116
Immigration: see Migralion
Income: see also names of elements of
income; definition, family living
studies, 332, 335
Index numbers: see also names of topics
cost of living: see also Cost of living
index numbers:
convention relating to economic
statistics, 249-251
and direct taxes, 237
frequency of compilation, 245, 249
groups, 236
methods, report, I.L.O., 22
national, by combining separate
indexes, 240
national, industrial workers, 238
principles, etc., stated, 248
references to source and methods,
251
separate for classes and areas with
marked differences in economic
conditions, 238, 239

Numerical references in italic type are to the numbers of the paragraphs of the resolutions as
given in Part I I ; other references are to page numbers.

160

INTERNATIONAL STANDARDISATION OP LABOUR STATISTICS

Index numbers {coni.):
cost of living {cant.)
towns covered, 250
use in index of real wages, 103
and wages;, 90
weights, 236:
aggregate consumption, 242
based on consumption of class
or area, 238, 239
family living studies, 241, 336,
337
use of theoretical budget, 243
earnings, 19; 90:
convention, 179-181
frequency of compilation, convention, 179
methods, convention, 181
earnings per hour:
averages for different categories,
135
comparable character of work and
efficiency, 128-137:
categories of workers, 130-131
choice of date, 128-129
frequency, 133
industries covered, 130-131
localities, 132
weights, 136-137:
effect of differences in different
countries, 136
economic statistics, Committee of
Statistical Experts, L. of N „ 252
hours, normal, convention, 197
labour cost per unit of production,
97, 98, 138-146
nominal wage rates, 90
purchasing power of wages, relative
levels, 277-278
real wages, 20; 90, 92; original data
to be published, 255
rents, 317, 318
wage rates, 19; convention, 196-199;
methods indicated, 199
wages, 19-20:
American report to Social Science
Research Council, 156
base period, post-war, common t o
all countries 120; same as for
index of production, 146
basic data to be published, 154
changes in standard of living, 108120:
categories of workers covered,
111-112
choice of districts, 113
choice of type of wage data,
108-110
frequency, 114-117
industries included, 111-112
method of compilation, 118-120
convention, 20; 179-181, 196-199
detailed indices b y sex, skill, and
industry, 155

Index numbers (cont.):
wages (cont.)
different industries or occupations,
121
elements to be considered, 105-106
forms, 153-156
general, 155
general for international comparisons, 156
International Conferences of Labour Statisticians, 19, 20; 90-92
International Labour Review, 20
method of construction, important
for international comparison, 152
notes and reservations published,
158
as percentage of national income,
102
publication of basic data, needed,
157
purposes, 94-98
separate industries and all industries, 100
skilled, 185
skilled, semi-skilled and unskilled,
100
Social Science Research Council,
Conferences, 19; 94-162
source references published, 158
variable weights, for different industries, 119
and base period choice, 119
wage earners as a whole, 111-112
weights, constant, 119:
problem, 100
simple or weighted average, 119
women, separate index, 162
women, whether included in general index, stated, 162
Industrial accidents: see also Industrial
diseases
agenda, First Conference, 2
cause classification, 341
classifications, 339-349
coal mining, report, 33-34
disability, classification, 342-345
general notes, 348
industry classification, 340
International Conference of Labour
Statisticians, First, 33; 339-353
location of injury, classification, 346
nature of injury, classification, 347
railways, report, 33, 34
rates, frequency, 350-351, 353
severity, 350, 352-353
reporting, 348
statistics, 33-34:
expanding scope, 42-43
methods, report, 33
report, 33-34
Industrial classification: see also Classification, Industries, List of industries, migrants, 497-499, 501

Numerical references in italic type are to the numbers of the paragraphs of the resolutions as
given in Part I I ; other references are to page numbers.

