Preview

INTERNATIONAL LABOUR OFFICE

PRICES, WAGES,
AND INCOMES POLICIES
IN INDUSTRIALISED
MARKET ECONOMIES

by
H. A. TURNER and H. ZOETEWEIJ

GENEVA
1966

STUDIES AND REPORTS
New Series, No. 70

PRINTED BY "IMPRIMERIES POPULAIRES ", GENEVA, SWITZERLAND

CONTENTS
Page
INTRODUCTION

CHAPTER I: Wages and Economic Stability
Scope and Definitions
The General Economic Context
Inflation : Brief History .
Post-War Recessions
Wages
Prospects for Economic Stability

5
5
8
10
13
19
22

CHAPTER II: Effects of Inflation
The Meaning of Inflation
Prices and the Cost of Living
.Price Increases and the Distribution of Income and Wealth
Price Increases, Saving and the Risk of Galloping Inflation
Price Increases as a Social Lubricant
Inflation and Economic Growth
Rising Prices and the Balance of Payments
Conclusions

28
28
29
31
40
49
50
52
54

CHAPTER III: Causes of Inflation and the Fixing of Prices and Wages . . . .
Demand Pull and Cost Push
Features of Modern Price and Wage Fixing
Government Price Fixing
Pricing in the Private Sector
Raw Materials
Manufactured Goods
Some Qualifications to Cost-Pius Pricing
Summary
Retailing
Wage-Fixing Processes
Collective Bargaining
Trade Union Criteria
Management Attitudes
Wages and the Demand for Labour

56
56
58
59
60
60
62
64
67
68
69
70
71
72
75

CHAPTER IV: Some Mechanics of Inflation
Introduction
Indicators of Demand Inflation and Cost Inflation
The Demand for Labour
The Relation between Profits and Wages
Factors in Inflation

77
77
78
79
85
86

CHAPTER V: Wage Policy—Methods and Institutions
Analysis and Information
Government Exhortation
Central Consultation
Central Wage Negotiation

92
92
94
95
99

PRICES, WAGES, AND INCOMES POLICIES
Page

Central Government Control
Central Arbitration
Differences of Form and Substance

103
107
109

CHAPTER VI : Criteria of Wage Adjustment in National Wage Policy
Adjustment of the Over-all Wage Level
The Productivity Criterion
Some Problems of the Productivity Criterion
The Scope for Wage Increases
Maintenance of Real Wages
The Distribution of Wage Increases
The Comparative Criterion or the Nature of the Work
Demand for and Supply of Different Types of Labour
Productivity Growth in Particular Industries
Repercussions Elsewhere in the Economy

112
112
113
115
117
119
120
122
124
125
127

CHAPTER VII: Some Problems of National Wage Policy
Equity in the Distribution of Income
Sectional Wage Increases
The Need for Productivity Growth
Wage Drift
Institutional Problems
The Need for a Consensus

129
129
132
134
136
142
146

CHAPTER VIII: Policy for Non-Wage Incomes

149

CHAPTER IX: Steps towards Stability—Improving Wage Systems
Greater Security of Wages and Wage Adjustments
Workers'Participation in Profits and Investments
Combined Wage and Price Negotiations
Increasing Control over Plant Bargaining
Public Influence over Wage Fixing

157
157
159
161
162
166

INDEX

171

LIST OF TABLES
I Indications of Post-War Growth and Stability, 1950-62
9
II Industrial Production and Consumer Prices during Periods of Rising
Output
10
III (a) Annual Changes in Consumer Prices, and (b) Distribution of
Annual Percentage Changes in Consumer Prices in 13 European
Countries and the United States 1950-63
12
IV Post-War Trends in Manufacturing Wages and Consumer Prices,
1950-62
16-17
V The Wage Drift in Selected European Countries, 1950-62
20
VI Average Annual Percentage Increases in Wage and Salary Incomes,
and in Other Incomes, in Selected Countries
21
VII Consumer Price Indices
29
VIII Distribution of the National Income at Current Cost, 1955-62 . . . . 33-35
IX Personal Savings as Percentage of Disposable Income 1954-62 . . . .
41
X Total Savings of Selected Countries, 1954-62
42-45
XI Percentage Growth of Output and Price Changes in Selected Periods,
Compared with the Preceding Decade
50

INTRODUCTION

This study concerns certain problems of wage policy in industrialised
countries aiming at high employment and steady growth of production
and incomes, largely in the framework of a market economy. Interest in this
subject has been aroused mainly by the fact that during the past generation several of these countries have experienced almost uninterrupted
price increases. More recently, questions of " the planned growth of
incomes " have also been examined, in one or two cases, in the context
of long-term over-all economic and social policy-making. But the most
common concern has so far been with wage policy as a tool of antiinflationary policy. Continuing price increases have been held responsible
for distortions in international payments and in other respects, and
attempts to tackle these problems by measures curbing demand and
production have hampered growth and threatened full employment,
thus endangering economic progress as a means to the ultimate goal of
social advance.
In the search for an alternative, less wasteful approach to the problem
of inflation, attention has come to be focused on the processes of cost
and price determination—on the procedures and substantive principles
of fixing wages, profits, rent and interest. Since all these cost elements
are also incomes, attempts at dealing with them in some systematic way
have in recent years become known as " incomes policies ". As an antiinflationary device such policies have as their objective the avoidance of
excessive growth in money incomes as compared with increases in output.
In concentrating on possible inflationary aspects, the present study
covers of course but a small—and by no means the most important—
area of income formation and distribution. Questions of equity, of
improving economic opportunity, and of tackling extreme poverty are
more basic and important. Indeed, at the time of writing, the richest
country in the world has launched a " war on poverty ". These questions,
however, are not the subject of this study except in one important respect.
When action is contemplated to guide or regulate income formation
with a view to avoiding inflation, inevitably the question arises : according to what rules are the incomes of various groups—wage earners,
farmers, professional workers—to be adjusted? And, as noted in Chapter
VI, at that stage, issues of equity cannot be avoided. Thus, in practice

PRICES, WAGES, AND INCOMES POLICIES

incomes policies intended as anti-inflationary devices have usually provided for priority increases in the wages of the lowest-paid groups of
workers.
The present study deals very largely with problems of anti-inflationary
incomes policies that arise in connection with wage determination. Nonwage incomes are discussed, especially in Chapter VIII, but the main
subject of this volume is wage policy. There are various reasons for this
limitation. First, questions of wage determination, and the specific
technical problems that it involves, are of more central interest to the
I.L.O. and can be discussed in its publications more fruitfully than other
aspects of incomes policy. Second, wages (including salaries) are by far
the largest category of incomes in any industrialised country and they
are inevitably at the centre of any policy for influencing over-all price and
income levels. Furthermore, since there are far fewer wage schedules than
there are price schedules, it may be less difficult to develop practical rules
for adjusting the former than for fixing the latter. At the same time it
seems clear that no really effective wage policy could be applied or would
be equitable, unless it were accompanied by effective policies relating
to other incomes. And in a broader perspective, incomes policies may
come to be regarded primarily as means of implementing social goals in
respect of income distribution, rather than as mere anti-inflationary
devices. Nevertheless, it seems useful to consider questions of wage
policy—its possible scope and functions, problems and probable effectiveness—while recognising that, to be effective and equitable, wage policies
need to be accompanied by appropriate measures relating to other incomes.
The first two chapters of the present study provide a general background. First, the basic obj ectives of economic progress—growth and high
employment—are discussed and attention is drawn to the post-war
record, in most countries, of growth that has been vigorous but not
uninterrupted, and accompanied by continuing price increases. In the
second half of the first chapter some factors are considered that are
likely to have a bearing on future growth, and it is suggested that, on
balance, the prospects for growth seem favourable, though they will
probably be somewhat marred by continued inflationary pressures.
Chapter II examines the question of whether, why and to what extent
continuing price increases are undesirable from the point of view of
economic growth, welfare, or both. The conclusion is drawn that some
of the drawbacks of inflation seem less serious than they have sometimes
been thought to be. But continuing price increases can simply not be
sustained when they cause a country to experience serious difficulties in
its balance of international payments. Present arrangements of international trade and payments being what they are, balance-of-payments

INTRODUCTION

problems have called for measures for curbing demand which have
tended to lead to recessions or interruptions of growth.
This leads to the question whether part of the solution to the modern
problem of inflation does not he in altering the existing arrangements
for international trade and payments. But for purposes of the present
study the more important question is whether or to what extent the new
inflation may have been due to existing methods of wage and price fixing
(and therefore, in principle, may be capable of correction by means of
incomes policy) or to excessive over-all spending (and therefore may
call for measures to restrict demand). This question is examined in two
parts. First, in Chapter III, after some comments on the " demandpull " and " cost-push " controversy, prevailing methods of price and
wage fixing are described in general terms. This leads to the tentative
conclusion that these methods may very well by themselves (i.e. in the
absence of excess demand) lead to continuing price increases. Then, in
Chapter IV, the difficult problem is considered of how, in any concrete
situation, one may determine whether inflation has indeed been caused
by wage and other cost increases, or by other factors.
Such information as is available does not appear to establish, with
the precision required in statistical analysis of scientific problems, the
parts played respectively by increases in demand and increases in costs
in causing inflation. But policy decisions can rarely be deferred until the
exacting standards of scientific inquiry have been fully met, and in
several countries approaches to wage policy have been made as part of
wider policies for dealing with specific problems arising from steady
inflation. Chapters V and VI give an illustrative analysis of these experiments, the former dealing with procedural questions, the latter with the
substantive rules that have been used as criteria or guidelines for wage
adjustments.
These experiences reveal a number of considerable difficulties regarding both the formal principles and the practical enforcement of a wage
policy. Many of the established practices of wage fixing that have been
held partly responsible for the post-war inflation in industrialised
countries also constitute obstacles to the introduction of any high degree
of central guidance in wage determination, while other features of existing systems escape any kind of central control. These problems are
discussed in Chapter VII.
In Chapter VTII the general nature and problems of policies regarding non-wage incomes are discussed briefly.
On the basis of the analysis in Chapters VII and VIII the conclusion
is reached that many considerable difficulties are bound to arise in formulating and implementing a comprehensive and effective incomes policy.

PRICES, WAGES, AND INCOMES POLICIES

However, a variety of possible causes of wage inflation can be identified
and it may be feasible to make considerable progress towards economic
stability by attacking some of these causes even before a " complete "
incomes policy (assuming it to be desirable) can be adopted. A number
of possibilities of making such partial progress are discussed briefly in
Chapter IX.
The main purpose of this study is to stimulate thought and discussion
on a number of important and controversial questions. The views
expressed are those of the authors and not necessarily those of the I.L.O.

CHAPTER I
WAGES AND ECONOMIC STABILITY
SCOPE AND DEFINITIONS

The bearing of wage trends and wage policies on economic stability
differs radically according to the type of economy considered. Broadly,
the problems of underdeveloped economies in thisfieldare very different
from those of industrial countries. And among the latter the problems
differ, largely according to whether or not the economy is centrally
planned. This study will be concerned with wage policy only in countries
which are relatively highly industrialised, and which also have a large
sector of private enterprise1, so that wages are mostly paid by private
employers. Within such countries varying distinctions are usually drawn
between salaries and wages ; here, salaries will generally be regarded as a
form of wage payment. At some points, however, it will be necessary to
examine these two forms of remuneration separately—some important
practices and factors are common to wage and salary determination,
others are not.
The centre of primary interest in this study is the question how movements in the level and structure of wages affect economic stability, and
how any negative effects they may have could be corrected through
appropriate policy measures. Action to that end will have to extend to
incomes other than wages, and Chapter VIII deals with this question;
but for reasons indicated in the Introduction the analysis in this report
primarily concerns wage problems.
Economic stability may be defined generally as including—
(i) the maintenance of a high level of employment, changing with
long-term variations in the size of the population and in labour-force
participation rates, but free from violent short-termfluctuations;
(ii) the avoidance of substantial fluctuations in production ;
1
A volume on problems of wage policy in developing countries is in preparation
in the I.L.O. and an I.L.O. contribution to the United Nations Report on the World
Social Situation, 1965, contains a substantial analysis of certain questions of wage
policy in centrally planned economies.

PRICES, WAGES, AND INCOMES POLICIES

(iii) the preservation of a reasonably stable average level of retail
prices ;
(iv) the achievement of a satisfactory balance of external payments.
The concept of economic stability used in this study also embraces—
(v) the notion of a relatively steady rate of economic growth.
There is a basic tendency for productivity to rise, enabling real income
per head also to increase. Sustaining a high rate of productivity increase
as well as maintaining high levels of employment—in other words,
maintaining a high rate of economic growth—is now a major policy
objective.1
Of course, each item of this definition raises problems of practical
economic policy. " A high level of employment " requires closer definition in relation to the labour force in particular countries and the structure of employment opportunities presented by these economies. This
raises in an acute form the general question of how far definitions are
translatable into measurable terms, as statistical indices which can reliably
serve as guides to policy. The aim of avoiding fluctuations in production
involves decisions as to how far governments should attempt to control
or offset fluctuations arising from a limited section of the economy,
which may or may not represent adjustments to changing patterns of
demand and supply, as opposed to more general swings in economic
activity. The preservation of stable prices opens the question of how
strictly this objective is to be regarded. Would a slight tendency for the
general price-level to rise or fall be acceptable ? Similarly, it will in practice
be necessary for a government to decide what level of exchange reserves
it wishes to maintain, and the extent to which it is prepared to allow occasional reductions over short periods.
Decisions on matters of this kind will usually reflect national economic
circumstances on the one hand, and prevailing social ideas and preferences on the other. Employment objectives, for example, will depend upon
prevailing ideas about how important it is to provide work for different
classes of workers, including juveniles and married women. Similarly, the
increases achieved in real income per head will depend partly on social
choices between longer working hours and leisure, and the over-all propensity to save out of current income, the latter in turn being determined
in some measure by social preferences for particular kinds of income
distribution. The changing distribution of income, particularly, is at the
1
Thus, in November 1961 the 20 member nations of the Organisation for Economic Co-operation and Development set themselves the target of attaining a 50 per
cent, increase in their combined real gross national products during the decade from
1960 to 1970, i.e. an annual growth rate of 4.1 per cent.

WAGES AND ECONOMIC STABILITY

same time a focus of social controversy and an important element in
economic stability both as cause and effect. Social preferences and pressures in this respect may therefore be expected to exert an especially
significant influence on the direction of economic development. Economic
objectives will likewise be influenced by more narrowly political events,
such as decisions to allocate capital resources to military rather than productive uses, or to foreign investment where the returns are expected to
be in some degree political rather than purely economic.
Aside from these general influences, however, there is the question
of the degree to which the stability objectives suggested above, even thus
broadly, are compatible with each other. Is high and rising employment
compatible with general stability of retail prices ? And how far can stable
prices be reconciled with a steady increase in output and productivity?
Are there reasons for believing that, in the short run at least, not all of
the objectives can be achieved simultaneously, so that some may have
to be relaxed temporarily, in order that the others can be fulfilled ?
Finally, the question arises of how far pursuit of the various objectives
included in our definition of economic stability is compatible with certain other objectives, such as a reduction in the inequalities in the distribution of income and wealth. An important question that will be considered in later chapters of this study is that of the extent to which, or
conditions in which, present methods of wage negotiation can be combined with economic stability.
These questions are central to contemporary wage policy problems.
A more detailed examination of the background and issues of recent
wage controversy will help to indicate directions in which solutions may
be sought. But while these questions may be stated in more precise terms,
it remains in some respects impossible to give a final answer to them, if
only because techniques of economic measurement are not yet sufficiently
precise and subtle to provide incontrovertible evidence of significant
relationships.
The above aspects of economic stability have posed major issues
throughout the history of private enterprise industrialism. The nature
and causes of business cycles and other economic fluctuations, inflation,
and unemployment have been important problems for at least 100 years.
But the setting of questions of economic stability during the past 15 years
or so has changed in several respects. Three of these are of special relevance to the present study. First, this period has been one of considerable
growth, with no more than relatively minor interruptions, and such
growth has become an explicit object of policy in all the countries considered here. Second, prices have been rising, more or less gradually but
with hardly any interruptions, and this " creeping inflation " has become

PRICES, WAGES, AND INCOMES POLICIES

an important concern of policy largely because it is regarded as a threat
or obstacle to steady growth. Third, present methods of wage and price
fixing have come into focus of interest as possible contributors to this
inflation, and attempts have been made, or proposed, to gain a measure
of control over the movements, especially of wages. While these factors
seem common to most countries, additional developments have occurred
of more specific interest to individual countries—for example, attempts to
combine some variety of " planning " with the processes and institutions of a market economy, and attempts to alter the distribution of
incomes. While the present study is concerned primarily with questions
arising from the more general features, some consideration will be given
also to such more specific issues.
THE GENERAL ECONOMIC CONTEXT

In most of the countries with which we are concerned, the past 15
years have been a period of remarkable economic expansion. In the
United States, and one or two other countries which had not suffered
great physical destruction or loss of economic assets during the Second
World War, this expansion continued a phase of growth initiated by the
war itself. Though there were some differences between European countries in their rates of recovery from wartime damage and disorganisation,
western Europe as a whole had about recovered its pre-war level of
industrial output by 1948-49 and has more than doubled it since (table I).
Over the same period there has been, in nearly all these countries, a significant, and in some cases very considerable, rise in employment. It is also
clear from a comparison of the output and employment indices in table I
that substantial increases in industrial productivity have been achieved.
In other words, the general tendency has been strongly upwards and
this upward trend extends back, in cases where it has not been overlaid
by adverse wartime effects, over a generation. But it has been accompanied by instabilities of two kinds, as table I suggests. First, there has
been a persistent or recurrent inflation in the sense of rising retail prices.
Second, recessions in economic activity have occurred. These usually
produced no more than a short and by no means universal interruption
in economic growth. Thus, in the United States four periods of contraction in activity occurred after the Second World War. But compared
with earlier periods the amplitude of fluctuations has been remarkably
small. The last of the four recessions (from May 1960 till February 1961)
was also the mildest and at the time of writing the period of growth
following this recession had become the longest peacetime period of
expansion of this century. In a number of western European countries

TABLE I.

Country

INDICATIONS O F POST-WAR G R O W T H A N D STABILITY, 1950-62
(1958=100)
1950

1951

1952

1953

1954

1955

France :
Industrial production
Non-agricultural employment .
Consumer prices

58
92
64

65
94
74

66
94
83

67
93
82

73
94
82

80
95
83

Germany (Fed. Rep.) :
Industrial production 2 . . . .
Non-agricultural employment 3
Consumer prices 4

48
70
84

56
74
91

60
78
92

66
81
92

74
85
92

Italy:
Industrial production
Manufacturing employment . .
Consumer prices

55
95
76

63
96
83

64
95
87

70
96
88

Sweden :
Industrial production 5 . . . .
Manufacturing employment . .
Consumer prices

80
100
67

84
103
77

82
100
83

United Kingdom :
Industrial production 8 . . . .
Non-agricultural employment 9 .
Consumer prices

82
94
68

85
95
74

United States :
Industrial production
Non-agricultural employment 1 0
Consumer prices

84
88
83

87
94
90

1956

1957

1958

1959

1960

1961

1962

97
85f

96
99
87

100
100
100

101
99
106

110
99
110

116
100
114

123
101
119

85
90
94

92
95
96

97
99
98

100
lOOf
100

107
104
101 t

119
106
102

126
109
105

132
112
109

77
97
91

84
99
93

90
101
96

96
102
97

100
100
100

111
100
100

128
107
102

142
114
104

156
120
109

84
98
84 f

87
99
85'

92
102
87

96
102
91

98
102
95

100
100
100

106
101
101

117
106
105

120
108
107

122
107
112

82
95
81

88
96
84

94
97
85

99
100
94

101
101
97

100
100
100

105
101
101

112
103
102

114
104
105f

115
105
110

90
96
92

98
98
93

92
95
93

107
102
94

107
103
97

100
100
100

113
104
101

116
106
102

117
105
103

126
109
105

i
g
C/3

>
ö
M

o
o

!
98
99
89 f
103
99
93

Source: IX.O. : Year Book of Labour Statistics (Geneva).
t=Restricted comparability of preceding with following figures.
"Revised series; former series linked to revised series. ! Excl. Saar. a Up to 1958 excl. Saar. ' Up to 1959 excl. Saar. • Excl. electricity and manufactured gas. ' New series
linked to former series. ' July-Dec. 8 Incl. construction. 9 Excl. Northern Ireland. I0 Prior to 1959 excl. Alaska and Hawaii.

v©

PRICES, WAGES, AND INCOMES POLICIES

there were mild recessions between 1951 and 1953, and between 1956
and 1958. All these recessions have been very limited in depth as well as
in duration.
An especially significant feature of the whole period has been that
instabilities of both kinds—price inflation and recession—have sometimes
appeared together.
INFLATION: BRIEF HISTORY

Periods of rising prices are nothing new, even in the present century.
There was of course a very sharp general inflation during the First World
War. And in phases of economic expansion before 1939 prices also rose,
at some times and in some countries as fast as or faster than in more
recent experience (table II).
Recent inflations differ from these cases in three main respects.
Firstly, the inflationary period has been very long, beginning generally
with the outbreak of the Second World War. Secondly, the incidence of
inflation has been general in the countries in question. No country in the
group here discussed has experienced any but the briefest phases of
TABLE II. INDUSTRIAL PRODUCTION AND CONSUMER PRICES
DURING PERIODS OF RISING OUTPUT
(Average annual percentage increase¡decrease)
1908-13

Country

France :
Industrial production

. . . .

6.2
3.0

Germany (Fed. Rep.) :
Industrial production
Consumer prices

. . . .

6.4
2.0

United Kingdom :
Industrial production
Consumer prices

. . . .

6.2
1.9

. . . .

9.8
2.4

United States :
Industrial production

1923-29

8.5
13.4
9.8 *
4.11

2.7
--0.8
7.0 1
i

1932-37

1952-57

1957-62

4.0
4.4

8.5
0.8

5.5
7.3

20.4 2
0.8 2

11.6
0.9

7.7
2-1

8.5
1.4

4.3
3.7

2.4
2.5

20.0
1.0

5.0 3
1.58

3.4
1.5

Sources: United Nations : World Economic Survey 1957 (New York), pp. 19-20, and Monthly
Bulletin of Statistics. I.L.O. : Year Book of Labour Statistics, op. cit.
— = negligible.
1
1924-29. • 1932-38. ' 1954-57.

WAGES AND ECONOMIC STABILITY

reduction in the general price level. Several countries have had rather
longer intervals (of up to about three years) of price stability.1 But these
intervals have sometimes been due to the offsetting of a general tendency
for prices to rise by a price fall in one section of output affected by special
circumstances as, for some years after 1951, for agricultural products.
In no case has such price stability proved permanent. Before the war,
by contrast, periods of rising prices were relatively short, and were subsequently offset by general reductions. This cyclical behaviour of prices
reflected relatively large cyclical variations in output and employment.
These variations have been much smaller in recent years but price levels
appear to have acquired a persistent upward bias even when minor
fluctuations in activity do occur. Thirdly, though persistent, inflation has
usually been relatively moderate.
The period since 1939 can be roughly broken into the following
phases. In the first year or two of the Second World War prices rose very
rapidly. In most countries price inflation was at least reduced to tolerable
proportions by about 1942, and prices continued to rise little, or at least
at a much slower rate, until 1945. In the immediate post-war period
there was another bout of sharp inflation, but this appears to have levelled
off in most countries about 1948-49. A new cycle of sharper general price
increases was set off by the Korean war in 1950. At varying times between 1951 and 1954 prices stopped rising (or even fell temporarily) in
most countries (see table III). However, the movement was then resumed
and from 1955 until 1958 most countries were subject to continued
inflation.2 From 1958 to 1963 most of the countries in the group with
which we are concerned had price increases ranging up to 5 per cent.
per annum (section (b) of table III), with a tendency towards more rapid
increases in the last two years.
In few cases during even the steeper phases of this long upward
movement did the rate of price increase approach " hyper-inflation ", i.e. a
pace at which the public generally becomes unwilling to hold the national
money at all, and foreign currencies or physical commodities tend to
replace it as a medium. These cases were ones of extreme economic
and social disorganisation following upon wartime defeat. In the later
part of the period (since about 1953-54) the typical rate of rise of retail
1
The United States has had approximately stable prices since 1958, with an average
annual increase of only 1.2 per cent, in the retail price index and virtually no rise at
all in the aggregate index of industrial wholesale prices.
a
This was considered particularly disturbing because, unlike earlier inflationary
periods, it seemed to have no obvious explanation, and therefore the application of
conventional anti-inflationary policy instruments (fiscal and monetary measures) did
not appear suitable.

TABLE III.

(a) ANNUAL CHANGES IN CONSUMER PRICES AND (b) DISTRIBUTION OF ANNUAL PERCENTAGE CHANGES
IN CONSUMER PRICES IN 13 EUROPEAN COUNTRIES AND THE UNITED STATES, 1950-63
1950-51 1951-52 1952-53 1953-54 1954-55 1955-56 1956-57 1957-58 1958-59 1959-60 1960-61 1961-62 1962-63
Countries

(a) Annual percentage increase/decrease
is

13 European countries (median)
United States
Amount of increase/decrease

9.7
8.0

4.2

-0.2

2.2

0.8

1.0
0.3

1.6
-0.2

2.8
1.4

3.1
3.5

2.1
2.8

1.1
0.8

1.9
1.6

2.5
1.1

3.1

fri

1.2

(b) Distribution of annual percentage changes (number of countries)

Increase of—
20 per cent, or over . . . ,
10 to (less than) 20 per cent.
5 to „ „ 10 „ „
2 to „ „
5 „ „
Oto „ „
2 „ „

>
ö
1
1
11
1

Decrease of—
Up to 2 per cent
More than 2 per cent. . .
Source: Year Book of Labour Statistics, op. cit., and national sources.
1

4.4
1.2

2
o

Austria, Belgium, Denmark, Finland, France, Germany (Fed. Rep.), Italy, Luxembourg, Netherlands, Norway, Sweden, Switzerland and United Kingdom.

o
o
3
10
1

WAGES AND ECONOMIC STABILITY

prices in western industrial economies seems to have been between 2 and
5 per cent, a year, with only one or two countries falling outside that
range. Moreover, it is possible that, whereas official price indices for the
war years and immediate post-war period tended to understate price
inflation because the authorities were sometimes reluctant to adjust
them to changed epxenditure patterns, recent price indices may overstate it by neglecting rapid improvements in the quality and choice of
goods.
POST-WAR RECESSIONS

Before 1939 the fairly sharp price inflations—like those traced in
table H—which often occurred in boom periods were largely offset by
subsequent price reductions in periods of slump. During the post-war
recessions, which have been very mild by comparison with the major
inter-war or nineteenth-century slumps and comparable in their dimensions
with some minor recessions of the epoch before the First World War,
it is apparent from the broad view of price movements given by table III
that during the 1951-54 recession such reductions in national price
levels as occurred did not offset the continued national inflations. Again
in 1958, although recession was more severe, there was no significant
reduction of price levels in any of the 14 industrial countries surveyed,
and several continued to experience substantial average price increases.
In other words, not only has resistance to price cuts (the so-called
" downward rigidity of prices ") appeared to stiffen markedly over this
quite short period, but in some cases quite significant recessions made no
very immediate impact on the rate of inflation.
The reasons for the minor character of post-war recessions seem fairly
straightforward. The " floor level " for total national expenditures is
very much higher than before the war on account of continued high
military expenditures and a generally increased level of social spending.
Automatic offsets to depression, in the way of public or private unemployment benefits, have substantially increased. The general increase in
progressive income taxation acts in rather the same way, since a fall in
personal incomes before tax produces a much smaller fall after tax.
While the quantitative effect of these " built-in stabilisers " has perhaps
been relatively small, the existence of a general state of confidence in the
ability and determination of governments to prevent any large or protracted decline in economic activity has certainly been a factor in promotion of rather rapid recovery. In addition, relatively stable sectors of
employment—public utilities, nationalised industries, salaried employment in general—have greatly increased their share of the total. Up to

PRICES, WAGES, AND INCOMES POLICIES

the recession of the later 1950s the level of private investment in fixed
equipment (fluctuations in which were the most important factor in
major pre-war slumps) had been comparatively little affected by spontaneous swings. Indeed, abrupt changes in government military and other
expenditure arising from the disturbed international relations of the postwar years has on the whole been rather more important as a factor in
economic fluctuations. A significant feature of the 1957-58 recession,
however, was the re-emergence of swings in private investment outlay
as a big influence on changes in the level of activity. In the United States
particularly, where productive capacity had significantly exceeded requirements throughout the preceding boom, the sharp reduction of private
investment expenditures in 1957 and 1958 contributed heavily to the
fall in production and employment.
The causes of the earlier post-war recessions were, in a sense, abnormal. The 1948-49 United States recession was due mainly to the subsidence of the post-war replacement boom (which was more protracted
in Europe because of the much greater destruction and the severe and
widespread wartime reduction in real incomes); while the 1951-53
recessions in western Europe and the 1953-54 recession in the United
States were largely associated with the aftermath of the Korean boom.
The more recent United States and European recessions resembled normal
trade cycle slumps in the sense that they followed on an investment boom
and the appearance of excess capacity. It is too early to determine whether
the old, general business cycle has reappeared. Certain individual industries experienced specific cycles of varying lengths in the past and there
is some evidence that the short-term cycles characteristic of the textile
and vehicle industries are reasserting themselves.
More important, probably, is the appearance in the post-war recessions of certain new factors—or, rather, factors which were formerly of
comparatively minor significance. One is the increase in importance, in
total consumers' expenditure, of fluctuations in demand for certain
types of durable goods which, although representing a relatively small
proportion of total consumer spending, may cause fairly substantial
variations in total output and employment because of their impact on
key industries at the centre of complex modern economic systems.1
The general increase in real incomes means that a bigger portion of
personal spending goes on durable gooes, the purchase of which can be
1

In North-America, for example, the real consumption of durable goods (including motor cars as a major item) increased by 22 per cent, from 19S4 to 1955 and fell
by 8 per cent, from 1955 to 1956. The comparable figures for personal consumption as
a whole were increases of 7 per cent, and 3 per cent, respectively. See United Nations:
World Economic Survey, 1957 (New York), Part II, Ch. 4, table 45, p. 140.

WAGES AND ECONOMIC STABILITY

postponed if there is a temporary fall in real income or if people feel an
immediate need to increase their savings (for instance as a safeguard
against an anticipated slump). Moreover, an increasing proportion of
expenditure on durable goods is financed by hire-purchase and is thus
particularly sensitive to any general pressure to reduce personal commitments.
Another factor, however, is the much greater impact of government
policy on economic trends. Thus, while the recession of 1951-54 was
due in part to a reversal from the sudden spurt of rearmament spending
following the outbreak of the Korean war, it was probably aggravated
in some countries by government measures to stop the decline in exchange
reserves caused by balance-of-payments difficulties. Similarly, a significant element in the 1957-58 recessions in the United States and western
Europe was the impact of government restrictive policies intended to
halt the continuing price rises of the preceding period. Again, in some
countries these policies were prompted in part by balance-of-payments
deficits and, in the United States, by budgetary problems which led the
Government to reduce defence spending.
The effects of these recessions have obviously been very much less
serious in social terms than the inter-war slumps. However, they have
usually involved a fall in output and sometimes in productivity, as well
as an interruption in the general growth of the labour force which has
been a marked feature of the post-war period in most countries. The
1957-58 recession, particularly, was characterised by a quite substantial
decline in fixed investment, affecting the subsequent rate of growth
of productivity and therefore of real income. The most important effect
of recent recessions, therefore, seems to be a disturbance of economic
growth in general.
Against this it can be argued that short-term recessions of this type
may have some beneficial effects. First, if it is true that in a protracted
boom serious distortions in the pattern of labour and resource allocation develop, a fall in demand may lead to a transfer of idle or
underemployed labour and resources to industries where they will be
better used. Secondly, periodical interruptions in an upward trend of
prices may tend to preserve confidence in the monetary unit as well as to
correct excessively high costs.
But these benefits would depend very largely on the depth and duration of a recession in relation to the length and intensity of the previous
inflation, as well as on the extent of the cost rigidities which already
existed. In so far as the general pattern of the past generation has been
one of continuous expansion interrupted only by very moderate and short
recessions, it is likely to have encouraged a growth of rigidities and cost

G WAGES x AND CONSUMER PRICES, 1950-62

TABLE IV. POST-WAR TRENDS IN
Country

1950

1955

Austria 2 :
Money wages
Consumer prices
Real wages

1957

195S

1959

1960

1961

1962

105
102
103

112
104
108

118
107
110

123
112
110

131
113
116

138
114
121

151
116
130

164
121
136

175
125
140

Belgium :
Money wages3
Consumer prices
Real wages

100
100
100

112
109
103

117
111
105

121
111
109

128
114
112

136
118
115

139
119
117

144
120
120

148
121
122

154
123
125

163
124
131

Canada :
Money wages
Consumer prices
Realwages

100
100
100

113
110
103

136
113
120

140
113
124

147
115
128

155
118
131

160
121
132

166
123
135

172
124
139

177
125
142

181
126
144

Denmark :
Money wages
Consumer prices
Realwages

106
101
105

111
107
104

121
113
107

116
109

132
117
113

142
119
119

153
120
127

172
125
138

188
134
140

Finland :
Money wages
Consumer prices
Realwages

104
99
105

108
96
112

121
109
111

127
123
103

133
129
103

141
132
107

150
136
110

161
138
117

170
145
117

2
o
>

France :
Money wages
Consumer prices
Realwages

100
100
100

128
118
108

162
130
125

174
131
133

187
134
140

202
138
146

226
157
144

240
167
144

257
173
149

276
179
154

300
187
160

Germany (Fed. Rep.) *
Money wages
Consumer prices
Realwages

100
100
100

113
108
105

127
108
120

135
111
122

148
114
130

163
116
141

174
118
147

184
120
153

205
121
169

226
124
182

252
129
195

tri

r
o
w

Italy:
Money wages 6 .
Consumer prices
Real wages . . ,

100
100
100

110
109
101

115
114
101

118
116
102

122
120
102

129
122
106

138
126
110

145
128
113

151
132
114

155
132
117

162
134
121

173
137
126

200
143
140

Netherlands :
Money wages 6 . ,
Consumer prices .
Real wages . . .

100
100
100

108
109
99

111
110
101

113
110
103

131
114
115

135
117
115

148
118
125

164
126
130

164
129
127

170
131
130

192
133
144

218
135
161

238
139
171

Norway :
Money wages 3 . .
Consumer prices .
Real wages . . .

100
100
100

114
115
99

127
126
101

133
128
104

140
133
105

148
135
110

159
141
113

168
144
117

177
151
117

192
155
124

200
155
129

215
159
135

235
168
140

Sweden :
M o n e y wages '. .
Consumer prices .
Real wages . . .

100
100
100

121
116
104

143
125
114

150
127
118

156
128
122

168
132
127

183
137
134

194
143
136

206
151
136

215
152
141

228
158
144

246
161
153

244'
169
144

M
O

Switzerland :
Money wages . .
Consumer prices .
Real wages . . .

100
100
100

102
104

105
107
98

109
106
103

115
107
107

117
108
108

120
109
110

127
111
114

133
114
117

137
112
122

143
115
124

151
117
129

161
121
133

o
g
o
H

United Kingdom :
Money wages 3. .
Consumer prices .
Real wages . . .

100
100
100

110
110
100

118
120

124
123
101

132
126
105

143
131
109

155
138
112

165
143
115

170
147
116

177
149
119

193
150
129

205
154
133

214
162
132

United States :
Money wages 9. .
Consumer prices .
Real wages . . .

100
100
100

108
108
100

114
110
104

120
111
108

121
111
109

127
111
114

133
113
118

139
116
120

144
120
120

149
121
123

154
122
126

158
124
127

163
126
129

>
O
ta
on

o
z

Source: Year Book of Labour Statistics, op. cit., and International Labour Review (Geneva, I.L.O.), Statistical Supplement.
1
Average hourly gross earnings of male and female workers, incl., as a rule, juveniles. a Monthly earnings. 3 Adult males only. 4 Prior to 1960 excl. Saar. 5 Incl. the value
of
payments in kind. * Production and related workers; Oct. of each year. 79 Incl. mining and quarrying, Incl. holiday and sick leave payments and the value of payments in kind.
8
Excl. holiday and sick leave payments and the value of payments in kind.
Prior to 1959 excl. Alaska and Hawaii.