INDEX
Industrial comparisons, accident rates,
350
Industrial diseases: see also Industrial
accidents, 349
Industrial disputes, 439-463:
agenda, Third Conference Labour
Statisticians, 3
classification:
importance: duration 457; number
of man days lost, 458; number
of workers affected, 456
industries affected, 453
matter in dispute, 449, 450
number of establishments affected,
454
result, 451
strikes and lockouts, 441
time of beginning, 44Z
and collective agreements, 433
collective bargaining, 449
definition, 440, 442
duration, 443, 446, 457
establishments affected, 443-444, 454
frequency of compilation, 463
importance, duration, 443, 446, 457:
number of establishments, 443444, 454
number of workers, 443, 445, 456
International Conference of Labour
Statisticians, Third, 38; 439-463
method of settlement, 452
risk, frequency and severity rates,
460-462
statistics, expanding scope, 42-43:
report on methods, 38
time loss, 38
wage loss, 459
workers affected, 443, 445, 456
Industrial status:
gainfully occupied, 24, 26:
classification, 58-66
sex, age, marital status, 68-72
migrants, 500-501
Industries: see also Economic activity,
Gainfully occupied, List of industries,
Occupation
classification: see also Gainfully occupied, classification, 9-11
agenda, First Conference, 2
Second Conference, 2
collaboration with League of
Nations and International Statistical Institute, 8
consultations with statistical offices of different countries, 6
definitions in different countries, 5
employment exchange statistics, 82
industrial accidents, 340
industrial disputes, 453
I.L.O. report, 9
primary and secondary production, services, 1-9
scientific basis, 6
use in labour statistics, 1-8

161

Industries: {cont.):
definitions, 5
list:
detailed, note to, 73
minimum, 49-57, 73
provisional (I.L.O.), 4, 6, 7, 9
Institute of International Cooperation, 6
Inter-American Statistical Institute, 6
International comparisons: see also
Cost of living, International
indices,
Rents, Wage rates, Wages
and annual volume of labour statistics, 276
Committee of Statistical Experts,
(I.L.O.), 2
cost of living, 24-25, 27-30; 272-286,
291-297, 298-309, 310-325:
on agenda of International Labour
Conferences, 285
Committee of Statistical Experts,
I.L.O., 27; 298-325
I nternational Conference of Labour
Statisticians, Fourth, 27; 272-286
methods, 280-284, 294-297
Social Science Research Council
Conferences, 27; 291-297
earnings per time unit, 268
food costs, 27-30; 298-309:
Committee of Statistical Experts,
I.L.O., 29; 298-309
I nternational Conference of Labour
Statisticians, Fourth 29; 272,
275, 280-284
methods, 300-309
gainfully occupied, 13-19, 70; uniform principles, 40
industrial accidents, rates needed,
350
industries, facilitated by agreed list,
4-6,8
labour cost of production, 140
occupancy of dwelling, and definition of occupant, 408
real wages, 16, 20, 24-30; 253-297:
actual earnings of working class
family, in relation to physiological needs, 254
agenda, Fourth Conference, Labour Statisticians, 3
British Ministry of Labour, 253
chief cities, 93
Committee of Statistical Experts,
I.L.O., 24
elements, 279
and family allowances, 271
International Conference of Labour Statisticians, Fourth, 266286
methods, 280-284, 293-297
sources, methods, resolutions, to
be published, 254

Numerical references in italic type are to the numbers of the paragraphs of the resolutions as
given in Part I I ; other references are to page numbers.