PRICES, WAGES, AND INCOMES POLICIES

pressures which would make a reallocation of labour and resources in a
recession more difficult, at least to the extent that it depended on adjustments in relative prices and costs. And, as will be seen in Chapter III,
the cost pressures themselves appear to have become more resistant to a
decline in demand. It seems, in any case, significant that in very few cases
have wage rates, or even hourly earnings and still less salaries, fallen
during any one of the post-war recessions. Indeed, they have usually
continued to increase, if at a reduced rate.
Against a general background of recurrent expansion rather than of
recurrent depression, therefore, it becomes more likely, as time goes on,
that at times when output falls money incomes will not fall or will fall
considerably less. The realism of this expectation is suggested by postwar United States experience—prices fell slightly in the 1948-49 recession,
were stable in the 1953-54 recession, and continued to rise, though
very slightly, in the 1957-58 recession.1 Since then, however, they have
remained approximately stable. But during this period unemployment
has remained high, in spite of very considerable growth of production
after the 1960-61 recession.
A special aspect of such situations is that the downward rigidity of
wage costs may have increased in other respects besides that of wage
rates per hour. On the one hand, various guarantees of weekly earnings
and, indeed, of future wage increases have been conceded in long-term
contracts during periods of high employment and strong labour bargaining power.2 On the other hand, the willingness or ability of employers
to dismiss workers seems to have been much reduced. Thus, it is much
more likely that in a recession output will fall more than employment,
and hence that productivity will decline. Total costs also are likely to
fall less than output, and this will mean that unit costs will increase—
a common experience when plants are working below capacity. In such
circumstances it becomes rather more likely that a recession may at
least temporarily intensify price inflation rather than reduce it, at least
if " administered " price fixing has developed to a point at which firms
attempt to adjust prices to cover total costs per unit (implying a downward rigidity of profits as well as of wages).
Of course, in the long run total costs will be reduced by the natural
wastage of the industrial labour force and of industrial plant (since
replacement is likely to lag) as well as by a declining ability to resist
1
It has, however, been pointed out that in 1957-58 part of the price increases were
due to bad weather causing higher food prices and to automatic increases under longterm wage contracts in key industries (O. ECKSTEIN in Economic Policy in Our Time
(Amsterdam, 1964), Vol. II, p. 38).
2
See below, p. 114.

WAGES AND ECONOMIC STABILITY

wage and profit reductions. But the long run is likely to get longer and
may be too long to be politically tolerable or (because of balance-ofpayments problems, for example) economically acceptable.
WAGES

The wages of manual workers have on the whole risen faster than
prices over the period, though there were some cases in which they lagged
during the war and others in which temporary lags have occurred since.
Table IV shows post-war trends in money and real earnings per hour in
manufacturing. Since the indices of real wages in table IV are composite
series obtained by dividing money wage indices by consumer price indices, they reflect the separate movements and characteristics of these two
series. In particular, the existence of black-market prices and rationing
in some countries throws doubt on the precision of the estimates given
for the earlier years, while for later years the increases in family earnings
associated with high employment levels and the growth of extensive
social benefits have probably increased the living standards of working
people more than the figures would suggest.
Comparable information about the movement of real wage rates (as
distinct from earnings) is not generally available. As shown in table V,
earnings have risen, in general, faster than wage rates, partly because of
overtime working, and also because of rising production bonuses and
many other factors. However, it seems clear that the general improvement in earnings is partly attributable to increases in average real rates
in several major cases, and such increases appear to have been particularly important since about 1953. They are, of course, not inconsistent
with falls in the real wage rates of particular sections of workers in individual countries.
There is, however, no evidence that post-war price inflation has tended
to restrict the growth of real wages in general. On the contrary, not only
was the inflation of the 1950s associated with rising real wages but this
rise appears in general to have been faster than the increase in all other
incomes taken together. Thus in all the 15 countries shown in table VI,
total wages and salaries rose faster than the total of other incomes over
the period 1953-54 to 1961-62, though there were three countries (Canada,
Switzerland and the United Kingdom) in which this was not true over
the latter half of the period. Of course, it does not follow that average
individual wages and salaries have risen faster than average other individual incomes—this depends on the relative increases in numbers of wage
and other income receivers. Furthermore, within the group of wages and
salaries, although the information available is inadequate to substantiate

TABLE V. THE WAGE DRIFT IN SELECTED EUROPEAN COUNTRIES, 1950-62
(Changes in hourly rates and average hourly gross earnings, 1958=100)
Country

Industrial coverage

Austria :
Rates 1
Belgium :
Rates

Sex

1950

1951

1952

1953

1954

1955

1956

1957

1958

1959

I960

1961

1962

man.
man., con.

m.f.
m.f.

91
85

95
91

99
97

100
100

105
107

112
114

119
126

125
137

man.
man. 5

m.f.
m.f.

100
105t 100

101
102

105
106

110
111

1163
119

France :
!• m.f.
m.f.

42
44

64
60

67
62

68
63

71
67

77
72

84
82

90
90

100 108
100t 106

115
115

124
124

133
135

man., mi., con., com., tran., serv. m.
man., mi., con., serv.
m.

63
53

69
61

73
66

76
69

77
71

82
76

88
83

94
92

100 104f 111
100 107f 117

120
129

132
143

man.
man.

m.f.
m.f.

69
66

75
73

78
76

80
78

83
81

86
86

91
92

95
96

100
100

101
102

106
108

111
115

123
131

man., mi., con.
man.

m.
m."

62
63

66
67

68
69

69
70

76
81

81
84

85
91

95
100

100
100

102
104

112 118
115f 133

129
146

man., mi., con.
man., com., tran.

m.f.
m. 10

85
78

89
84

90
86

91
88

95
92

97
96

100
100

102
103

105
108

110
116

116
125

man.
man.

m.
m.

64
59

71
64

75
69

76
73

80
78

85
85

92
91

97
97

100 101
100 105

106
114

108
121

112
126

< man., con., com., tran., serv.
Germany (Fed.Rep.)
Rates

Italy :
Netherlands :
Rates
Switzerland
Rates
United Kingdom :
Rates l l

Principal sources: German Federal Statistical Office: Statistisches Jahrbuch für die Bundesrepublik Deutschland and Series 12: Verdienste und Löhne in Ausland.
... = not available, f =* Restricted comparability of preceding with following figures. Abbreviations: man.=manufacturing; mi.=mining; con. = construction;
com.=commerce; tran.=transport; serv. = services; m.=male wage earners; m.f. =male and female wage earners
1
Collectively
agreed wages for married wage earners (with two children) in Vienna, after deduction of taxes and social security contributions payable by the worker, excl. family
a
allowances.
Vienna. 3 June. 4 Oct. of each year. 6 Prior to 1958, incl. mining and construction. e10Rates of pay actually applied, 1 Jan. of each year. 7 1950-52: Nov.; 1953-62
8
Sep. Prior to 1960, excl. Saar. * Data for 1961 and 1962 relate to male and female wage earners.
Skilled workers. " Weekly rates, 31 Dec. of each year.

TABLE VI. AVERAGE ANNUAL PERCENTAGE INCREASES IN WAGE AND SALARY INCOMES, AND IN OTHER INCOMES,
IN SELECTED COUNTRIES

Country

Austria
Belgium
Canada
Denmark
Finland
France 6
Germany (Fed. Rep.) 6 .
Ireland
Luxembourg
Netherlands
Norway
Sweden
Switzerland
United Kingdom . . .
United States

Period '

Wage and
salary
incomes 2
(percentage)

1953 -54/1957-58
1953 -54/1957-58
1953 -54/1957-58
1953 -54/1957-58
1953 -54/1957-58
1953 -54/1957-58
1953 -54/1957-58
1953 -54/1957-58

12.5
6.7
7.3
5.6
9.9
11.2
11.1
3.6

8.1
-0.1
4.1
4.3
8.0 "
9.5
11.0
1.8

1957-58/1960-61
1957-58/1961-62
1957-58/1961-62
1957-58/1960-61
1957-58/1961-62
1957-58/1960-61
1957-58/1961-62
1957-58/1961-62

7.7
4.3
4.9
9.5
9.9
11.0
12.1
6.9

4.0
3.1
5.2
7.4
8.8 4
10.0
8.1
5.5

1953--54/1957-58
1953--54/1957-58
1953--54/1957-58
1953--54/1957-58
1953--54/1957-58
1953--54/1957-58

11.1
7.8
7.9
6.2
7.3
5.3

7.9'
6.1 s
4.1
4.5
3.4'
4.7

1957-58/1961-62
1957-58/1961-62
1957-58/1961-62
1957-58/1961-62 '
1957-58/1961-62
1957-58/1961-62

7.8
7.7
8.5
7.7
6.1
5.1

7.0'
4.9 8
5.0
8.5
7.2'
4.1

Other
incomes 8
(percentage)

Period

Wage and
salary
Other
incomes 2 incomes 3
(per(percentage) centage)

Wage and
Other
salary
incomes a incomes 3
(per(percentage) centage)

1953-54/1960-61
1953-54/1961-62
1953-54/1961-62
1953-54/1960-61
1953-54/1961-62
1953-54/1960-61
1953-54/1961-62
1953-54/1961-62
1956-57/1960-61
1953-54/1961-62
1953-54/1961-62
1953-54/1961-62
1953-54/1961-62 8
1953-54/1961-62
1953-54/1961-62

10.4
5.5
6.1
7.3
9.9
11.1
11.6
5.2
5.0
9.4
7.7
8.2
6.9
6.7
5.2

7.3
1.5
4.7
5.7
8.4 4
9.7
9.8
3.7
2.9
5.9'
5.6 8
46
6.5
5.3'
4.5

Source: United Nations: Yearbook of National Accounts Statistics.
1

Average annual percentage increases are calculated as average annual geometric rates of growth between the two terminal dates. In order to minimise the influence of single
terminal years on the computed
growth rates, the data for the first two and the last two years are averaged and the rate is calculated as between the two averages. All incomes are
measured in current terms. 2 Incl. pay and allowances of members
of the armed forces before payments of taxes and deduction of social security contributions ( = " Compensation of
employees"
in
the
United
Nations
Yearbook terminology. 8 Income from unincorporated enterprises and income from property (rent, interest, dividends and corporate transfer payments). 4 Includes saving of corporations and deducts interest on consumers ' debt. 6 Change in statistical
techniques reduces comparability of pre-1959 with post-1959 figures. ° Inclusion
of the Saar and West Berlin figures from 1960 onwards reduces comparability with pre-1960 figures. 7 Net of interest on consumers'debt. 8 Excludes corporate transfer payments and income from unincorporated enterprises in industries other than agriculture, forestry, fishing, liberal professions and personal services. " Change in statistical techniques reduces comparability of pre-1960 figures with those for later years.

3
>

O
ta
Vi

•z
a
m
o
O

•z
o

o
CA

>
3
F
H

PRICES, WAGES, AND INCOMES POLICIES

any precise statement, it seems likely that during part of the period
average salaries have risen less than wages proper, though in more
recent years the reverse may have occurred.1 But the share in national
income of wages and salaries together appears in no case to have declined
over the period, and in one or two cases to have increased markedly.
PROSPECTS FOR ECONOMIC STABILITY

What is the probable balance of inflationary and deflationary factors
for the years ahead ? Some more or less foreseeable forces will strengthen
demand, or lead to cost increases, thus making for upward pressure on
prices. Others will tend to produce the opposite result. The former may
be expected to outweigh the latter.
First, population changes. The post-war recovery of reproduction
rates in most Western industrial countries suggests that the phase of
population decline indicated by the inter-war reproductive trends is
unlikely to materialise and that instead there will be a period of continued
population growth. A growing population involves a demand for new
investment. Structural changes in the populations of these countries may
also exercise an upward pressure in other ways. In several important
cases the proportion of old people is likely to rise quite sharply in the
near future. In the Netherlands, for example, those aged 65 and over are
expected to number 10.3 per cent, of the population in 1970 compared
with 8.9 per cent, in 1960; in Sweden the corresponding figures are 13.5
per cent, and 11.6 per cent., in France 12.7 per cent, and 11.6 per cent.
This is likely to make increased demands on social security. It may
mean a reduction, other things being equal, in net personal savings
through a greater volume of withdrawals by this sector of the population.
On the whole, social expenditures are likely to rise in any case.
Improvements in social security have become part of the programmes of
the national political parties. Demand for increased public expenditure on
education is created by the needs of modern industry, by the fact that
with higher incomes people can afford to have their children stay longer
at school, and the fact that countries do not want to fall behind others
in their educational level. Increased education, of course, also reduces
the total supply of labour, while raising its quality in compensation.
Modern motor transport is making very large and increasing investment demands for roads and parking facilities, and calls for extensive
replanning of cities. There also appears to be a steady rise in the demand
1

1964.

For United Kingdom experience see The Economist (London), 23 and 30 May

WAGES AND ECONOMIC STABILITY

for housing—quantitatively (space per person) and qualitatively. More
generally, potentialities for profitable new investment may be increased,
and are at least likely to be maintained. Technical development has proceeded at a remarkably rapid pace over the past generation and there is
no reason for supposing that it will lose its momentum within the foreseeable future; indeed the contrary seems more likely. New industries in
thefieldsof power production, transport and materials may develop fast,
and this is likely to be encouraged by the interest to which governments
have been stimulated by international political competition.
New investment may be stimulated by the general growth of longterm economic planning. At the level of individual enterprises this
arises from the changing character of private management and the
generally favourable domestic political-economic environment. At the
national level, long-term investment planning has been stimulated by the
general increase in the public sector. Opportunities for foreign investment
are opened by the development plans of newly industrialising countries.
Investment may also be stimulated by freer international trade, in some
cases through the opening of new markets, in other cases through
pressure for re-equipment to meet stronger international competition.
Political factors probably indicate a general expansionist tendency at
several levels. The degree to which a moderation of international tension
or other factors may lead to a substantial reduction in military expenditure is not predictable, even in the near future. But if direct defence
expenditure is reduced, the post-war experience suggests that attempts to
compensate for a reduced scale of armaments by a more rapid technical
replacement may still have a stimulating industrial effect. Moreover,
international political competition in other terms—of national economic
development and of assistance to underdeveloped countries—may in
some measure replace rival weapon-piling.
Among institutional developments which favour an expansionist
inflationary perspective is the change in the general character of management towards a professional or technological direction of enterprises,
which may make for investment plans favourable to long-run growth
rather than to short-run profit considerations. And with the growth
of planned production in larger enterprises, management generally may
become more reluctant to accept interruption of output by industrial
disputes, and thus more willing to make wage concessions.1 There has
been steady pressure from trade unions for longer paid holidays and
increasing pressure in recent years for shorter working hours with no
1

The impact of modern management views on price and wage fixing is further
discussed below: see pp. 62 ff. and 72 ff.

PRICES, WAGES, AND INCOMES POLICIES

reduction in weekly wages. The higher costs involved might be offset by
increased hourly productivity—it is not wise to be too pessimistic about
this possibility. But wage costs may be increased by more overtime
working when standard hours are reduced. Trade union organisation
may extend—if slowly—because an important feature of the present
century's growth in collective organisation has been that the rise of nonmanual trade unionism has continued fairly steadily, without much
interruption by recessions in employment ; and a very large proportion of
salaried employees still remains unorganised.1 Moreover, on the whole,
the employment market situation is likely to favour such groups ; in the
post-war period the demand for people with higher educational and
technological qualifications seems to have been visibly outrunning the
educational system's capacity to supply them.
Shortages in the labour market would be bound to lead to upward
pressures on the salaries (or wages) for the groups concerned. But when
the income structure is rigid, because maintenance of traditional relativities and relationships is a major criterion of wage and salary determination, such shortages could provoke significant general cost pressures.
A similar effect may follow from another consequence of inter-war
reproductive trends. In some countries there will soon be a relative shortage of certain key age-groups—particularly from the mid-20s to the mid30s—which may also push up wages generally.
Economic organisation is also likely to extend among other sections
of the community. The strength of organised farmer interests during the
past decades has been quite remarkable and in several countries similar
pressure-group organisation, is spreading to other self-employed people.

The effect of increasing productivity to offset such pressures may
diminish because the former may rise at a slowing rate. Attention has
often been drawn to the increase in the proportion of expenditure devoted
to services (where productivity has been increasing less than in physical
production) which accompanies general increases in real income.2
The factors which may reduce inflationary pressures—whether from
the side of the demand-resources relationship or from that of costs
and incomes—are, to some extent, the obverse of the above. Productivity
1

See below, p. 145.
See, for example, United Nations: Some Factors in Economic Growth in Europe
During the 1950s (Geneva, 1961), Ch. Ill, p. 57, table A8. Of 13 western countries there
was only one (Ireland) in which labour productivity in services increased as rapidly
over the period 1949-59 as it did in the economy as a whole. In some countries the
discrepancy was quite large. In Western Germany labour productivity in services
increased by 1.9 per cent, per annum, as compared with 4.9 per cent, in the economy
as a whole including services. In France the corresponding figures were 2.7 per cent.
and 4.3 per cent., in Italy 1.3 per cent, and 4.9 per cent., in the United Kingdom 1.1
per cent, and 1.8 per cent., in the United States 0.4 per cent, and 2.1 per cent.
2

WAGES AND ECONOMIC STABILITY

growth may actually increase, and operate effectively as a disinflationary
influence. The prospective increase in population means wider markets
and increasing returns in secondary industries. Investment may itself
become more productive because of the accelerated research expenditure
notable in recent years. The stage of diminishing returns in primary
industries has proved to be further off than was formerly supposed —
agricultural productivity has been rising quite fast in some countries
recently, and substitutes for natural raw materials are already cheaper
in several cases. It may be that the tendency for productivity to rise more
slowly in tertiary than in secondary industries will not continue. The
application of " scientific " industrial methods to administration and
service occupations is comparatively recent, and it is perhaps not impossible that mechanisation and rationalisation in these fields will halt or at
least slow down the general tendency towards an increase in the proportion of employees not engaged in actual production.
The propensity to save may increase because of generally rising real
incomes.1 (It is possible that until quite recently this effect was concealed
by that of greater income equality.) The growth of institutionalised
" personal " saving may also increase saving in general by making it more
easy and secure—or more difficult to avoid (as in the case of industrial
pensions schemes).
The prospect of manpower scarcity in some age brackets of the
labour force may be more than offset in the quite near future by sharp
increases in the supply of young people as a result of the wartime and
post-war stimulus to reproduction. At the other end of the scale, medical
interest is becoming as much concerned to extend adult life and capacity
as to reduce infant mortality, and the rise in the proportion of old people
may be balanced by an increase in their ability and willingness to stay
at work.
In some respects economic competition is likely to grow. Increased
variety in consumer goods has usually accompanied rising living standards, and this—combined with an increased propensity to save—may
make the markets for the products more competitive. Freer international
competition may also restrain price increases. It might create significant
structural unemployment in countries that failed to adopt active policies
to promote and facilitate adaptation. In the quite near future the centrally
planned economies are likely to appear as important new competitors in
international trade in manufactures—or even services (tourism, finance,
etc.)—as they already have done for certain primary products.
1

But see H. F. LYDALL in Australian Economic Papers, Vol. II, No. 2, Dec. 1963,
pp. 228 ff.

PRICES, WAGES, AND INCOMES POLICIES

On the cost side, the assumption that institutional development (union
organisation, collective bargaining, etc.) is likely to encourage costinflationary pressures may be wrong. The general trend appears to be
towards greater centralisation of wage fixing, and more governmental
participation in it. This might slow down the rise of money wages. The
possibility of sectional gains from cost inflation depends on the relative
incapacity of other groups to retaliate. Conceivably, the general growth
of pressure group organisation may have reached in some countries a
position of balance in which these possibilities tend to cancel each other
out. This may make the need for voluntary restraints on cost inflation
more apparent to the groups concerned. It may also increase the practical possibility of achieving such restraint, with or without government
participation in the process as mediator between opposing interest
groups.1
These are factors which may limit the inflationary impact of continued
expansion or restrict expansion itself. There remains the possibility of
general contractions. Any large reduction in military expenditures might
set off a broader decline in economic activity because of its more than
proportional effects on military stocks, production and investment. The
problem would be much more limited than, and also quite different
from, that of " reconversion " after the Second World War because in
that case the immense arrears of civilian demand involved an upward
" accelerator " which offset the downward military one.
Moreover, the general business cycle (as opposed to short-term or
sectional cycles, which are likely anyway) might reappear; and it is possible that we overestimate the capacity of modern economic techniques to
deal with it. The Keynesian prescriptions have not yet been tested against
a major slump; at any rate, neither they nor more old-fashioned devices
have succeeded in eliminating inflation. Even assuming the basic analysis
to be correct, it is not obvious that a sufficient array of anti-depression
measures is ready. Attempts to stimulate consumption directly may not
be very effective in a recessionary situation, and are likely to be inadequate to offset a really steep fall. It would seem amply possible to offset
a collapse either in military demands or in private investment by social
investment, of which there are great arrears in some countries, and for
which in all countries large and justifiable opportunity exists. But, even
if detailed plans for such counter-slump activity are prepared (and they
usually are not) a lag before they exercise a sizeable effect on economic
activity is inevitable. And the lag might be sufficient to destroy that confidence in the government's capacity to maintain high levels of demand
1

Economic Policy in Our Time, op. cit., Vol. I, pp. 174 ff.

WAGES AND ECONOMIC STABILITY

which seems to have been an important element in the post-war buoyancy
of the economic situation.
This can be perhaps reckoned as an outside risk. Another possibility,
however, involves certain of the same elements. The failure of governments generally to reconcile expansion with price stability may stem
not from the imperfection of economic techniques themselves but from
the extent to which their impact is modified by psychological factors
which are by nature incalculable (" expectations ", " confidence " and the
like). It is possible, moreover, that the balance of public pressures on
government policy is changing as memories of mass unemployment recede
and inflation replaces it as the more familiar economic evil. Thus, fear
of inflation might hold back a government from adequate measures to
forestall an incipient recession until it was too late to prevent this becoming a major slump. Particularly, perhaps, there is a possibility of
periods of economic stagnation, because while governments will still be
unwilling to permit a recession to go very deep (and able to prevent it
from doing so), they may at the same time hesitate to launch measures
which would pull their economy quite out of it, from fear of inflation.
No clear conclusion emerges from these considerations. But it may be
thought that the general balance of economic, social, political and institutional factors favours a long-run growth of demand, and continued
inflationary pressures. There is, however, a possibility of not infrequent
interruption by limited recessions; and a marginal possibility that any
one of these might develop into either a substantial slump or a period of
economic stagnation. This possibility would be increased by a failure to
resolve the problem of combining expansion with price stability, since
that failure might inhibit adequate government action against incipient
or actual unemployment. The desire to maintain economic growth and
to avoid major recessions thus provides the best reason for developing
an effective control of inflation.

CHAPTER II
EFFECTS OF INFLATION
THE MEANING OF INFLATION

Strictly speaking, the term " inflation " means an increase in the supply of money in relation to that of commodities. It is thus usually and
mainly associated with a sustained rise in one or more price levels (wholesale, retail, export, etc.), and price increases are the most important
aspect of inflation to be examined in this study. But it is useful to bear in
mind certain other possible connotations—especially those of generalised
excess demand, and excess demand for foreign exchange at official rates,
the latter being, of course, a common consequence of the former.1 As
noted in the previous chapter, balance-of-payments problems have
frequently been the occasion for government action to restrict demand
and thus, probably, to contribute to the occurrence of recessions.
Sustained price increases, whether due to excess demand or to rising
wages and other costs, are a possible cause of balance-of-payments
problems, and preventing or halting them is often necessary for this
reason alone.
The present chapter is concerned with the effects of a condition of
more or less continually rising prices from whatever causes. As mentioned
above, there are several price levels in which a sustained rise may be
considered indicative of inflation. What is the best indicator in any
particular case depends on the purpose of the inquiry. Thus, wholesale
prices are more relevant to problems of foreign trade and the balance of
payments, while retail prices are of greater interest to the consumer and
therefore to a number of problems related to the cost of living.
The various price indicators need not move in a parallel fashion. For
example, when wholesale prices are stable the consumer price index
will normally rise slightly because of increases in the prices of services
1
Some writers prefer to regard a state of generalised excess demand as the essence
of inflation and rising prices as a common but not inevitable symptom of it. When
there is widespread and effective price control, excess demand may be cut back through
queues or rationing instead of through price increases. Black markets as well as lengthy
delays in meeting demands are, however, characteristic features of such " suppressed
inflations ".

EFFECTS OF INFLATION

that are not included in the wholesale price indices and, consisting largely
of wages and other earned incomes, tend to rise at the same rate as living
standards.
A number of effects of rising prices are discussed below—effects on
the cost of living, on the distribution of income and wealth, on social
security, on savings, on the ease or difficulty with which certain social
conflicts can be resolved, on the rate of economic growth and on international balances of payments.
PRICES AND THE COST OF LIVING

The recent movements of consumer price indices in a number of
countries are shown in table VII. It will be seen that of the 18 countries
in the table, only six experienced increases of less than 50 per cent, between 1948 and 1963. Eight had increases of from 50 to 100 per cent.,
four had increases of over 100 per cent. There has been a general slowing
down since 1953, and 12 out of the 18 countries experienced slower
TABLE VII. CONSUMER PRICE INDICES

1953
(1948 = 100)

Country

Canada
Finland
Germany (Fed. Rep.) .
Italy

New Zealand
Norway

....

United Kingdom . . .
United States . . . .

177
165 l
106
119
123
157
144 !
108
126
117
121
129
134
136
130
104
130
111

1958
(1953 = 100)

113
114
108
108
117
128
121
110
117
113
107
117
118
118
119
107
119
108

1963
(1958 = 100)

109
114
106
106
121
117
126
112
110
117
105
113
111
114
115
111
112
106

(1948 = 100)

218
214 !
121
136
174
253
219 J
133
162
155
136
170
175
183
178
123
173
127

Sources: Year Book of Labour Statistics, op. cit; and United Nations: Monthly Bulletin of
Statistics, op. cit.
'Base: 1949 = 100.

PRICES, WAGES, AND INCOMES POLICIES

increases from 1958 to 1963 than from 1953 to 1958. Except for Denmark, Finland, France and Sweden, no country had an average yearly
increase as high as 3 per cent, over the period 1953-63.
A rise in the cost of living is an increase in the cost of maintaining a
certain level of living in the elusive, but real, sense of a certain level of
satisfaction of wants. A consumer price index registers changes in the
cost of buying a certain collection of goods and services. While consumer
price indices are useful indicators of changes in the cost of living, they
are not perfect for this purpose because buying a constant collection of
goods is not the same as maintaining a constant level of satisfaction.
Indeed, for a number of reasons which we will examine, it is usually
advantageous for the consumer to change the pattern of his purchases
over time. And for each of these reasons a rise in the consumer price
index tends to overstate the increase in the cost of living at any date compared with earlier dates. (These reasons apply to consumers in all income
categories, not only to wage earners. They do not constitute reasons for
exercising more restraint over wages than over other incomes, but need
to be borne in mind when evaluating tie hardship entailed in a rise in
the cost of living.)
First, individual prices vary at different rates and even in contrary
directions. Thus, when beef becomes more expensive but pork does
not, people are likely to buy more pork—thereby revealing that they
regard this as a lesser evil (i.e. as a smaller fall in satisfaction) than
increasing their expenditure on meat by the full amount of the price
increase for beef. Hence, the increase in the cost of living is less than
a consumer price index assuming constant beef consumption would
suggest.
Second, consumer price indices do not reflect the continuing appearance on the market of new products. Suppose that the price of men's
worsted suits has not changed for some years but that one can now buy
(for the same, a lower or some higher price) a suit of man-made fibre
that is washable and crease-resistant. If people buy the new things, this
shows that they think they get better value for their money by doing so ;
the cost of living would seem to have gone down, but the consumer price
index will not show it. Or suppose that plastic buckets can now be bought
for 10 cents instead of the conventional zinc pail costing 1 dollar.1
Clearly buckets have become cheaper, although one would hesitate to
say that there was a price fall of 90 per cent., but the consumer price
index may not show any reduction at all.
Third, the consumer price index does not normally reflect improve1
Willard L. THORP and Richard E. QUANDT: The New Inflation (New York,
McGraw Hill, 1959), p. 7.

EFFECTS OF INFLATION

ments in quality.1 If, for example, medical services or motor cars have
increased in price but have clearly improved in quality over a period of
time, then in terms of consumers' satisfaction the fall in the purchasing
power of money has been less than the rise in prices would suggest.
For these reasons it may be concluded that consumer price indices
tend to give an exaggerated impression of increases in the cost of living.
Indeed, when the index shows very small increases it is legitimate to ask
whether the cost of living has risen at all. An annual increase in the index
of 1 per cent, or perhaps 2 per cent, might be fully matched by improvements in quality and the availability of new products. It is difficult to
say to what extent this has actually been the case.2 But the reality of the
factors mentioned may be illustrated by imagining that an individual
were given $1,000 and a choice of ordering goods either from an early
post-war department store catalogue (say 1948) or a current (1966)
catalogue.3 The 1948 catalogue would have substantially lower prices,
but also less advanced products. Different people would undoubtedly
answer this kind of question differently, but despite the increase in the

prices of consumer goods it is by no means certain that an overwhelming
majority would choose to spend their money on 1948 products at 1948
prices rather than on 1966 products at current prices, even if they confined
their choice to things that do not go out of fashion. And this comparison
limited to products would not take account of new or improved services
in the medical and educational fields, etc., or of the possibly higher
value of government services provided per dollar or franc of taxes paid.4
PRICE INCREASES AND THE DISTRIBUTION OF INCOME AND WEALTH

This leads to a discussion of another and very familiar effect of price
increases—namely, that they lead to or are associated with changes in
1
" There are certainly some cases where a price rise is disguised by rather superficial quality improvements
Furthermore, there are cases, though infrequent, of a
deterioration in quality, which may be insufficiently allowed for in recording price
changes. These considerations partly offset the failure to make full allowance for
quality improvements." (Organisation for European Economic Co-operation: The
Problem of Rising Prices. A report by a group of experts (Paris, 1961), p. 27.)
2
A group of experts who examined the course of prices in western Europe, Canada
and the United States in recent years felt that " the upward bias in the price indices is
not likely to have been enough significantly to affect the indices during the period
under review " (ibid., p. 27). This group, however, considered only the element of
bias introduced by inadequate allowance for quality improvements.
8
See Richard and Nancy D. RUGGLES in United States Congress Joint Economic
Committee : The Relationship of Prices to Economic Stability and Growth, Compendium
of papers submitted by panellists, 31 March 1958 (Washington, 1958) p. 298.
4
During the period 1947-58 productivity in a sample of United States federal
government agencies was found to have risen by about 2.2 per cent, a year (Henry D.
LYTTON in Review of Economics and Statistics (Cambridge, Mass.), Vol. XLI, No. 4,
Nov. 1959, pp. 341-359.

PRICES, WAGES, AND INCOMES POLICIES

the distribution of income. At times when consumer prices are rising,
most money incomes tend to rise too, but not equally or without exception. Some people's incomes rise faster than prices, others more slowly,
if at all. Who may be in the former group and who in the latter depends
to some extent on the cause of the price increases. For example, when
there is much excess demand, increases in sales at constant (and, still
more, at rising) profit margins first yield rising profits, wage increases
ending to follow somewhat later. If prices rise as a consequence of
wage increases one might expect the reverse to happen.
It has already been noted (table VI) that the share of national incomes
going to wage and salary earners in the countries with which this study
is concerned has been increasing, but that it is difficult to know how
far this is accounted for by an increase in the proportion of wage and
salary earners to total income receivers. Further details on income
distribution are shown in table VIII. Total compensation of employees
ranges from below 60 per cent, of the national income in Belgium and
Ireland and (in some years) Austria, France, the Federal Republic of
Germany and the Netherlands, to over 70 per cent, in the United Kingdom and (since 1957) the United States. In the great majority of the 13
countries included in table VIII the share of employees' compensation
was higher on the average of the last three years of the eight-year period
1955-62 than on the average of the first three years. A part at least of the
explanation for this is probably to be found in the continued shrinkage
in the proportion of the labour force engaged as small-scale proprietors
in agriculture, and their transfer to wage-earning or salaried employment.
For some purposes it is appropriate to group together employees'
compensation and income from unincorporated enterprises. Most of the
latter incomes are small, and there is no doubt that the greater part of
income in this category is in fact a retura to the labour, managerial and
productive, of small-scale owner-managers, and corresponds to the
wages or salaries they would have earned if they had not been working
on their own account. This category of incomes is affected by fluctuations,
particularly those due to variations in agricultural prices and yields,
in the profits of persons working on their own account. Even so, if
we take employees' compensation and income from unincorporated
enterprises together, as constituting a better measure than the former
alone of the share of income from work in national incomes (and call
this total "pay")1, we find that (a) the share of "pay" varies much
less from country to country than the share of employees' compensation,
1
Following the convenient terminology adopted in E. H. PHELPS BROWN and
M. H. BROWNE in Economic Journal (London), Vol. LXX, No. 280, Dec. 1960, p. 731.

EFFECTS OF INFLATION

TABLE VIII.

DISTRIBUTION O F THE NATIONAL INCOME
A T C U R R E N T COST, 1955-62
(Percentage shares in the national income)

Country

1955

1956

1957

1958

1959

1960

1961

61.5

61.6

64.6

62.9

63.0

(A.2

64.9

21.6
83.1

18.3
82.9
9.0
2.4
2.2
3.1

20.0
82.9
9.0
2.4
2.4
2.9

19.1
82.1
9.1
2.2
2.4
2.9

18.5
82.7
9.2
2.3
2.5
2.8

17.7
82.6

3.0

22.0
83.6
8.5
' 2.2
. 2.0
2.9

58.4

60.1

60.2

60.6

61.8

59.6

60.8

33.2

31.5
91.6
0.5
-0.4

31.2

91.6
0.5
-0.4

91.4
0.7
-0.4

30.7
91.3
1.0
-0.4

28.3
90.1
1.3
-0.3

29.2
88.8
1.5
-0.2

29.3
90.1
1.9

> 0.7

1.0

1.1

1.3

1.6

1.7

Í 1.0
\ 0.9

53.9

55.8

57.4

56.6

57.1

57.0

29.7
83.6
13.7
6.8
3.9
2.8

27.6
81.5
14.8
8.3
3.3
2.9

26.9
82.7
14.4
7.9
3.3
3.0

26.5
83.9
14.5
7.9
3.6
2.7

26.7
83.3
14.4
7.8
3.9
2.5

26.2
83.3
14.3
7.3
4.1
2.7

26.4
83.4
14.4
7.0
4.3
2.9

65.8

66.2

68.9

68.1

67.9

68.5

69.5

14.5
80.3

14.7
80.9
9.0

12.7
81.6
9.7

13.3
81.4
9.9

12.6
80.5
10.6

12.4
80.9
11.1

11.5
81.0
11.4

• 7.8

7.3

7.9

8.1

8.9

9.3

9.6

1.5

1.4

1.5

1.6

1.5

1962

Australia x
1. Employees' compensation pretax
2. Income from unincorporated
enterprises 2
Total of 1 and 2. .
3. Income from property . . . .
of which: Rent
Interest
Dividends
. . . .

8.6
. 4.3

9.6
2.5
2.7
2.8

Austria :
1. Employees' compensation pretax
2. Income from unincorporated
enterprises 3
Total of 1 and 2. .
3. Income from property . . . .
of which: Rent
Interest
Dividends*
. . .
Belgium :
1. Employees' compensation pretax
2. Income from unincorporated
enterprises
Total of 1 and 2. .
3. Income from property . . . .
of which: Rent
Interest
Dividends6
. . .
Canada :
1. Employees' compensation pretax
2. Income from unincorporated
enterprises
Total of 1 and 2. .
3. Income from property . . . .
of which: Rent
Interest
Dividends
. . . .
For footnotes see p. 35.

9.6

(Table continued overleaf)

PRICES, WAGES, AND INCOMES POLICIES

Country

Finland :
1. Employees' compensation pretax
2. Income from unincorporated
enterprises
Total of1 and 2...
3. Income from property . . . .
of which: Rent
Interest
Dividends . . . .
France :
1. Employees' compensation pretax 6
2. Income from unincorporated
enterprises
Total of 1 and 2...
3. Income from property . . . .
of which : Rent
Interest
Dividends . . . .
Germany (Fed. Rep.) 7 :
1. Employees'
compensation pretax 8 9
2. Income from unincorporated
aivu
enterprises 8
3. Income from property.

Ireland :
1. Employees'compensation pretax
2. Income from unincorporated
enterprises
Total of 1 and 2...
3. Income from property . . . .
Luxembourg:
1. Employees' compensation pretax
2. Income from unincorporated
enterprises
Total of 1 and 2...
3. Income from property . . . .
of which: Rent
Interest . . . . 1
Dividends . . . J

1955

1956

1957

1958

1959

I960

61.3

62.3

63.1

62.3

63.7

63.1

23.9
85.2
4.4
1.3
1.9
1.1

23.7
86.0
4.4
1.3
1.9
1.2

22.6
85.7
4.3
1.3
2.1
0.9

23.0
85.3
4.1
1.3
1.9
0.9

23.3
87.0
3.6
1.2
1.7
0.5

23.8
86.9
3.3
1.2
1.6
0.5

58.4

59.2

59.1

59.0

59.5

58.3

30.0
88.4
7.2
2.2
2.2
2.8

29.5
88.7
7.0
2.0
2.1
2.9

29.8
88.9
6.3
1.7
1.8
2.7

29.8
88.8
6.0
1.5
2.0
2.4

28.8
88.3
6.1
1.8
1.8
2.3

29.4
87.7
6.0
2.0
1.8
2.1

59.6

59.5

59.7

60.5

60.2

60.8

32.2

32.4

31.9

31.4

31.6

30.Í

53.0

57.0

55.2

55.6

54.6

55.0

31.6
84.6
10.6

29.4
86.4
10.0

31.4
86.6
10.9

28.7
84.3
11.5

29.4
84.0
11.8

28.6
83.6
11.2

57.6

58.5

58.6

63.3

61.4

58.3

26.6
84.2
6.7
2.5

27.0
85.5
6.8
2.5

26.7
85.3
6.8
2.4

27.5
90.6"
6.3
2.7

26.0
87.4
7.4
3.1

23.8
82.1
7.6
3.1

3.6

3.8

3.3

3.5

EFFECTS OF INFLATION

Country

Netherlands :
1. Employees' compensation pretax
2. Income from unincorporated
enterprises
3. Income from property10 . . .