162

INTERNATIONA!, STANDARDISATION OP LABOUR STATISTICS

International comparisons (conk):
International indices: see also Food
Costs, International
comparisons,
rents, 30; 310-325:
World indices, 30
Committee of Statistical Experts
food costs, 298
I.L.O., 30; 310-325
purchasing power of wages, 277-278
International Conference of Lareal wages, 25, 27; elements other
bour Statisticians, Fourth, 30;
than wages, 279
273, 275-277
standard units recommended, 273
International Institute of Agriculture:
total wages, proportion of national
represented on Committee of Statisincome, 150
tical Experts of the League of
unemployment statistics, 83
Nations, 4
wage rates, 86-91; time rates and
International Labour Conference, 1919
piece work basic time rates, 269
(First Session), Emigration Commiswages, 24-27:
sion, 39
adult male unit in earnings per
family, 118
International Labour Conference, 1922
on agenda of International Labour
(Fourth Session):
Conference, 285
recommendation on migration stacoal enquiry, 18
tistics, 39
Committee of Statistical Experts
recommendation, unemployment of
young persons, 12
(I.L.O.), 4, 24,
between countries or districts, 107
International Labour Conference, 1938
per hour of work, 99
(Twenty-Fourth Session):
index numbers, 156; construction
Convention concerning Statistics of
important, 152
Wages and Hours of Work, 3, 5, 6,
International
Conferences
of
17; 163-218
Labour Statisticians, 24, 26;
resolution referred to Member States,
253-255, 266-286
96; 229-234
occupations, 259-265
resolutions outside scope, 95
occupations, similar, 130-131
supplementary resolutions, wages
occupations, variations in numand hours of work, outside agenbers of workers, 134
da, 95
Social Science Research Council,
International Labour Office: see also
Conferences, 24, 26; 256-265
Convention concerning Statistics of
International Conference relating to
Wages and Hours of Work, InterEconomic Statistics: see also Appennational Labour Organisation and
dix; 4, 56; base period, 23; 249-252;
passim
gainful occupation, 24
base period for cost of living, agreeInternational Conference of Labour
ment of Governments in collaboraStatisticians: see also Appendix for
tion with I.L.O., 247
topics covered at each session, and
duties, 1
names of topics, 1-4, 8, 56, 100, 108;
international comparisons of real
First, 2, 9, 16-17, 19-22, 24, 56; 1-5,
wages, 253, 255 ; capital cities, 93
86-93, 339-353
provisional list of industries, 4, 6,
Second, 2, 10, 12, 14, 22-24, 26, 74,
7,9
118; 6, 74-85, 235-248, 253-255
to receive statistics, 228
Third, 2-3, 10, 29, 31, 37-38, 118;
represented on Committee of Statis7-8, 326-337, 421-438, 439-463
tical Experts of League of Nations,
Fourth, 3, 4, 16, 20, 24, 26-27, 29-30,
4, 10
41, 107, 110; 266-286
statistical publications, 6 and passim
Fifth, 3, 5, 16-17, 96; 219-228
unemployment statistics, methods
International Conference on Migration
and changes, forwarded to I.L.O.,
Statistics (1932), see also Appendix,
83
p. 151; 39-40; 464-508
wages, annual statement of defects,
International Conferences of Statis258; d a t a for calculating forwarded
ticians convoked by the Social Scito I.L.O., 191
ence Research Council: see Social
I.L.O. Year Book, 18, 23:
Science Research Council
predecessor to Year Book of Labour
International Convention relating to
Statistics, 41, 44
Economic Statistics: see Convention
relating to Economic Statistics
International Labour Organisation:
International Emigration Commission,
see also International Labour Office,
39
1, 5
Numerical references in italic type are to the numbers of the paragraphs of the resolutions as
given in Part I I ; other references are to page numbers.

INDEX
International Labour Review, 6-7:
family living studies, summary, 31-32
statistics, 41, 44-45:
cost of living, 22
employment, 14
food cost comparisons, 30
gainfully occupied, 11
hours of work, 21
industrial accidents, 33
industrial disputes, 38
migration, 39
rents, 30
unemployment, 13
wages, 17-18, 20, 26
International standardisation: passim
techniques of I.L.O., 1-8
International Statistical Institute, 6,
56; 25:
and base period for economic indices,
252
collaboration with, in classification
of industries, 8
migration statistics, 491
International trade statistics, minimum
list of commodities, 57
International Union of Towns, Conference on Housing Statistics: see also
Housing, Appendix, 1, 3
Committee of Experts in Housing
Statistics, 35-36
Juveniles: see Young persons
Kitchen: see also Rooms, 366; as room,
315, 358, 388, 389
Klezl, Felix, 22, 24
Labour Camps: and definition of gainfully occupied, 10, 22
Labour cost per unit of production:
index numbers, 97, 98, 138-146:
aggregate earnings as basf, 143
changes, interpretation, 141
industries, 144
localities, 144
period, 145
weights, 146
as percentages of wages to total cost
of production, 138-140
international comparisons, 140
price level data, necessary, 103
Labour legislation, and collective agreements, 37 ; 423, 428, 438
Labour organisations, membership, 77:
statistics of employment and unemployment, occupations, 82
unemployment data, 77
workers, as percentage of total in
industry or occupation, 77
Labour relations: see Collective agreements, Industrial disputes
Language, of migrants, 495

163

Laspeyres, formula for index, 29
Lavatory: see Toilet, Bathroom
League of Nations: see also Committee
of Statistical Experts of the League of
Nations, and names of organs and
departments, etc.
collaboration with, classification of
industries, 8
organs of, collaboration with I.L.O., 6
studies on housing, mentioned, 325
Liberal professions: see also Gainfully
occupied, Branches of economic activity, 25
Light: see also Retail prices, prices, 272
Lindberg, John, 13, 15
List of industries: see also Classification, Gainfully occupied, Industries,
9-10
detailed, Committee of Statistical
Experts of L. of N . (note to 73)
minimum, 49-57:
Committee of Statistical Experts
ofL. ofN., 73
principle, 52
provisional, I.L.O., 4, 6, 7, 9
Literacy, and migration, 506
Localised industries: see also Industries, and wages for specific occupation, 263
Lockouts: see also Industrial
disputes,
441
Man-days: see also Employment,
Exposure to risk, for accident rates, 353
Man-days loss: see also
Industrial
disputes, industrial disputes, 447, 458
Manufacturing industries: see also
Industries, and names of topics, e.g.
Earnings,
principles of grouping,
55-56
Marital status:
gainfully occupied, 68, 71
migrants, 502
Maritime Statistical Handbook, 19
Merchant marine, wages and wage
rates, 19
Method: see names of topics: Group
basket method, Representative selection, Sample, and passim
Migration: see also Migration statistics
Conference on Migration Statistics,
1, 3-4, 39; 464-508
definition, 465-467, 474
internal, 473, 507
International Labour Conference,
1922, 39
permanent, 466, 468
temporary, 467-470

Numerical references in italic type are to the numbers of the paragraphs of the resolutions as
given in Part I I ; other references are to page numbers.

164

INTERNATIONA!, STANDARDISATION OF LABOUR STATISTICS

Migration statistics: see also Migration, 39-40; 464-508
classifications, 492-507:
emigration and immigration, 471,
493
industrial status and industry or
occupation, 500, 501
nationals and aliens, 471, 494
permanent and temporary, 466470
coupons detachable from certain
documents, 486-488
declaration of residence, 482-483
expanding scope, 41-44
frequency of compilation, 508
frontier control, 484-485
methods, 476-491
port statistics, 479-481
recommendation of the International
Labour Conference, 1922, 39
uniform international method, 489491
Military service, gainfully occupied, 22
Minimum List of Commodities for
International Trade Statistics, mentioned, 57
Mining: see Coal Mining and names of
topics, e.g. Earnings
Nathan, Robert R., 13
National Bureau of Economic Research,
40
National income:
wages as part of, 20
index numbers, 102, 147-151
Netherlands, unemployment days in
relation to possible days of work, 76
Nichols, Charles K., 4
Nixon, J. W., 5, 11, 1 9 , 3 2 , 3 3
Nomenclature: see List of industries
Nominal wages: see Wages
Non-family household: see also Households
definitions, 362, 396-398
and occupation of dwellings, 364
Non-localised industries: see also Industries, wages for specific occupations, 263
Normal hours: see Hours, normal
Number of rooms: see Rooms, number
Nutritive value: see also Food
as basis for food price comparisons,
284
food, family living studies, 335
Occupancy of dwellings: see also Density of occupancy, Dwellings, Housing,
Rooms
definition, international comparisons, 408
more than one family household, 406

Occupants: see also Density of occupancy, Dwellings, Housing,
Rooms
classification :
character of habitable building,
369
density of occupancy, 370, 374,
408-414
rent paid, 379
size of dwelling, 369
definition, 360, 399
number, as basis of classification of
dwellings, 371
and rent restrictions on dwellings,
380
Occupational status: see
Industrial
status
Occupations: see also Gainfully occupied, Industries, List of industries
classification, 9-11; 1:
gainfully occupied, 24, 27, 28
housing censuses, 416
migrants, 497-499,501
wage rates and normal hours, same,
convention, 188-190
definitions, 5
earnings, 229
employment exchange statistics, 82
list, of I.L.O., 9
nomenclature, not attempted by
Committee of Stat. Experts, L. of
N., 67
remunerated, as definition of gainfully occupied, 10
similar, wages, 130
at time of census, gainfully occupied,
11
unemployment and employment, 82
wage data from payroll census, 91
wages :
additions to list of I.L.O., 260-265
international comparisons, 27 ; 130131, 134, 259-265
October, recommended as base month
for enquiry of wages, prices and
rents, 275
October enquiry: see also Cost of living,
International
comparisons,
Retail
prices, Rents, Wages
expanding scope, 42-43
hours of work, 21
occupations, additions, 259-265
rents, 30
retail prices, 27; 93, 272
wages, 18, 26; 93
wages, extension recommended, 267:
origin in British Ministry of
Labour enquiry, 24
Overcrowding: see also Density of
occupancy. Standards, 414
Overtime: see also Hours, Wage rates
amount permitted, frequency of
compilation, convention, 194

Numerical references in italic type are to the numbers of the paragraphs of the resolutions as
given in Part I I ; other references are to page numbers.