1955

1956

1957

1958

1959

1960

1961

53.5

55.3

56.7

57.8

56.7

56.6

58.9

}3,4

34.7

33.2

33.4

32.7

32.8

31.4

60.8

62.2

61.8

61.3

61.6

60.7

61.3

19.4
80.2
10.5
3.7
3.4
3.4

18.0

80.2
10.0
4.0
3.1
3.1

18.4
80.2
9.7
3.7
3.0
3.0

18.6
79.9

18.5
79.2
11.5
3.2
5.4
2.9

18.3
79.6

3.6
3.6
2.8

18.4
80.0
9.9
3.4
3.4
3.0

72.7

73.0

73.2

72.8

73.0

73.7

74.4

9.4
82.1
10.2
1.4

9.0
82.0

8.9
82.1
8.5

8.7

81.7
9.6
1.7

8.6
82.3
10.5
1.6

8.3
82.7
10.8
1.6

Switzerland :
1. Employees' compensation pretax
2. Income from unincorporated
enterprises
Total of 1 and 2...
3. Income from property . . . .
of which: Rent
Interest
Dividends . . . .
United Kingdom :
1. Employees' compensation pretax
2. Income from unincorporated
enterprises
Total of 1 and 2...
3. Income from property10 . . .
of which: Rent
Interest
Dividends . . . .
United States :
1. Employees' compensation pretax
2. Income from unincorporated
enterprises
Total of 1 and 2...
3. Income from property . . . .
of which: Rent
Interest
Dividends . . . .

10.0

11.2

2.9
5.4
2.9

1.1
7.2

;.;
7.4

8.6
81.4
9.1
1.6
7.4

68.4

69.8

70.3

70.7

70.3

71.7

71.8

12.8
81.2
11.9
3.2
4.8
3.4

12.5
82.3
12.1
3.1
5.0
3.5

12.2
82.5
12.6
3.2
5.4
3.4

12.6
83.3
13.0
3.3
5.7
3.4

11.7
82.0
12.8
3.0
5.9
3.4

11.2
82.9
13.3
2.9
6.3
3.5

11.4
83.2
13.5
2.8
6.5
3.6

8.3

7.8

9.1

Source: Derived from Yearbook of National Accounts Statistics, op. cit.
. . . = Not available. — = Negligible.
The percentages do not add up to 100 because the concept of national income includes in addition income received
by financial intermediaries, corporate savings, direct taxes on corporations and general government Income from property
and entrepreneurship, minus interest on public and consumers' debt.
1
Fiscal year begins in July. * Changes in farm livestock are not taken into account. s Incl. a statistical discrepancy.
4
Incl. corporate transfer payments. 67 Incl. directors' fees. * Figures after 1959 are anot strictly comparable with those
before because of statistical revision. After 1960 the Saar and West Berlin are incl. Income from professional services
is included in employees' compensation. 9 Excl. supplements, other than those paid directly to employees or to the
government. " Net of interest on consumers* debt.

PRICES, WAGES, AND INCOMES POLICIES

mostly accounting for from 80 to 85 per cent, of the total (but for
rather more in Finland and France and considerably more in Austria
and the Federal Republic of Germany); (b) there was no discernible
trend towards a change in the share of " pay " over the short period
considered; and (c) there was no discernible association between
changes in the share of " pay " in national incomes on the one hand and
more or less rapid increases in consumer prices on the other.
This confirms the results of an earlier inquiry covering 25 countries
over the period 1950-57, the authors of which conclude that—
Inflation does not bring with it any strong tendency to raise pay relatively
to the return to property, or vice versa, but is compatible with shifts in either
direction. Perhaps we can go farther, and surmise that resistances to inflation,
which hold back prices or money incomes in some sectors of the economy, tend
to bring distributive shifts, but that where inflation is least impeded and most
general its distributive effects (between those broad categories alone that we
are considering here) are least. ... Over the whole field, what is remarkable
is the absence of any general distributive change in the presence of general
inflation.1
It is, of course, entirely compatible with this conclusion that there
may have been large redistributional shifts within the broad income
categories distinguished. Within the category of incomes from work,
there was rather generally a sharp narrowing of wage differentials on
the basis of skill during the war and the years immediately after it, and
of differentials between weekly wages and monthly salaries. More recently
these differentials have tended to become wider again in many countries.
In different sectors of the economy, profits and wages may rise sooner
or later, more or less, than consumer prices ; but they do in general tend

to rise, under the influence of market forces and collective bargaining,
at times when consumer prices are rising. No similar forces, however,
operate at such times to push up certain other categories of incomes,
notably those of retirement pensioners, people in receipt of other social
security benefits and rentiers.
People drawing social security benefits or private pensions have little
control over their income. The benefits may depend on legislation, or
have been stipulated in some contract, or they may be at the discretion
of the administrators of company pension and welfare funds. Since rising
prices are often accompanied by rising real income per head, arrangements may be made to provide pensioners or other recipients with
increases in money incomes at least sufficient to prevent a fall in the real
value of the benefits. Unless there is an increase in the proportion of
social security beneficiaries in the total population, the whole of the rise
1

PHELPS BROWN and BROWNE, op. cit., pp. 731-732 and

745.

EFFECTS OF INFLATION

in real income per head could still accrue to the active population,
without this being at the expense of the standard of living of the pensioners.1 Or benefit rates may be linked to indices of money wages, ensuring
that social security beneficiaries share in the same proportion as wage
earners in general improvements in the standard of living.
The problem of adaptation of social security benefits to rising prices
is particularly important in connection with long-term benefits (pensions),
though the question also arises in respect of short-term benefits (e.g.
sickness and unemployment benefits), especially when these are not wagerelated.
Under most statutory social security schemes there is some machinery
for adapting pensions to changes in the general level of wages or in the
cost of living. Three methods of adaptation may be distinguished:
(1) ad hoc adaptation, where the law does not contain any provision
for adaptation; (2) adaptation in principle, where the law provides for
periodic review of the problem of adaptation of pensions to economic
variations without specifying the procedure, mechanism or degree of
adaptation; and (3) systematic adaptation, where the law prescribes the
procedure, mechanism and degree.
An example of the first method is Australia, where old-age pensions
under the general scheme are fixed at flat rates and changed by amendment of the law from time to time in the light of general economic
conditions, budgetary commitments, etc. The rate of benefit increased
by 147 per cent, from 1950 to 1963 2, during which period the cost of
living increased by 83 and the general level of wages by 107 per cent.
The flat-rate pensions under the National Insurance Scheme in the
United Kingdom may be considered to belong to the second category.
The law provides forfive-yearlyreviews of the scheme, at which time the
benefit rates are to be reviewed. In practice, the pension rates have been
revised more frequently. The rate of retirement pension (for a single
person) rose by 160 per cent, from 1950 to 1963 3, while the increase in
the cost of living was 64 per cent, and the increase in the level of wages
121 per cent, during the same period.
Among schemes of the third type could be mentioned the general
pension scheme in Luxembourg, where pensions are adjusted by the cost1
If inflation were to be gradual and indefinitely prolonged, it might be argued that
one generation's advantage at the expense of the old would be matched by its being
disadvantaged in relation to the next generation. See H. S. HOUTHAKKER: Protection
Against Inflation, Study Paper No. 8 for the Study of Employment, Growth and Price
Levels by the United States Congress Joint Economic Committee (Washington, 1959),
p. 119.
2
From £2 2s. 6d. a week in 1950 to £5 5s. Od. in 1963.
3
From £1 6s. Od. in 1950 to £3 7s. 6d a week in 1963.

PRICES, WAGES, AND INCOMES POLICIES

of-living index, being increased or reduced by five points when the index
shows a rise or fall of five points on the average for the previous sixmonthly period.
In the Netherlands the general state old-age pension is linked to an
index of wage rates, thus ensuring adjustments to price changes to
approximately the same extent as those obtained by wage earners, and
also increases in the purchasing power of the pensions as real wages rise.
In France pensions under the general scheme for industrial and
commercial workers are adjusted each year according to the average of
insured wages for the previous year. As insured wages are subject to a
ceiling, the pensions would not necessarily follow the actual general
level of wages. In fact, while the cost of living increased by 52 per cent.
from 1955 to 1963, and the general level of wages by 92 per cent., benefit
rates increased by 137 per cent. In this case, as in others cited above,
pension rates were not merely adjusted to price increases but were also
raised for reasons unconnected with them.
An even more refined system of adjustment of pensions was introduced in 1957 in the Federal Republic of Germany. An author describing
this system commented that the task of the pension is to maintain the
living standard achieved by employees during their active Ufe, though
taking account of reduced needs. In addition, the retired worker should
also share in the further economic development in the founding of which
he co-operated.1
Nevertheless, although there are therefore a number of social security
schemes under which increases in benefits have matched or exceeded price
increases, this is not always the case, and a relatively large increase in
benefits during a given period may in fact merely correct their depreciation in an earlier period. In the United States, old-age and widows'
pensions were not changed between 1935 (when the Social Security Act
was introduced) and 1949. As a result, between 1940 and 1949 their real
value fell by about 40 per cent.
Little information is available on adjustments under private pension
schemes, but it is likely that they occur less frequently than under public
schemes. Under one quite widespread type of private scheme, however,
there is a measure of automatic adjustment to general increases in pay.
In schemes of this type, which have sometimes been set up by collective
agreements and are quite common for salary earners in several countries,
a contribution equivalent to a given percentage of each employee's
current wage or salary is paid by the employer (or by the employer and
1
Heinz EICHER in Bundesarbeitsblatt, Vol. 11, No. 1,1960, p. 19, where the methods
for achieving these ends are also described.

EFFECTS OF INFLATION

employee together) into a fund or insurance scheme from which pensions
are paid on retirement. The pension received by an employee is thus
proportionate to his or her total earnings during working life under the
scheme, and the average of the pensions paid rises with the general level
of wages or salaries concerned. However, for each individual the pension
received remains constant after retirement, so that employees covered by
such schemes may suffer a steady erosion of their living standards through
rising prices once they have actually retired.
Moreover, not all retired or disabled people are covered by social
security schemes or private schemes of the type referred to, and even
those who are may suffer losses from the falling real value of interest
payments derived from their savings (e.g. from life insurance annuities, or
houses subject to rent controls) since price increases favour debtors
and inflict losses on creditors.1 Thus, it has been estimated that between
1939 and 1952 about $500,000 million of debts (at 1952 prices) were
wiped out in the United States by price increases2—nearly equivalent
to one-and-a-half times the 1952 gross national product. The precise
effects of the distortions of the debtor-creditor relation are complex.
Most households and firms are simultaneously debtors and creditors.
Also, as noted in the next section, the real value of some debts may be
maintained to some extent by " index-linkage ".
It used to be thought that businessfirmsand banks were the principal
net debtors in industrialised countries and therefore the main beneficiaries
of inflation. But in the United States at least this no longer (i.e. since the
Second World War) seems to be the case. The principal debtor now
appears to be the Government, and the principal creditor the household
sector. The main redistributive effects of inflation occur, therefore, between these two. As noted in the United States study referred to above,
since people holding government bonds also pay a large part of the taxes
needed for the payment of interest and redemption of debt, it may
be said that a good deal of their losses as creditors are offset by gains
that they make as taxpayers.
If this is true in the aggregate, sizeable redistribution may nevertheless take place among (and within) various groups of taxpayers and
government bond holders. The impact of inflation has been roughly
estimated in the United States for different income groups, comparing
average income with " net monetary asset position " (i.e. total fixed1
Savers and pensioners may also be regarded as creditors in the sense that they
refrained from spending money in one period, postponing till a later date their claims
on goods and services represented by money.
2
G. L. BACH and Albert ANDO in Review ofEconomics and Statistics, Vol. XXXIX,
No. 1, Feb. 1957, p. 5.

PRICES, WAGES, AND INCOMES POLICIES

price assets minus debts). It appeared that, as a percentage of income,
losses from inflation had been a heavier burden in the lower income
brackets than in the higher.1
A further comparison may be made between losses on monetary assets
due to inflation and the burdens of measures that could be taken to
avoid inflation. For example, what would happen if instead of permitting
inflation to reduce the real value of household assets by a specified total
amount the government increased taxation by the same total amount
(thus eliminating or substantially reducing inflation and its effects) and
distributed the burden among income groups in the same proportions as
existing taxes ? Again, it appears that in the United States inflation is a
heavier burden on lower income groups than certain types of taxation
would be.2 Similar conclusions were reached regarding other antiinflationary policies, such as raising the rate of interest and reducing
government expenditure—inflation being more regressive than these
alternative policies.3
To summarise: although the redistributional effects of price increases
may not be very serious for broad groups of the population, many
individuals may be hurt badly. On balance, these effects, especially since
they tend to injure or benefit individuals in an arbitrary and haphazard
manner, seem to be undesirable.
PRICE INCREASES, SAVING AND THE RISK OF
GALLOPING INFLATION

Since prices have been rising almost without interruption for about
a generation, and also because the question has become more and more
a subject of public discussion, one would expect people to find ways of
protecting themselves from, or taking advantage of, this situation. When
inflation is expected to continue, it is good to owe money and to own
things, and bad to own or lend money. So one effect of inflation might
be that price increases discouraged saving and the holding of cash
balances. This, in turn, would provoke further inflation. If people spent
all their money on consumption goods as soon as they received it, and
all investment had to be financed out of bank loans, excess demand would
increase and further price rises would be provoked.
1
Franklyn D . HOLZMAN in Inflation, Growth and Employment (Commission for
Money and Credit, 1963).
2
B. P. PESEK in American Economic Review (Menasha, Wis.), Vol. L, No. 1, Mar.
1960, p. 153.
s
Oswald BROWNLEE and Alfred CONRAD in ibid., Vol. LI, No. 2, May 1961, pp.
74 ff.

EFFECTS OF INFLATION

In fact this does not seem to have happened. Tables IX and X set
forth available information about personal savings as a percentage of
personal disposable incomes, and total savings as a percentage of gross
domestic product at market prices, in a number of countries.
TABLE IX. PERSONAL SAVINGS AS PERCENTAGE OF DISPOSABLE
INCOME 1954-62
Country

Austria
Belgium
Canada
Denmark
Germany (Fed. Rep.)

1954

1955

1956

1957

1958

1959

5.9
10.0
5.4
7.9
5.1

9.5
9.7
5.4
7.6
6.6

8.3 13.2
6.8 8.4

7.6
9.8
7.3
7.3
6.4
13.0
6.5
13.2
10.2
9.4

9.8
10.1
6.3
10.5
7.5
14.0
8.7
12,8
10.7
10.0

8.8
12.7
7.7
9.3
7.2
14.4
3.8
13.8
13.5
8.3

6.9
9.3
6.3
11.3
6.4
14.3
8.0
12.3
11.9
7.8

3.5
6.5

3.6
8.0

3.6
7.8

2.7
7.9

. .
7.2

5.8

.. .
Netherlands
Sweden

1.9
7.4

I960

1961

9.3
12.1
6.8
11.4
8.4
14.3
6.6
13.8
14.4
10.2
9.5
3.7 6.0
7.1 6.3

8.4
10.7
6.4
11.8
7.3
13.4
8.1
13.4
12.5
10.6
9.9
7.5
7.7

1962

11.1
8.9
8.9
8.0
12.7
8.6
11.9
11.0
8.7
6.0
7.7

Source: Yearbook of National Accounts Statistics, op. cit.
. . . = Not available.
Personal savings are denned as savings of households and private non-profit institutions. Disposable
income is denned as total income of households and private non-profit institutions net of direct taxes,
of other current transfers to general government and of current transfers to the rest of the world.

The figures in table IX range from under 5 per cent, in the United
Kingdom before 1960 (and in Ireland in 1958) to over 12 per cent, in
the Federal Republic of Germany and Luxembourg and in some years
in Belgium and the Netherlands. But if the percentage of personal
savings in, respectively, the first three years and the last three years of the
period are averaged the followingfiguresemerge:
Country

Austria
Belgium
Canada
Denmark
France
Germany (F.R.)

Average
percentage
1954-56 1960-62

7.7
9.9
6.0
7.6
6.0
. . . 13.8

8.2
11.3
7.4
10.7
7.9
13.5

Country

Ireland
Luxembourg
Netherlands
Sweden
United Kingdom . . .
United States . . . .

Average
percentage
1954-56 1960-62

6.5
13.3
10.6
8.2
3.0
7.3

7.8
13.2
12.9
10.6
6.5
7.2

The general tendency towards increases in personal savings in this
period of rising prices is clear—the only exceptions being in the Federal

TABLE X. TOTAL SAVINGS OF SELECTED COUNTRIES 1954-62
Country

4»

1954

1955

1956

1957

1958

1959

1960

1961

1962

6.5
1.5
3.7
11.7
13.4

6.9
1.8
6.9
15.6
15.5

7.3
4.6
6.1

7.6
5.1
8.6

6.8
5.4
8.1

6.9
6.8
6.6

9.8
8.7
9.9

13.9
6.8
9.7

14.0
6.7
8.9

18.0

21.3

20.3

28.2

30.4

29.6

15.2

16.3

14.9

14.2

17.5

17.2

15.9

-9.4
1.4
36.9
28.9
6.3

-2.2
3.4
37.2
38.4
8.0

4.8
0.8
12.5
37.1

11.1
0.8
9.4
40.1

-1.1
0.9
6.3
52.2

-2.7
0.9
9.7
39.0

-2.3
0.8
10.6
54.2

3.8
0.8
13.1
49.6

5.3
0.9
13.0
54.0

55.2
IIA

61.4

58.3

46.9

63.3

67.3

73.2

12.0

11.3

8.8

11.2

11.3

11.6

430
97
571
919
2 017
8.0

702
182
962
999
2 845
10.4

1 112
153
1 131
1489

164
83
854
1353

749
49
876
1786

548
45
986
1 512

387
21
873
1723

713
53
778
1680

631
68
835
2 551

3 885

3 411

2 875

3 292

3165

2 898

4 085

12.5

10.5

8.6

9.3

7.6

10.0

Austria (1,000 million schillings)
Savings : general government
public and private corporations
households, etc. x
Total . .
Total savings as percentage ofG.D.P. 2 .
Belgium (1,000 million francs)
Savings: general government
public corporations
private
„
households, etc. 1

. . . .

Total .
Total savings as percentage ofG.D.P. 2
Canada (million dollars)
Savings : general government
public corporations
private
„
households, etc. 1

. . . .

Total3.
Total savings as percentage ofG.D.P. 2

8.6

Denmark (million crowns)
Savings: general government
public corporations
private
„
households, etc.1
Total . . .
Total savings as percentage ofG.D.P,* . .

1076

1235

377

390

1651
3104
11.2

1629
3 254
11.3

6.2

5.8

1402

1556

2 066

2 385

1536

2 653

2 582

2 361

3 196

3 457

4 029

3 312

3 984
VIA

3 917
11.4

5 262
13.8

5 842
14.2

5 565
12.2

5 965
11.7

4.4

9.1

10.0

10.6

11.4

11.0

France (1,000 million francs)
Savings: general government
public corporations
private
„
households, etc.1
Total . . .
Total savings as percentage of G.D.P.2 . .

2.9

3.7

4.7

5.6

7.2

9.8

9.3

9.7

5.8
14.9
9.3

8.1
17.6
10.3

11.4
20.5
9.6

12.3
26.9
11.0

11.5
28.6
10.7

17.3
37.7
12.7

16.3
37.1
11.6

20.1
40.9
11.6

12.0

14.0

14.9

12.9

16.1

22.9

26.2

27.6

2.5

3.9

4.1

5.3

6.4

5.5

4.6

12.1
26.6
16.9

15.8
33.7
18.6

20.9
39.9
18.4

23.2
40.2
17.4

24.4
45.8
18.2

28.4
57.7
19.4

29.0
60.6
18.5

29.5
61.8
17.4

Germany (Fed. Rep.) (1,000 million marks)
Savings: general government
public corporations
private
„
households, etc.1
Total . . .
Total savings as percentage ofCD.P.2
. .
Italy (1,000 million lire)
Savings: general government
public corporations
private
„
households, etc.1
Total . . .
Total savings as percentage ofG.D.P.2 . .
(For footnotes see p. 45)

183

1162
9.3

462

460

1 330*

1542

1513
11.1

2 002
12.5

1863
2 325
13.6

336
4

2 371
2 707
14.8

645
4

953

873

2 429*

2 767

3 074
15.5

3 640
16.6

3 049 4
4 002
16.3

(Table continued overleaf)

Country

1954

1955

1956

1957

1958

I960

1961

301
50
1764
2115

595
357
1737
2 689

1 367

686

105

-426

12.2

14.1

1 189
271
1 778
3 238
15.5

439

1 733

2 159

350

1858
3 330
14.8

100

2 096
2 356
10.6

1917
2 439
10.6

2 272
4 355
17.5

2 265
4 524
17.6

1418
83
1452
1467
4 420
16.6

1030
60
1740
2 692
5 522
18.5

1381
80
1420
2 230
5111
15.8

1797
90
1660
2 489
6 036
17.2

1035
120
1 710
3 294
6159
17.3

1944
200
2 100
3 012
7256
19.2

2 270
280
2 220
4 053
8 823
20.8

2 770
240
1960
3 660
8 630
19.5

1959

1962

Luxembourg (million francs)
Savings : general government
public corporations
private
„
households, etc. 1

. . . .

Total .
Total savings as percentage ofG.D.P. 2

Netherlands (million guilders)
Savings: general government
public corporations
private
„
households, etc. 1

. . . .

Total .
Total savings as percentage ofG.D.P.2

2 420
180
2 020
3 720
8 340
17.7

Norway (million crowns)
Savings : general government
public corporations
private
„
households, etc. 1

. . . .

Total .
Total savings as percentage ofG.D.P.2

1315

1562

1761

2 231

2 286

2 130

2129

2 535

2 698

2136

2 415

3 379

3 012

1792

1989

2 375

2 516

2 149

3 451
15.3

3 977
16.5

5140
18.8

5 243
18.0

4 078
14.1

4119
13.4

4 504
13.8

5 051
14.2

4 847
12.7

Sweden (million crowns)
Savings: general government
public corporations
private
,,
households, etc.1

. . . .

1967

7 121

6 317

6 055

4 740

4 299

4 810

2 804

2 812

2 709

2 783

4 460

5 224

5 330

6 421

1775

3 159

3 555

3 113

3 032

7 878

8 820

10 423

11591

11152

12 236

15152

16 805

17325

18.8

19.5

21.2

22.0

20.2

21.0

23.8

24.2

23.1

-26
-90
1063
228

182
-154
995
466

112
-126
1 214
518

293
-189
1 185
547

346
-212
1233
419

286
-203
1311
605

37
-111
1709
1062

107
-134
1233
1425

490
-153
1 106
1 166

Tl
ta

1175

1489

1718

1836

1786

1999

2 697

2 621

2 609

6.6

7.8

8.3

8.4

7.9

8.4

10.7

9.7

9.0

4 136

Total 5 .
Total savings as percentage ofG.D.P.2

5 940

2 222
4 520
1925

4 536

5 464

United Kingdom (million £ sterling)
Savings: general government
public corporations
private
,,
households, etc.1

. . . .

Total .
Total savings as percentage ofG.D.P.2

9
in

United States (million dollars)

Savings: general government
public corporations
private
,,

. . . .

households, etc.1

Total .
Total savings as percentage ofG.D.P.*

1 146

10 592

11 845

7 954

-1406

8 139

12 987

4 782

6 575

6 684
18 860

10 084
17 508

8 631
23 025

8 159
23 627

6151
24 726

10 322
23 607

7 678
21657

6 488
27 605

8 265
29 080

26 690
7.3

38184

43 501

39 740

29 471

42 068

42 322

10.4

9.0

6.6

8.7

8.4

38 875
7.5

43 920

9.6

Source: Yearbook of National Accounts Statistics, op. cit..
. . . =Not available.
1
Incl. private non-profit institutions. a At market priées. 3 Before adjustment for stock valuation.

7.9

* Obtained as a residual.5 Incl. provision for depreciation of fixed capital.

PRICES, WAGES, AND INCOMES POLICIES

Republic of Germany and Luxembourg, where they came down nearer
to, but remained much above, the general average, and in the United
States where there was practically no change. Although personal savings
sometimes showed substantial fluctuations from year to year, these
bore no close relationship to yearly price changes. Of course, the
statistics usually include in personal savings profits invested in unincorporated businesses and, particularly, the more or less involuntary
savings represented by contributions to industrial pension schemes, etc.
Turning to the figures in table X, of total savings as a percentage of
gross domestic product, the general picture is similar, with a range of
from under 8 per cent, in certain countries in certain years (mostly rather
early in the period) to over 19 per cent, (or well over 20 per cent, in
Sweden, where however thefiguresinclude provision for depreciation of
fixed capital). Again, savings percentages were generally higher on the
average of the last three years than of the first three years of the period;
only Canada, Norway and the United States do not show an upward
trend :
Country

Average
percentage
1954-56 1960-62

Austria
14.7
Belgium
8.6
Canada
10.3
Denmark
10.8
France
9.4
Germany (F.R.). . . . 18.1
Italy
10.6

16.3
11.4
8.8
12.7
12.0
18.4
16.1

Country

Luxembourg
Netherlands
Norway
Sweden
United Kingdom . . .
United States

Average
percentage
1954-56 1960-62

13.6
17.0
16.9
19.8
7.6
9.1

15.2
19.7
13.6
23.7
9.8
7.9

One may conclude that a continuing fall in the value of money at
rates such as those experienced in these countries seems not so much to
make people save less as to lead them to invest their savings in different
types of assets. A major traditional purpose of personal saving is to provide for future needs (education for the children, a house) and for a
possible fall in income (due to old age, ill health, etc.). People are not
likely to abandon this purpose of saving but rather to look for ways of
protecting the real value of their savings.
In periods of hyper-inflation people use their available money to
invest in anything tangible they can find. Money is the one form in which
savings are not held. The moderate price increases of the 1950s have given
rise to less dramatic, but not insignificant, shifts in savings habits. Government bonds at conventional interest rates have lost a good deal of their
attractiveness and shares have become a much more desirable outlet for
savings (including those accumulated in pension funds, etc.) than they
used to be. On the one hand, it has been difficult for some governments
to raise money in the bond market: they have had either to pay higher

EFFECTS OF INFLATION

rates of interest or to issue short-term debt.1 On the other hand, in many
countries share prices have risen very considerably, probably reflecting
in part an increase in demand for this type of investment, but also perhaps reluctance of firms to issue many shares, and a preference for loans
because interest charges are deductible for tax purposes while dividends
are not.2 Popular interest in house ownership may also well have increased
in recent years partly because it is considered a suitable form of investment in times of rising prices.
Since debtor-creditor relations are so obviously distorted by continuing price increases, it might become necessary to raise rates of interest
so that they would cover the decline in the value of money. For example,
if the " normal " rate were 3 per cent, and if it were generally accepted
that prices would continue to rise by 2 per cent, a year, the actual rate
of interest could be set at 5 per cent. In fact, as people become more and
more convinced that prices will continue to rise, the rate of interest ought
to rise and has shown a rather widespread tendency to do so.
Another approach would be to apply to the principal or to the interest
on loans, or both, an arrangement similar to that often made in wage
contracts and in some social security schemes, linking them to some price
index. For example, in post-war Finland government loans have been
linked to the consumer price index, and in France loans have been issued
by the Government and public enterprises linked, for example, to the freemarket price of gold, the price of a kilowatt-hour of electricity or the
price per mile of second-class railway travel. In this case, loans can
be made in the belief that some price increases will occur, but no specific
figure needs to befixedin advance.3
Index-linked loans may be less dangerous than a rate of interest
including a fixed allowance for price increases; similarly, sliding scales
1
See for example United Kingdom: Report of the Committee on the Working of
the Monetary System (Radcliife Report) (London, 1959), paras. 68 and 572 ff. (Cmnd.
827), and United States Congress Joint Economic Committee: Staff Report on Employment, Growth, and Price Levels (Washington, 1959), pp. xxxvi and 28.
2

HOUTHAKKER, op. cit., pp. 121 and

131.

As long as people have different expectations about the future of prices one
could also envisage the coexistence of loans that are, and others that are not, linked to
indices. For example, the government might issue some bonds at 5 per cent, interest and
others at 2'/i per cent, but tied to the cost-of-living index. Anybody expecting that prices
would rise faster than 2Y2 per cent, a year could buy the latter while the government, if
it expected that it could keep price increases below this rate, would be happy to sell the
low-interest bonds. Probably the issue of two types of bonds would, however, be interpreted as a government belief that prices would not increase by much less than 2'/2
per cent, a year, and this belief might give rise to widespread speculation. On the other
hand, if people had generally believed that the rate of price rises would be much larger
than 21/2 per cent., a government loan of this type might cause them to revise their
estimates and, by making them less inflation-minded, take some of the inflationary steam
out of the economy. For further discussion of index loans see Guy ARVIDSSON in
Skandinaviska Banken Quarterly Review (Stockholm), Vol. 40, No 1, Jan. 1959, pp. 1-14.

PRICES, WAGES, AND INCOMES POLICIES

in wage contracts may involve smaller wage increases than agreements for
stipulated future wage increases based on some liberal estimate of future
rises in the cost of living; and building contracts providing that increases
in actual cost are to be added to the contractor's bills are likely to lead
to lower building prices than contracts in which the contractor has to
protect himself in advance from the risk that wages and materials prices
will go up. Yet official acceptance of the inevitability of price increases
is repugnant to many people. They fear that, through "escalator" clauses
in wage agreements, pension schemes, agricultural price programmes, etc.,
increases in some prices will automatically and immediately be followed
by more and more other price rises, and creeping inflation will degenerate
into galloping inflation. The Radcliffe Committee hoped " that the Government of the United Kingdom will not be driven to resort to an
expedient (namely the issue of 'index bonds') which would too plainly
be a confession of failure to maintain a reasonable degree of stability in
the value of money and might easily have disruptive consequences for
our economic system".1 In France a decree issued late in 1958 makes it
illegal for wages and salaries to be linked to changes in the guaranteed minimum wage (itself linked to the cost of living).
The main reason why the extension of index-linkage and other protective devices might lead to accelerated price increases is that inflation
would cease to make it possible for the government or some private
groups to increase their share in the national product. This is possible
only if some other group can be made to take a smaller share.
Ways in which this result can be achieved through inflation are familiar. When a government wants to carry out a public works or other
programme, it may obtain the necessary funds from taxes or borrow
from the public in ways that do not increase the supply of money.
Alternatively, however, it may borrow from the central bank or in other
ways increase the supply of money. Excess demand then occurs, prices
rise and citizens withfixedmoney incomes are unable to buy some goods
which they would not have been willing to forgo if merely invited to
subscribe to a loan. Similarly, when a rise in the over-all wage level is not
matched by higher productivity, profits may fall or prices rise and the
employers or consumers (other than wage earners) or both can buy less,
while the wage earners can buy more. But when these initiatives are followed by increases of incomes in other groups—profits, wages, pensions,
rents, interest rates or dividends—they lose their effect. Money created
for the government or applied to meet wage increases will no longer buy
1

573.

Report of the Committee on the Working of the Monetary System, op. cit., para.

EFFECTS OF INFLATION

what it was intended to buy. More new money may then be created or
further increases made in wages, but the attempt to increase government
programmes or to raise real wages will be frustrated again, and either
inflation will become galloping, or the objectives will have to be
approached in different ways or be abandoned.
PRICE INCREASES AS A SOCIAL LUBRICANT

It can be argued that price increases help in mollifying social conflict.
Higher prices enable farmers or manufacturers to enlarge their money
incomes without directly reducing anybody else's money income. It is
easier to reach agreement on a wage increase when prices can be raised
and the cost passed on to more or less anonymous (and usually unorganised) buyers than when profits have to be reduced. It has been suggested
that " if managements are to secure settlements below an average figure
of 4 or 5 per cent, a year, there will probably be more strikes over money
at contract-renewal time ". 1 If wages and other incomes had to be determined without any risk of causing inflation, they would have to be fixed
in precise real terms such that any gain by one group would be seen
to be clearly at the expense of other claims or existing shares of the national income. In the Massé report it is pointed out that very serious social
conflicts might occur on the eve of major decisions regarding the global
distribution of incomes that such an approach would involve. This risk
would have to be weighed against the disadvantages of inflation, including
the frustration and irritation occurring when rising money incomes are
found to be illusory because of the price increases they provoke.2 It must
be added, however, that the struggle for higher group incomes, while
causing price increases, is not in the short term a fight for illusions as
long as some groups have greater bargaining power than others and price
increases do not immediately follow income increases. Moreover, any
group that is left behind in the struggle suffers a clear disadvantage.
Some people, objecting to interest (and especially compound interest)
as an " unearned income ", may also feel that the gradual expropriation
of savings through price increases is socially desirable. A good deal of
interest income goes, however, to pensioners and other people of small
means. And, as mentioned above, savers can, and perhaps increasingly
will, find means of protecting themselves from price increases by investing
their savings in a different way.
1

J. T. DUNLOP in The Reporter (New York), Vol. 20, No. 11, 28 May 1959, pp.

11-12. It should be added, however, that this prediction has not come true in the United
States in recent years.
2
Pierre MASSÉ: Rapport sur la politique des revenus, établi à la suite de la Conférence des revenus (La Documentation française, Recueils et monographies N° 47,
Paris, Feb. 1964), pp. 8-10.

PRICES, WAGES, AND INCOMES POLICIES

INFLATION AND ECONOMIC

GROWTH

Rising prices are often thought to stimulate economic growth, but
they are evidently not a necessary condition for rapid increases in output.
Consider, for example, the figures shown in table XI.
TABLE XI. PERCENTAGE GROWTH OF OUTPUT AND PRICE CHANGES
IN SELECTED PERIODS, COMPARED WITH THE PRECEDING DECADE
Period

Growth
of output

Price change

Italy

1884-1893
1904-1913
1924-1933
1934-1943

19.4
30.7
40.6
8.8

- 8.7
+ 11.4
+ 35.0
+ 18.5

Japan ,

1892-1902
1903-1912
1923-1932

64.8
34.7
67.4

+ 35.4
+ 38.2
- 7.1

United Kingdom

1885-1894
1905-1914
1925-1934
1939-1953

37.6
16.5
21.1
22.7

- 15.0
+ 9.6
- 16.6
+ 51.3

United States

1884-1893
1924-1933
1944-1953

54.8
29.1
52.0

- 16.0
0
+ 84.8

Country

Source: Taken from series given by Otto ECKSTEIN in The Relationship of Prices to Economic
Stability and Growth, op. cit., p. 362.—

The figures do not point to any clear conclusion as to whether rising
or falling prices tend to accompany economic growth. This is confirmed
by the survey, already referred to, of 25 countries over the period
1947-57. The authors concluded—
Differences in the rate of inflation have not been systematically associated
with differences in the rate of growth of real income per head, whether we compare different countries or different periods; but within the present material it
is instances of a lower rate of inflation being associated with a higher rate of
growth of real income that predominate.1
Theoretical discussion of this question has included the reasoning,
referred to above, that continuing price increases may cause a fall in
savings or their diversion into the speculative hoarding of goods. This
would reduce productive investment and slow down the rise in output.
However, it does not seem to have happened in fact, except in hyperinflations, and there is no evidence as yet that continuing small price
1

PHELPS BROWN and BROWNE, op. cit., p.

725.

EFFECTS OF INFLATION

increases are bound to degenerate into an inflationary trot or gallop,
though growing " indexisation " may increase this danger.
On the other hand, it is clear that growth of output and incomes requires a level of demand high enough to ensure reasonably full employment of labour and other resources. It is certain that the full potentialities
of growth do not materialise when demand is deficient. It is not altogether
impossible that their full exploitation requires some continuing slight
excess in demand. Again, it is certain that the maintenance of full employment is easier when demand is slightly excessive than when it is
slightly deficient or, in some sense, "just adequate ".
When, in response to an increase in demand, prices rise more than
costs (such as wages) so that profits also rise, businessmen are likely to
increase their output and to invest in new capacity. In this case price
increases may be accompanied by fairly rapid growth.
The figures in table XI show, at least for the United Kingdom and
the United States, rapid growth accompanied by falling prices towards the
end of the nineteenth century. But there are reasons for doubting that
this experience is likely to be repeated. In the early phase of the industrial
era tremendous new resources of materials became available and revolutionary changes in techniques of production helped to raise output
rapidly at steeply falling costs. New resources and techniques are, of
course, still being discovered but, as will be noted farther on1, modern
methods of price and wage fixing make it much less likely that cost
reductions will lead to substantial price reductions.
It seems likely that nowadays price increases are to some extent caused
not by rising demand but by price- and wage-fixing decisions.1 Price
rises following an increase in cost are less likely to stimulate growth
than those following an increase in demand. At the same time, price and
wage-fixing arrangements causing an upward pressure on prices are
probably more widespread and effective when demand is high than when
it is low.
So special problems arise when—in the face of wage- and price-fixing
practices that fail to translate cost reductions into lower prices and that
may cause cost increases—it is desired to maintain full employment and
fairly rapid economic growth. On the one hand, the maintenance of full
employment may lead to higher prices and wages. On the other hand, it
will be necessary to keep the level of demand high enough to enable full
capacity output to be sold although its price is rising. In such conditions
economic expansion might well make some rise in costs and prices
inevitable.
1

See below, Ch. III.