INDEX
Overtime {coni.):
average wages per hour, 128-129
and frequency of compilation of
earnings per hour, 133
rates, frequency of compilation, convention, 194
Paasche, formula for index, 29
Pan-American Sanitary Conference,
Rio de Janeiro, 1942, 32
Payrolls, used for wage censuses, 91
Permanent Committee on Road Traffic, of League of Nations, 6
Personal status: see Industrial status
Persons per room: see Density of occupancy
Physiological needs, climate and race
and international comparisons of
real wages, 254
food cost comparisons, 298
Piece rates: see also Earnings, Wage
rates
changes, 88
convention, 196
Piece work, earnings, 92
Postal Services, hours of work, 21
Pribram, Karl, 3
Production: see also Industries, List of
industries
primary, 2, 9
secondary, 2, 9
Public services: see also
Industries,
List of industries, Occupations; not
including establishments in other
branches of economic activity though
belonging to public authority, 48
Purchasing power: see also Cost of
living; of wages, 277
Quality: see also International comparisons, Retail prices
items in cost of living, principles,
244, 248
precautions, for price comparability,
297
Quantities: see also Food, Family
living studies; consumed, family
living studies, 336
Race, physiological needs, international comparisons of real wages, 254
Railways: see also Industries, List of
industries
industrial accidents, report on statistical methods, 33-34
wages, in general index, 156
Rates: see also Taxes, payable by
tenant, and rent, 376
Rates of wages: see Wage rales

165

Real wages: see also Cost of living,
International comparisons, Remuneration, Salaries, Wage rates, Wages
index numbers, 90, 92; original data
to be published, 255
international comparisons, 253-297:
agenda of Fourth Conference of
Labour Statisticians, 3
capital cities, 93
Committee of Stat. Experts,
I.L.O., 24
elemtnts other than wages, 279
International Conference of Labour Statisticians, Fourth, 266286
I.L.O., 27-28
methods, 293-297
race, climate, etc., 254
Regions: see also Food costs, Group
basket method, International comparisons, Rents
in international comparisons, 301309; food costs, 29
rents, 321
Remuneration: see also
Earnings,
Family allowances, Salaries, Wage
rales, Wages
allowances, earnings, convention,
172-173; and time rates, convention, 194, 195
elements other than wages, 26-27287-290:
Committee of Statistical Experts,
I.L.O., 26; 287-290
International
Conference
of
Labour Statisticians, Fourth,
26; 271, 279
in real wage comparisons, 279
per hour, problems, 101
special allowances, taken into account in earnings, 129
supplementary items, and frequency
of compilation of wage indices, 117
Rents: see also Cost of Living, Dwellings, Housing, Family living studies,
Rooms
per capita, 312
classification of dwellings, 378
Committee of Statistical Experts,
I.L.O., 417
definition, 376-377
enquiry in towns included in wages
and prices enquiry, 273; by I.L.O.,
322
and floor space, 317-319, 324
housing census, 417
international comparisons, 30; 310325:
Committee of Stat. Experts,
I.L.O., 30; 310-325
International Conference of Labour Statisticians, Fourth, 30;
273, 275-277

Numerical references in italic type are to the numbers of the paragraphs of the resolutions as
given in Part I I ; other references are to page numbers.

166

INTERNATIONAL STANDARDISATION OP LABOUR STATISTICS

Rents (cont.):
Rooms {cont.y.
regions, 321
number, in calculating densitv of
resolutions, Committee of Statistical
occupancy, 372-373, 410-414 *
Experts, I.L.O., 310-325
number, in classification of dwellrestrictions, and classifications of
ings, 365, 371, 403
dwellings and occupants, 380
number, in classification of occuper room, 311, 315, 317
pants, 370
statistics of vacant dwellings, 387
rooms per person, 414, see also Denand supplementary services, 323
sity of occupancy
Representative selection: see also
subsidiary, definition, 359
Sample
Running-water: see also
Housing,
character of data on total wages, 225
Standard dwelling unit, 30
districts, wage index data, 113
classification of dwellings, 404
earnings and hours, 169
Safety:
see also Industrial accidents
establishments:
coal mines, Report of Committee of
actual earnings, 89
Experts in Safety, 34
actual hours, 89
employment, 80
Salaried employees: see also Gainfully
family living studies, 327
occupied, Industrial status
industries, time rates of wages and
gainfully occupied,
classification,
normal hours (convention), 182
62-65
industries and categories, in wage
salaries and hours, statistics recomindexes, 111
mended, 221
labour cost per unit of production,
Salaries: see also Total wages and salalocalities and industries, 144
ries, Wages
localities, earnings per hour indices,
annual, by industries, 226
132
statistics recommended, 221-222
number of employers, employment, 80
total, industries, 225
total wages, employed persons, 232
Sample: see also Representative selection
towns, for comparisons of consumpperiods, earnings and hours, 92
tion, 304
representative, earnings and hours,
towns, industries and occupations
convention, 169
wage data, 267
statistics, unemployment, 81
typical occupations, wages and hours,
Sanitary accommodation: see Toilet,
92-93
unemployment statistics, 83
Bathroom
wage earners, for wage statistics,
Seasonal changes:
convention, 167
food prices, and October as enquiry
Retail prices: see also Cost of living,
month, 275
International
comparisons, October
wages, 160
enquiry, Quality
Seasonal migration: see also Migration,
capital cities, 93
470
defects, an obstacle to international
Services: see also Gainfully occupied,
comparison, 299
Industries, List of industries, in clasfood, fuel and light (working class
sification of industries, 2, 9
families), 272
Severity rate, accidents, 350, 352-353
food, seasonal changes, 275
methods of collection and calculaindustrial disputes, 462
tion to be unchanged, 244
Sex: see also Women
October enquiry, 27
classification of gainfully occupied,
quality, 244, 248; precautions, 297
68-72
statistics, expanding scope, 41-45
earnings, and actual hours, convention, 177, 224
Ricci, Umberto, 19
occupations in each industry, 224,
Riches, E.J., 38
230
Rooms: see also Density of occupancy,
unemployment, 82
Dwellings, Housing, Rents
wage censuses, 91
definitions:
wage rates and normal hours, conCommittee of Stat. Experts, L. of
vention, 192
N., 388-390
wages, index numbers, 155
International Union of Towns,
Size of dwelling: see also Density of
357-359
occupancy,
Dwellings,
Housing,
habitable, 315; definition, 358
Rooms
Numerical references in italic type are to the numbers of the paragraphs of the resolutions as
given in Part I I : other references are to page numbers.

INDEX
Size of dwelling (cont.):
classification of dwelling, 365, 371,
403
in classification of occupants, 369-370
density of occupancy, 372-373, 410414
in housing censuses, 401
and rent paid, in classification of
occupants, 379
vacant dwellings, 387
Size of rooms: see Floor space
Skill: see also Wages
wage rates and normal hours, convention, 189
wages, index numbers, 100, 155
Snow, E. C , 22
Social insurance: see also Unemployment insurance
contributions of employers and
employees, 25-26:
and relative real wages, 279
and wage statistics, 287-290
Social Science Research Council: see
also Appendix, 1, 3
Conferences, 16, 19, 20:
resolutions, 19:
cost of living, 291-297
real wages, 256-265
wages and hours, 94-151, 152162
First Conference, 26:
base for wage statistics, 23
wages as part of national income,
20
international
comparisons,
real
wages, 24
Second Conference, cost of living,
international comparisons, 27
Social status: see Industrial status
Staehle, H., 28-29
Standard dwelling units: see also
Dwellings, Housing, 30; 273, 314-315,
322
Standard of living: see also Cost of
living, Family living studies, Housing,
Salaries, Wages
changes, different industries, 121
descriptive details, typical groups,
292
fluctuations, and index numbers of
wages, 97, 98
hours and conditions of work, 110
and wages, 103, 104; index numbers
to measure, 100, 108-120
Statistical Year Book of the League of
Nations:
base period, 23
geographical list of countries, for
classifying migrants, 493
Strikes: see Industrial
disputes

167
Taxes: see also Rates
and calculation of rents, 376
direct, effect of inclusion or exclusion
from index numbers of cost of
living, 237
separate item, family living studies,
335
Tenure: see also Dwellings, Housing
dwellings, housing censuses, 418
Textile industry: see also Industries,
List of industries
hours of work, 18
Tripartite Technical
Conference
1937, 5
wages, 18, 26
Toilets: see also Dwellings,
Housing,
Standard dwelling units, 30
classification of dwellings, 322, 366,
404
subsidiary room, 359, 389
Total wages and salaries: see also
National income. Salaries, Wages
frequency of compilation, 149, 225,
231
industries, 225
by industries, mining and manufacture, 231-232
International Labour Review, article,
20
part of national income, 20; 102,147151
related to value of production, 139;
to net value, 147
Social Science Research Council,
Conference, 20; 147-151
statistics, expanding scope, 42-43
Trade unionism: see also Labour organisations, and industrial disputes, 449
Trade unions: see Labour organisations
Transit migration: see also Migration,
472
Unemployed : see also Employed, Gainfully occupied, Unemployment
family living studies, 32
and gainfully occupied, 11-12:
international comparisons, 21
national censuses, 21
in gainfully occupied, 20:
according to last occupation, 21
Unemployment: see also Employment,
Unemployed
agenda, 2nd Conference, Labour
Statisticians, 2
census, special needed, for all aspects, 23
censuses, results reviewed, 13
days, in relation to total possible
days of work, 76
definition, for statistical purposes, 83
in industrial or occupational census,
81:
by sex, 81