PRICES, WAGES, AND INCOMES POLICIES

RISING PRICES AND THE BALANCE OF PAYMENTS

The various effects of continuing price increases considered so far
include some drawbacks regarding the distribution of income (which
may be inconvenient but need not, on the whole, impose intolerable
burdens) and some possible advantages in limiting social conflicts and
promoting growth. But the most important and most clearly adverse
result of modern inflation is that, if prices rise faster in one country
than in others, this endangers the balance of payments. There are
three reasons why this is the most important difficulty arising out of
inflation.
First, whereas the other problems are largely questions of policy
choice and political compromise, at some point restoration of the balance
of payments becomes a matter of economic necessity, namely when
exchange reserves and international credit facilities are exhausted.
Second, since international borrowing may weaken a country's external
political position, governments will not normally want even to use all the
credit facilities that are open to them. Third, when an acute balance-ofpayments problem arises, present arrangements for international trade and
payments being what they are, the obvious remedy consists of measures
that restrict demand, output and employment. As a result imports will
fall and producers may make greater efforts to sell abroad. However,
in order to correct in this manner a balance-of-payments deficit of a certain size, total production and incomes have to be reduced by an amount
several times larger. Moreover, measures taken to restrict demand are
apt to be inequitable as well as harmful to growth. A credit squeeze, for
example, is apt to be felt especially by young firms dependent upon loans
to finance the exploitation of new inventions or ventures into new
methods. It would be better if continuing price increases could be stopped
by different measures—such as an incomes policy.
From this point of view it does not seem surprising that the question
of incomes policy came to be considered seriously in the United Kingdom after a succession of deliberate restrictions of output (" stop-go
measures ") had proved a frustrating method of coping with a series of
balance-of-payments crises.1 Similarly, in the United States, " guideposts " for wage and pricefixingwere formulated by the Council of Economic Advisers in 1962, after a period of sustained gold losses that had
1
From 1950 to the end of 1964 the bank rate was raised on 12 occasions and on
ten of these by one full percentage point or more. During the same period the official
discount rate was raised seven times in the Federal Republic of Germany, only once
in Norway, and ten times in the Netherlands, but only once by as much as one full percentage point.

EFFECTS OF INFLATION

caused considerable concern.1 Conversely, it could hardly have been
accidental that the major revisions of wage-policy procedures in the
Netherlands since 1959 occurred in a period when the balance of payments was very strong.2 In the Federal Republic of Germany, which has
not had a balance-of-payments problem, the Government has shown a
limited interest in an incomes policy. In France, interest in an incomes
policy arose when, under the Fifth Republic, the balance of payments
showed signs of weakening, earlier post-war inflations in this country
having given rise to devaluation.
If the problem of inflation is primarily that of maintaining balanceof-payments equilibrium rather than of preserving a constant price level,
two major implications follow. First, since mild inflation has now become
internationally widespread, the avoidance of balance-of-payments problems does not require absolute price stability. It will be sufficient if a
country's price level does not, over long periods, rise faster than that in
major competing countries. Second, the balance-of payments problem is
conditioned in large part by the existing arrangements for international
trade and payments, so that its solution may also lie in changing these
arrangements. The three main aspects involved are: the fact that, in the
course of the 1950s, direct import restrictions were abolished; the system
of fixed exchange rates; and the slow growth of international credit
facilities.
Larger international credit facilities would give countries more leeway
so that measures to solve acute balance-of-payments problems could be
less frequent and abrupt. Réintroduction of direct import controls would
enable countries to protect their currency reserves while taking gradual
measures for restoring balance. But the most interesting possibility of
solving international problems arising from continuing price increases
seems to be that of adjusting rates of foreign exchange more frequently
than has been done in the past. If a country's prices kept rising faster than
those in other countries, this could be corrected by a corresponding
reduction in the price of its currency. Conversely, if a country's price
level kept rising less than that in other countries, its currency could
be revalued. In accordance with these principles the French franc was
devalued in 1957 and again in 1958, while the monetary units of the
Federal Republic of Germany and the Netherlands were upvalued in 1961.
1
Though, as noted above, prices in the United States have been remarkably steady
since 1958. Indeed, public preoccupation with inflation problems in the United States
had arisen during the period 1955 to 1957 when prices were found torisepersistently in
spite of the existence of substantial unemployment. The balance-of-payments difficulties
referred to above, however, do appear to have strengthened this preoccupation significantly.
2
See below, p. 105.

PRICES, WAGES, AND INCOMES POLICIES

However, frequent adjustment of rates of exchange is viewed with
disfavour by banking authorities as well as by political leaders of otherwise widely diverging opinions.1 The reasons include considerations of
monetary theory and of national prestige. It has been suggested that,
while a system of moreflexibleexchange rates may be useful in the case of
small and medium-sized countries, it would be unacceptable if extended
to the currencies of large countries.2
CONCLUSIONS

Inflation clearly has some undesirable effects. For purposes of policy
making it is necessary to appraise these effects, to see whether and how
they might be mitigated, and to compare the disadvantages of the
remaining distortions of inflation with the problems that might arise if
attempts were made to avoid the inflation.
While price increases have been almost uninterrupted during the last
15 years or so, they have been moderate. Through imperfections of consumer price indices, the cost of living has risen less than these indices
would suggest. But even a 3 per cent, annual increase leads to a doubling of prices within 24 years ; at 5 per cent, this period becomes less than
15 years, so that the value of money would fall to less than a quarter
within a lifetime. Such increases in the price level produce distortions
in the distribution of incomes and wealth which tend to be undesirable
because they are arbitrary and haphazard, depending on how different
income groups invest their savings and how promptly they can
achieve adjustment of their incomes. Tins is likely to be different for
different groups such as manual workers, teachers or pensioners.
Such arbitrary effects, however, can be and often have been corrected
(for example by indexisation) at relatively low cost. If prices rise at a rate
of 5 per cent, a year, complete compensation of those who suffer losses
would perhaps involve redistribution of about 1 per cent, of the national
income a year. In industrialised countries price rises of this order have
rarely been sustained for long periods. Furthermore, while such corrections have in fact been frequently made, they have not yet caused moderate price increases to degenerate into disruptive " galloping " inflation.
However, this may well have been due in part to the fact that governments have from time to time taken measures to halt price increases
through restrictions on demand and on the rate of economic expansion.
Continuing price increases have the important advantage of providing
a moreflexiblecontext for social adjustments. At the political level this
1

Economic Policy in Our Time, op. cit., Vol. 2, p . 103.
For instance, A. SMITHIES and J. K. GALBRAITH in Review of Economics and
Statistics, May 1964, pp. 113 and 117.
2

EFFECTS OF INFLATION

is also a great help to governments and parliaments—choices among
politically attractive projects need not be as uncompromising as they
would have to be if government budgets had to conform to strict criteria
of economic stability. Thus there are important compensations for the
internal disadvantages of the rising prices.
There has perhaps been more concern about the effects of inflation on
balances of payments. The outstanding feature in this case is the economic
necessity of equilibrium; while there is some scope for decision regarding
the extent to which a country is prepared to let its foreign exchange
reserves dwindle or to seek loans abroad, balance-of-payments problems
eventually leave no room for choice. Problems of external balance arise
primarily from excess demand, but they may be much aggravated by
price increases. Difficulties occur when demand or prices in one country
increase faster than in other countries. Even then they might be overcome
by, for example, abandoning the existing system offixedrates of exchange,
or by imposing temporary import restrictions. Such devices, however,
have disadvantages of their own and are not regarded with favour.

Improved international short-term lending provides more leeway but
does not in itself remedy a balance-of-payments deficit. For the rest, the
obvious remedy for such a deficit is restriction on internal demand, output and employment—i.e. slowing down of growth, which may possibly
provoke a recession. From this point of view the main negative effect of
inflation is the threat it holds for economic growth and full employment
through the creation of balance-of-payments problems under present
arrangements for international trade and payments. Moreover, while the
distributional effects of a 5 per cent, annual increase could be corrected
by transfers equivalent to a fraction of 5 per cent, of the national income,
when output during a recession is 5 per cent, below capacity, the full
5 per cent, is lost, totally and irrevocably.1

A. P. LERNER in Review of Economics and Statistics, Vol. XLV, No. 1, Part 2,
Feb. 1963, p. 146.

CHAPTER III
CAUSES OF INFLATION AND THE FIXING OF
PRICES AND WAGES
DEMAND PULL AND COST PUSH

Explanations of inflation, and policy recommendations based on
them, differ mainly in the relative importance given to two groups of
factors—those causing price increases through " the pull of demand "
and those operating through " the push of costs ".
Pure demand inflation would occur if national output had increased
as a result of growing private and public demand for consumption and
investment, without any rise in costs and prices, until a certain clearly discernible full-employment level of production was reached beyond which
further increases in demand did not lead to any further growth of output,
and therefore only caused prices and wages to go up. If this excess demand
were maintained, a prolonged rise in prices and wages would occur even
in the absence of any trade union pressures for higher wages or of
monopolistic price fixing. The only effective policy to cope with this kind
of inflation would be to curb demand by reducing government spending,
increasing taxes, raising the rate of interest, etc. .
Pure supply or cost inflation would occur if, for example, due to
action by monopolistic or oligopolistic enterprises, by trade unions, or
under a government agricultural price support programme, wages and
prices were rising in depressions and in boom periods alike, without any
relation to the level of employment and output. Thus, if all prices were
fixed by adding a certain profit margin to wage and other costs of production, while all wages were fixed by adding a certain increase in real wage
rates to the cost of living, there could be a perpetual rise in prices and
wages even in conditions of serious unemployment and underutilisation
of plant and equipment. Measures to reduce demand would be powerless
against such inflation. The only effective policy would be to alter methods
of price and wage determination.
But the explanation of post-war inflation in industrialised countries
has proved much too complex to fit either of the above extreme models.
A more promising approach, therefore, is to assume that normally both

CAUSES OF INFLATION AND PRICE AND WAGE FIXING

demand and supply (or cost) factors have played a role. On the one hand,
beyond some level of employment and capacity utilisation which may be
considerably short of full employment, increasing demand, instead of
resulting either in larger output or in higher prices, is likely to be associated with both. Output may continue to increase, but at gradually
increasing costs caused by the appearance of bottlenecks (which, as far as
labour is concerned, would give rise to overtime work, the drawing on
marginal labour reserves—housewives and pensioners—and competitive
wage increases for particularly scarce types of labour). On the other hand,
the extent to which enterprises and trade unions may be able or willing
to raise prices and wages, far from being wholly independent of the level
of employment and plant utilisation, is likely to be smaller at low levels
of employment and output than at higher ones.1
Reference may also be made to the view of inflation as a joint result
of demand and cost factors that is known as the " sectoral " or " demand
shift " theory.2 This is based on the fact that, even when total demand
does not rise faster than the total capacity to produce, normally the
demand for labour and capital goods is increasing in some sectors of the
economy while it is declining in others. In the expanding sectors, wages
and other costs may then rise, while in the contracting sectors, for
reasons discussed below, they do not fall. When such shifts in demand
for labour and capital are large and sudden, there may be significant
increases in the over-all wage and price level.
If modern inflation is indeed best understood as the combined result
of demand and supply factors, the conventional monetary and fiscal
means of dealing with inflation by manipulating demand will not be
ineffective, as they would be in the case of pure cost inflation. But their
effectiveness will be both more uncertain and more costly than it would be
in the latter case.
It will be more uncertain because of the absence of any exact association between any given level of demand and a particular rate of price
increase. The latter will depend largely on the many factors determining,
in any specific situation, at what level of employment and output costs
will begin to rise and at what rate, as well as on the factors determining
the degree to which labour organisations and industrial enterprises will
push up wages and prices. In other words, even if the monetary and
fiscal authorities were able to manipulate the level of demand with preci1
Cf. BRONFENBRENNER and HOLZMAN in American Economic Review, Sep. 1963,
pp. 595-597.
2
See especially Charles L. SCHULTZE: Recent Inflation in the United States, Study
Paper No. 1 for the Study of Employment, Growth and Price Levels by the United
States Congress Joint Economic Committee (Washington, 1959).

PRICES, WAGES, AND INCOMES POLICIES

sion, they would not necessarily be able to control the price level within
an adequate margin of tolerance.
Measures for curbing demand as a method of halting increases in the
price level will, further, be costly, in the sense that they will affect not
only the price level but also employment and output. And under the
impact of modern methods of price and wagefixingthe level of demand
and output at which the price level would stop rising at all might now
be so low as to be unacceptable in terms of the policy objectives of
economic growth and, especially, full employment.
From this point of view perhaps the outstanding feature of post-war
inflation is that a situation may arise in which countries cannot attain
full employment and a full utilisation of physical production capacity
without incurring unacceptable increases in the price level. Such situations may be caused by bottlenecks that occur as demand increases, and
by the increasing power of private organisations to push up wages and
prices. The latter aspect may be clarified by an examination of modern
price- and wage-fixing procedures in industrialised market economies.
FEATURES OF MODERN PRICE AND WAGE FIXING

Price and wage adjustments are the results of decisions : who takes
these and on what basis ? A reply to these questions should help in understanding why particular prices and wages behave as they do, and if this
behaviour is considered undesirable because it is inflationary, an analysis
of the mechanism through which it is brought about may point to possible
corrective measures.
The analysis of price- and wage-fixing processes is, however, a very
large and difficult task. There are millions of prices and wage rates.
Methods of fixing them vary as between countries, as between different
types of commodity and labour markets, and as between individual
firms. Furthermore, while concrete decisions (such as an announcement
that the price of a particular kind of steel is raised by so much per ton, or
a collective agreement scheduling wage increases over a three-year period)
can often be observed, it is much more difficult to detect how they were
arrived at. Why was the steel price not increased at an earlier (or later)
date and why is the increase not larger (or smaller) than it is ? The answer
to such questions certainly depends on, and varies with, the decisionmaker's objectives ; on his knowledge, ignorance and beliefs about facts
that he regards as relevant; and on his judgment and temperament in
dealing with facts, unknowns and uncertainties. These largely intangible
factors are difficult to analyse and the discussion below can only be
sketchy. Even so it may help to clarify certain aspects of the analysis

CAUSES OF INFLATION AND PRICE AND WAGE FIXING

in later chapters. The following cases are briefly reviewed below: government price fixing, private pricefixingfor raw materials and manufactured
products, retail pricing, and certain aspects of wage fixing.
GOVERNMENT PRICE FIXING

It is not necessary to review here those somewhat special prices the
fixing of which is part of over-all economic regulation by governments
and other central authorities : rates of exchange, tax rates and the rate of
interest of the central bank. Nor do we here discuss government control of
private prices (such as rents and the prices of certain essential consumer
goods) as part of anti-inflationary policies. Even when these matters
are left aside, the government is the most important single price-setter in
most countries. It sells goods and services through such agencies as the
post office, gas and electricity undertakings and public transport corporations or, where such public utilities are in private hands, it reserves the
right of controlling rates and fares. In some countries mining and parts
of the manufacturing sector are nationalised. In most countries the
government fixes minimum prices, or price ranges, for key agricultural
products, to assure the farmers a minimum income. In the case of some
commodities (for example military equipment) the government is the
only or the principal buyer and may have a considerable influence on
prices.
Government pricing policies are complicated by considerations of
political belief and expediency, of the country's general economic and
social conditions and, more broadly, of " the national interest ". These
things tend to give to many decisions more of an ad hoc character than
in private business.
For the purpose of this discussion, and at the risk of over-simplification, it may, however, be said that government price decisions follow
rather than lead price changes in the private sector. This is deliberate,
and explicitly so in the case of agricultural price policies based on some
" parity " concept, minimum prices being adjusted in relation to nonagricultural prices or incomes. In the case of public utilities an important
aspect of their role of serving the national interest is often understood to
be that prices should not be raised unless there are compelling reasons
for doing so. The services concerned are sometimes allowed to incur
substantial over-all deficits when costs are rising, and losses on particular
operations (such as unremunerative railway lines or the handling of
ordinary mail) may be tolerated or even encouraged for long periods.
And when the enterprises break even, their prices are often not intended
t o cover povision for investment or expansion (as is the practice in

PRICES, WAGES, AND INCOMES POLICIES

private industries). In order to limit inflation governments sometimes are
particularly reluctant to increase their own prices when (and because)
private prices are rising.1
On the other hand, government prices may be even slower in following
downward movements in private prices. When public utilities and nationalised enterprises have been losing money, they may not quickly reduce
their prices when costs are falling. Downward adjustments in guaranteed
agricultural prices are unpopular (and therefore politically difficult to
bring about) even when there may be good economic reasons for them.
The fact that government prices seem to follow rather than to lead
private prices does not necessarily mean that this policy helps to achieve
over-all price stability. For example, in several countries productivity
in agriculture has been rising faster than over-all productivity, and in
such cases over-all price stability would require a fall in agricultural
prices. In Sweden guaranteed prices have been reduced from time to time
to allow for productivity increases in agriculture, but in other countries
this has not often been done.2 Again, when the prices of nationalised
enterprises are kept low in the face of rising costs, and deficits occur,
people may spend on other things the money they save on cheap public
services, thus perhaps causing other prices to rise faster than they would
have done otherwise.
PRICING IN THE PRIVATE SECTOR

Raw Materials
Prices of many raw materials (not including manufactured substitutes
such as synthetic rubber or rayon) are set in " open markets " where
buyers and sellers meet, indicating the prices at which they are willing
to make transactions. The resulting prices are quoted publicly and often
internationally; they may vary substantially even in the course of one day.
These fluctuations may be limited by various devices: e.g. in the tin
market, when prices fall to a certain level, the manager of an international
buffer stock may start buying what is offered at the price and, as long as
1
An example is the 1956 " price freeze " for British government-operated services
involving, inter alia, a government decision to make special advances to the British
Transport Commission to cover its deficits arising from this policy.
2
Mention may, however, be made of another device for taking account of productivity increases in government price fixing. In the United States the government applies
" learning " or " improvement " curves to test the reasonableness of prices for government contracts, for example in the procurement of military aircraft. These curves
express the fact that often cost of production falls as an operation is maintained for
some time as the result of growing experience and other factors. Cf. Murray L. WEIDENBAUM in The Relationship of Prices to Economic Stability and Growth, op. cit., p. 538.