Numerical references in italic type are to the numbers of the paragraphs of the resolutions as
given in Part I I ; other references are to page numbers.

1 6 8 ' INTERNATIONAL STANDARDISATION OF LABOUR STATISTICS
Unemployment (cont.):
International Conferences of Labour
Statisticians, 12; 74-85
international index, 13
partial, 85
in population census, 81
in receipt of benefit, 75
short-term employment, 85
special enquiries, 81:
by sex, 81
statistics, 12-13:
comparisons of different sources,
84
expanding scope, 41-45
ideal field, 83
methods, 83:
and changes in, 83
forwarded to I.L.O., 83
I.L.O. report, 12
trade unions, 77
unemployment insurance, 74-75
young persons: recommendation of
International Labour Conference,
1935, 12; seeking work, in census
of gainfully occupied, 12
Unemployment Convention, 1919
recommendation, to send statistics
t o I.L.O., 12
Unemployment insurance: see also
Social insurance; as basis for unemployment statistics, 74-75
Unemployment relief works:
in definition of gainfully occupied, 10
gainfully occupied, 22
Unpaid family workers: see also Gainfully occupied
classification, gainfully occupied,
10,15, 60-61
Unskilled workers: see also Gainfully
occupied, employment exchange statistics, distinguished separately, 79
Vacant dwellings: see also Buildings,
Dwellings, Housing census
annual returns, 386
classification, 387
classification and definition, 407
Vacations with pay: see Holidays with
pay
Variez, L., 40
Wage-earners: see also Gainfully occupied; classification, gainfully occupied, 62-65
Wage payment system, and frequency
of compilation of wage data, 116
Wage rates: see also Earnings, Piece
rates, Remuneration, Salaries, Wages,
86
changes, representative of actual
earnings, 109

Wage rates {cont.):
classification :
occupation, convention, 188-190
skill, convention, 189
defects in information, annual statement of I.L.O., 258
definitions, 185: uniformity needed,
257
and earnings, 234:
both needed, 256
both published, 159
differences, special enquiry, 257
divergencies, 159
relation, 227, 234, 270
in economic forecasting, 122
frequency of compilation, 187
hourly, 161
index numbers, 19:
convention, 196-199
industries, mining and manufacturing, convention, 182-199
industries, representative selection,
convention, 182
international comparisons, 86-91
nominal, index numbers, 90
notes, accompanying, 234
occupat-'ons in each industry, convention, 187
overtime, frequency of compilation,
194
piece work:
changes, 88
convention, 196
time, and piece work basis time
rates, for international comparisons, 269
time period, same as for normal
hours, convention, 191
time, types, convention, 183-185
for transport, commerce, administration, recommendation, 219-220
types, 87
typical categories of time workers, 92
typical occupations, capital cities,
wage comparisons, 93
weekly, 161
young persons:
in census, 91
convention, 192
Wage statistics, and Committee of
Statistical Experts of I.L.O., 286
Wages: see also Earnings, Index numbers, October enquiry,
Remuneration,
Salaries, Total wages and salaries,
Wage rates
agenda, First Conference, Labour
Statisticians, 2
agriculture, 18; convention, 200-202
American report t o Social Science
Research Council, 156
censuses, 19; 91

Numerical references in italic type are to the numbers of the paragraphs of the resolutions as
given in Part I I ; other references are to page numbers.

INDEX
Wages (coni.):
changes, from changes in rates, 88
classification, by industries, districts,
occupations, sex and age division,
91
coal mining, 18, 26
Committee of Stat. Experts, I.L.O.,
16-17
defects in information, and statement by I.L.O., 258
details for industries, occupations
and localities, 127
and economic forecasting: see Economic forecasting
frequency of compilation, 160, 201
per hour of work:
international comparisons, 99
problems, 101
as purpose of index numbers, 97-98
index numbers: see Index numbers
international comparisons: see also
International comparisons, 24-27;
on agenda of International Labour
Conference, 285
Committee of Stat. Experts,
I.L.O., 24
between countries or districts, 107
per hour of work, 99
International
Conferences
of
Labour Statisticians, 24, 26;
253-255, 266-286
occupations, similar, 130-131, 134,
259-265
Social Science Research Council,
Conferences, 24, 26; 256-265
International Conferences of Labour
Statisticians, 16-17; 86-93, 219228, 266-286
International Labour Conference,
1938:
Convention, 17; 163-218
Recommendations
for further
action, 229-234
limitations of quarterly statistics of
I.L.O., 259
localities, additions to I.L.O. list,
260-264
national income, share in, 20; index
numbers, 102, 147-151
nominal, 16-19; given in compiling
index numbers, 90
occupations:
additions t o I.L.O. list, 260-265
localised and non-localised, 263
overtime, and average wages, 128129, 133
seamen, 19
seasonal fluctuations, 160
Social Science Research Council,
Conferences, 3, 16; 94-162

169

Wages (cont.):
statistics, 16-20:
expanding scope, 41-45
methods, report of I.L.O., 16
reports by I.L.O., 17-18
supplementary allowances, 287-290
textile industry, 18, 26
Water closet: see Toilet
W. C. : see Toilet
Women: see also Gainfully occupied,
Sex, Unpaid family workers; wages,
separate index, 162
Woodbury, Robert Morse, 28-29, 32,
116
Workers' organisations: see Labour
organisations
Working on own account: see Gainfully
occupied,
Industrial
status
gainfully occupied, 60, 63
World indices: see also International
indices
employment, 15
unemployment, 13
Woytinsky, W., 37
Year Book of Labour Statistics, 2, 6-7,
11,41-45:
annual volume recommended, 4th
Conference, 276
cost of living, 23
employment statistics, 14
family living data, 32
gainfully occupied, 11
hours of work, 21
index numbers, nominal and real
wages, 20
industrial accidents, 34
industrial disputes, 38
international indices, food costs, 30
I.L.O. Year Book as predecessor,
41,44
migration, 39-40
rents, 30
scope, 41-44
unemployment, 13
wages, 18, 26
Young persons: see also Age classifications, Gainfully occupied, and names
of topics
earnings and hours, convention, 177
seeking work, and definitions of
gainfully occupied, 12
as unemployed, 83
unemployment:
recommendations of International
Labour Conference, 1935, 12
wage data, in census, 91
wage rates and normal hours, convention, 192

• Numerical references in italic typeiLne to tha.numbers of the paragraphs.of the resolutions as
given in Part II; other references are to page numbers.

Publications of the International Labour Office
Year Book of Labour Statistics 1942
(Seventh Year)
This annual publication, now issued in a trilingual edition
(English, French and Spanish), presents in tabular form the most
complete data obtainable in the principal countries of the world
relating to:
Gainfully Occupied Population
Employment and Unemployment
Hours of Work
Wages
Cost of Living and Retail Prices

Family Budgets
Migration
Industrial. Accidents
Production and wholesale price
indices, exchange rates

The tables cover the last thirteen years. They are compiled
from the official statistics of more than 50 countries in all parts of
the world, supplemented in some cases by data obtained from
private sources. Special care has been taken to include the relevant
series, so far as available, for Latin Amercian countries. Recent
enquiries of the I.L.O. on wages, hours of work and cost of living
in various countries are reproduced in full.
An introductory note to each chapter calls.attention to the
principal questions of method to be borne in mind in interpreting
the data, especially when utilising them for international comparisons. The statistics are presented in a systematic way with
this object in view.
"Indispensable to statisticians of all countries . . . Illustrates the wealth
and complexity of the knowledge collected for experts to interpret . . . Through
the co-ordination, and so far as possible, comparison of the official figurespublisned by different countries it is possible to gauge world economic trends in a way
which was out of the question twenty years ago."
—The Times, London.
" . . . a valuable work for everybody concerned with social policy . . . The
statistics provided tell not only the story of the great world depression, 1929. to
1935, but cover also the year 1941, a great performance, because wartime difficulties most likely hamper every statistical service."
—The Standard, Wellington, N.Z. (referring to the 1941.edition).
". . . constituye una obra de consulta y de referencia en materia del trabajo
en sus más variados aspectos."
—Cuba Económica y Financiera.

xii + 222 pages.

Price: paper bound
cloth bound

$2 or 8s.
$3 or 10s. 6d.