CAUSES OF INFLATION AND PRICE AND WAGE FIXING

he has funds, prevent the price from falling further. When the price rises
to the " ceiling " specified in the International Tin Agreement, he may
offer what is demanded at that level and, as long as he has tin, prevem
further price increases.
The organisation and operation of commodity markets is complicated.
For the purposes of this discussion the important thing is that prices in
these markets are quite sensitive to changes in supply and demand. Even
in the case of regulated markets (such as tin, sugar and wheat) a significant increase in demand usually causes a prompt increase in prices.
Furthermore, demand depends a good deal on expectations. When people
expect prices to rise (and a price rise itself may be a ground for expecting
further increases) they will often decide to buy quickly and increase their
stocks before their expectation comes true. But in so doing they tend to
make it come true: due to increased demand the price will in fact rise.
The reverse happens when demand falls or prices are expected to become
lower. However, large purchases for stocks in anticipation of price
increases will cause a fall in demand and prices when stocks are liquidated
again. In short, raw material prices may be quite volatile, responding
quickly and violently to fluctuations in demand.
Many raw materials are produced largely in countries other than
those which use them for their manufacturing industries. Few industrial
countries are self-sufficient as regards their key raw materials. Since many
raw materials are homogeneous and can be described simply in a few
standard classifications, their markets are " open " also in the sense of
being of international scope : they are " world markets " with " world
prices ". Prices of raw materials have a substantial impact on costs and
prices of manufactured goods but they are largely beyond the control of
importing countries. The United States economy, accounting for an
important percentage of world industrial production, has a strong influence
on the demand for and prices of several raw materials, but in a global
way: neither the government nor individual firms or industries have
direct control over the demand for materials. Western Europe, though
its industrial production is smaller than that of the United States, is much
more dependent on imports, and its economy has a larger effect on world
trade. 1 But the impact of individual European countries on the total
demand for raw materials is for the most part rather small. So if one such
country experiences an inflationary boom, this will probably not cause a
significant rise in international raw material prices. Conversely, a country
that succeeds in avoiding domestic inflation may none the less experience
1
General Agreement on Tariffs and Trade: International Trade 1957-58 (Geneva,
1959), p. 97.

PRICES, WAGES, AND INCOMES POLICIES

rising costs of imported raw materials. In short, raw material prices are
not only volatile, they are often also beyond the reach of stabilisation
policies in individual countries. From the point of view of these policies,
changes in raw material prices are in the nature of acts of God. Furthermore, through the process of speculative stock-building referred to above,
the existence of these price fluctuations is liable to cause considerable
variations in the volume of imports of industrialised countries, which may
lead to balance-of-payments problems and thus impel the government to
curb demand, employment and output (partly with a view to provoking
the liquidation of stocks).
Manufactured Goods
Prices set in international open markets are even further beyond the
control of individual producers : the management of any tea or rubber
plantation may decide not to sell at any particular market price (hoping
that next week the price will be better) but it cannot influence the world
tea or rubber price.1 In this respect its position differs sharply from the
manufacturer's. The industrial producer often decides how much buyers
of his product (sometimes including the ultimate buyer purchasing from
retailers or other middlemen) shall have to pay for it.
The price policies of large manufacturing enterprises (which are of
special concern to the present discussion) naturally depend on the overall management objectives and policies of these firms. In the light of
some recent studies, especially in the United States, it appears that
expansion of the enterprise is a common basic management aim. This
aim could be achieved by increasing sales, and by matching this expansion
by a growth of the firm's productive capacity. Expansion of sales is,
indeed, a major goal of businessmen. A high volume of sales may be
sought even if it means a reduction in total profits: " a programme which
proposes explicitly any cut in sales volume, whatever the profit consideration, is likely to meet a cold reception ".2
Yet although somefirmsmay therefore act in a way similar to certain
public utilities in maintaining unprofitable lines, there seems to be a
1
Producers have of course been able to influence prices when acting not individuually but jointly. Such joint action has often been difficult to organise, and has necessitated governmental and inter-governmental intervention.
2
William J. BAUMOL in Review of Economics and Statistics, Vol. XL, No. 3, Aug.
1958, p. 213. Cases in which profits were sacrificed in order to maintain sales included
firms losing money in distant markets, where local competition forced the price down
to a level which did not cover cost of transportation, and a firm whose sales to small
retailers in sparsely populated areas were so few and far between that gross revenue did
not even cover the salesmen's wages.

CAUSES OF INFLATION AND PRICE AND WAGE FIXING

widespread tendency for industrial firms to aim at a certain level of overall profits. The interest of this finding lies in the words " a certain level ".
Manufacturers in newly developing countries are often said to seek the
highest possible profits from each transaction, and in economic theory
producers are commonly assumed to maximise profits. However, a significant sector of business in high-income countries appears to be more
interested in a particular over-all level of profits that is regarded as a
minimum but also, from certain aspects, as a ceiling which it is not necessary or desirable to exceed. Indeed, a strict policy of maximising profits
should often lead to the reinvestment of gross profits (both depreciation
and net earnings) not in the firm itself but in some other enterprise with
better prospects. But where the overriding aim is expansion of the existing firm, the objective of profit maximisation must be replaced by the
less ambitious one of achieving a " profit target " which is compatible
with that of expansion of the firm. For a number of large United States
firms profit targets were found to range from 10 to 20 per cent, on capital
after tax. 1

Profit planning seems to be undertaken usually on the basis of
standard costs, an estimate of unit cost of production assuming a normal
rate of output over a period of years. This normal volume of production
(of importance in estimating overhead costs per unit) may, for example,
be set at 75 or 80 per cent, of productive capacity. The price is then
determined by adding to the standard cost a margin of mark-up to achieve
the profit target. Alternatively, a firm may make an estimate of the price
at which it expects to be able to sell a suitable amount of a product,
deduct the desired profit margin, and then design the product in such a
way that it can be made at a cost thus calculated in advance. A good deal
of pricing is on a " cost plus " basis.2 This fact is of central importance
to the analysis of inflationary processes and has a major bearing on the
question of anti-inflationary policies. It is, however, useful first to consider
briefly the functions of profits as a planned margin on top of the cost of
production.
The first and obvious purpose of making profits is, of course, to provide the owners of the firms with what they (or their managers) decide
to be a " fair ", " necessary " or " economic " return on capital invested,
risks incurred and entrepreneurial services rendered. But profits are not
fully distributed among shareholders, directors and managers. A considerable portion is usually retained to finance advertising (promotional
campaigns) and to acquire new plant and equipment. Both these pur1
Robert K. LANZILOTTI in The Relationship of Prices to Economic Stability and
Growth, op. cit., p. 444 and the reference cited there.
2
Ibid., pp. 445 and 446.

PRICES, WAGES, AND INCOMES POLICIES

poses follow from the objective of promoting the firm's growth: the
former expands markets, the latter enables the increased demand to be
met. Since, as noted below, modern industry relies little on price competition as a means of increasing sales, it depends all the more on promotion
campaigns. Since the actual effects of advertising on sales are fairly obscure, it is also reasonable tofinancepromotion campaigns from retained
profits: it would be hazardous to obtain the necessary money from
fresh issues of bonds, bank loans or shares. But in many industrialised
countries retained profits have also become a major source of financing
the purchase of new plant and equipment.1
As long as business pursues its profit targets by means of a percentage
mark-up on costs, prices will move in the same direction and proportion
as costs. Since costs depend on other prices (of raw materials and equipment as well as labour), this practice serves as a vehicle for conveying
price changes from one stage of production to the next. A price change
for hides and skins will affect prices in leather, and then in footwear
and so on. An important factor is the speed with which piice increases
are passed on from one stage to the other. In practice, a few months may
elapse between an increase in costs and the price rise.2
Some Qualifications to Cost-Pius Pricing.
The principle offixingprices on a cost-plus basis may have an appeal
of fairness not dissimilar to that of wage adjustments to changes in a
cost-of-living index. It may, therefore, be useful to note here three ways
in which the practice of this price policy may deviate from any simple
theory on which it may be based.
In the first place, although most larger firms have detailed current
data on costs, produced and handled by competent accountants, price
fixing seems often to be based on predetermined " standard costs "
rather than real costs of production. When costs are calculated for some
1
See United Nations, Economic Commission for Europe : Economic Survey of Europe
in 1955 (Geneva, 1956), pp. 88 ff. and Organisation for European Economic Co-operation: The Supply of Capital Funds for Industrial Development in Europe (Paris, 1957), pp.
145 ff.
2
It has been suggested that the maintenance of percentage mark-ups in the face of
cost increases may be more typical of western European countries than of the United
States (Jean MARCHAL in John T. DUNLOP (ed.) : The Theory of Wage Determination
(London, Macmillan, 1957), p. 159 and the sources quoted there). However, the Governor of the Commonwealth Bank of Australia noted a tendency of " cost-plus pricing "
in his country's industry which he ascribed in part to " the influence of American
business thought " (H. C. COOMBS in Economic Record (Melbourne, University Press),
Vol. XXXV, No. 72, Dec. 1959, p. 345). For a study of mark-up pricing in Danish
manufacturing see Bjarke FOG: Industrial Pricing Policies (Amsterdam, North Holland
Publishing Company, 1960).

CAUSES OF INFLATION AND PRICE AND WAGE FIXING

predetermined " normal " rate of output rather than for actual production, overheads per unit may be over- or underestimated depending on
the real utilisation of capacity. It has been noted that standard costs
are often not corrected for rising productivity. For example, standard
costs may be raised when wage rates are increased but not reduced when
output per man-hour rises. Furthermore, the methods of accountants
themselves (as distinct from the pricing officials using their information)
have been criticised: an accountant's conception of costs is different
from that of an economist and often not relevant to policy decisions.
"There is, in any large corporation, a full-time job for an economist
in undoing the work of the accountant." 1
Again, some firms may calculate standard costs taking expected
rather than actually prevailing prices of materials and labour as the basis
of calculations.
A second reason why prices may not closely follow changes in costs
is that manufacturers may be less willing to adjust their prices to falls
in the cost of raw materials than they are to pass on increases in these
costs. The reasons for this asymmetry include a general suspicion of price
reductions as a competitive device, and the fear that they may arouse
among buyers expectations of further reductions and a postponement
of purchases. When costs are falling and prices are not reduced the widening margin may be used for sales promotion, improvements of the product or other purposes.2
A third case in which cost-plus pricing does not reflect true changes
in cost occurs, of course, when profit margins are changed deliberately.
It appears that during the years 1955-57 margins (and prices) in the
United States were increased to cover a large rise in overhead costs in a
period of slowly rising or stagnant output. Rising overhead costs increase
the scale of output at which cost per unit of output is lowest. But actual
output did not rise substantially during the period 1955-57, and firms
seem to have decided to " recapture " overheads more quickly by increasing profit margins.3
Again, in many cases profit margins may be changed in response to
variations in demand, or in order to increase demand. As was noted
above, prices of raw materials tend to fluctuate strongly with changes in
1
Sidney S. ALEXANDER in Economics and the Policy Maker, Brookings Lectures,
1958-59 (Washington, 1959), p . 21.
2
Ruth P. MACK in The Relationship of Prices to Economic Stability and Growth, op.
cit., p. 281 and in Review of Economics and Statistics, Vol. XLI, No. 3, Aug. 1959, p. 230.
3
SCHULTZE, op. cit., p. 91. MACK (in Review of Economics and Statistics,
loc. cit.) suggests that increasing margins due to lower raw material prices not translated
in lower product prices may have stimulated the increase, particularly in overhead
labour.

PRICES, WAGES, AND INCOMES POLICIES

demand. Manufacturers could make their prices (that is, with given
costs, their profit margins) behave in a similar fashion. In fact they do so
rarely. When demand increases, producers do not typically raise prices
immediately (though they will when costs rise and, in the process, may
" make some hay while the sun shines ") but accept longer unfilled order
lists instead. Managers interested in the long-run expansion of their
firms will not lightly raise prices when there is no better reason for doing
so than that the market could bear higher prices. Also, such price increases
might provoke wage claims. Yet, once wages or raw material prices have
risen anyway, the maintenance of a constant percentage mark-up means,
of course, an increase in absolute profits. The enterprise may then
take advantage of the opportunities offered by rising demand to increase
the margin while, perhaps, " the public relates the price increases at such
times to the pay increases and thus attributes the advance to the unions." 1
If industrial price-fixers seem to be slow in responding to increases
in demand, they appear to be even more sluggish in reducing prices when
demand is falling.2 Two reasons have been given for this policy. First,
the effet of price reductions by any one firm in restoring demand is
quite uncertain. Secondly, whatever this effect may be, it will certainly
become much smaller if competitors follow suit, leaving everybody with
at best a slightly larger volume of sales but at lower prices. If prices are
reduced at all in response to declining demand, this often tends to take
the form of secret rebates rather than of overt price cuts. 3 On the other
hand, when demand is falling, management price policy may take account
of this by not raising prices when costs increase.
Uncertainty about the effect of price reduction and fear of retaliation
also explain managerial caution in trying to promote sales in a stable or a
growing market through lower prices. A new product may first be sold at
a high price leaving enough profits for advertising and the building-up of a
servicing and sales organisation, the price being reduced more or less drastically at a later stage (a large reduction may be expected to be more
effective than a series of small ones). But there are many other ways of
improving the market position of a product, less hazardous, more lasting
in their effects; and these, apparently, are often preferred: quality
improvement, credit and servicing facilities, advertising, etc.
1
J. K. GALBRAITH in Review of Economics and Statistics, Vol. XXXIX, No. 2,
May 1957, p. 129.
2
However, margins and prices may respond to changes in demand more or less
informally if special offers, discounts, etc. are more frequent when demand is slack than
when it is brisk. Furthermore, when waiting lists are long, " grey markets " may
develop for such products as steel, real prices being higher than list prices.
3

SCHULTZE, op. cit., p. 56.

CAUSES OF INFLATION AND PRICE AND WAGE FIXING

Summary.
The following general picture of pricing in large-scale manufacturing
seems to emerge from the above:
(1) There is a widespread tendency to set prices by adding a percentage profit mark-up to some concept (which may not be very accurate) of
costs of production.
(2) The profit mark-up does not necessarily aim at maximising profits
(and it rarely aims at maximising short-run profits) but at providing
some fair or stable gross return; a large part ofthat return may be used to
finance sales campaigns and investment in additional capacity, in accordance with a management aim of growth (as a result of this method of
price and profit determination, the price paid by the consumer includes
the cost of advertising and of new capital formation; both aspects have,
of course, given rise to criticism).
(3) Prices react slowly if at all to increases in demand; they are even
less responsive to falling demand.
(4) As a means of increasing a firm's share in the market, price competition is used less frequently than other forms of competition.
Three final comments may be made. First, this, of course, is at best
a pattern typical of a sizeable proportion of pricing practices in manufacturing. There are exceptions and they may be important. For example,
although manufacturing prices generally differ from raw material prices,
particularly in that they respond little to changes in demand, some weaker
firms may try during a. recession to improve their position by sharp price
reductions. Or when a fall in demand is accompanied by substantial
reductions in cost (which will often not be considered to be the case
because average overhead costs rise when output falls) prices may be
reduced somewhat more easily. Again, in a recession a strong buyer may
be able to put his suppliers under effective pressure to lower their prices.
Secondly, firms and industries may differ in the degree to which they
adhere to the practices mentioned above. It may safely be said that
prices of many manufactured goods, new buildings and transport services
respond much more slowly to variations in demand than raw materials
traded in open markets, butin any country, textiles, apparel and processed
food may show greaterflexibilityin their prices than metal products and
steel. Again, while most prices will reflect increases in costs, the adjustment may be prompter and more complete in some industries than in
others.
Finally, in most countries there is a manufacturing sector in which
prices are raised or reduced simply on the basis of price adjustments of

PRICES, WAGES, AND INCOMES POLICIES

other firms. Although the above analysis may be more typical of fairly
large firms, in practice smaller firms may nevertheless adjust their prices
in a very similar way even when their basic pricing policy may be different: to follow the lead of others rather than themselves determine
margins, sales promotion and investment policies.
Retailing
Retailers often have little freedom in fixing prices. In several countries
it is still legal for manufacturers to decide what prices they are to charge
to the customer (though the shopkeeper may naturally refuse to carry
commodities whose margins he finds insufficient). But resale price maintenance clauses are forbidden in Denmark, France, the Federal Republic
of Germany1, Sweden, and (since July 1964) the United Kingdom.2
Many retailers, especially smaller ones,findno difficulty in following the
producers' price prescriptions; it protects them from price competition.
Large retailing undertakings and co-operative chain stores may,
however, reverse the roles of shopkeeper and manufacturer in fixing
prices. They may carry articles under their own trademark and at their
own prices, ordering their supplies from manufacturers of their own
choice and at prices that they determine themselves.
Since retailers' margins are often fixed as a percentage, price increases
raise the absolute size of retail mark-ups. Manufacturers may also offer
more than average margins as part of a promotion campaign for new
products without reducing them once a market has been established and
when product improvement has reduced servicing and repair requirements. Many retailers themselves prefer wide margins to lower prices so
that they can promote sales through non-price competition. For example,
the Canadian Royal Commission on Price Spreads of Food Products
ascribed the increasing proportion of distribution costs in retail food
prices after the war largely to increasing retail services (" contests, giveaways and gimmicks ") which it regarded as rather wasteful. There has
been controversy over " trading stamps " in the United Kingdom. In
general, high retail margins appear to have had the result of making
retailing an overcrowded trade, in which productivity is quite low. In
1

In this country, however, the prohibition of resale price maintenance seems to
have remained a dead letter and under Danish and Swedish laws the manufacturer may
" suggest " a resale price, but cannot legally enforce it.2
Where, however, existing resale price maintenance agreements, if registered with
the Registrar of Restrictive Practices, remain enforceable until the Restrictive Practices
Court has passed judgment on them. A number of substantial price reductions were
reported after the 1964 Act came into operation.

CAUSES OF INFLATION AND PRICE AND WAGE FIXING

recent years, however, modern forms of retailing (self-service stores) have
developed, offering less service but cheaper goods.
WAGE-FIXING PROCESSES

Procedures for the adjustment of wages differ considerably as between
countries and, sometimes, as between industries. Yet a few important
common features may be discerned that are of interest to the problem
of price stability.
Wages may be fixed unilaterally by the employer; this may be called
" management-administered wage determination ". 1 Other procedures
include collective bargaining at one or more levels—plant, company, industry or economy—or some legal machinery in the nature of minimum
wage-fixing institutions or compulsory arbitration procedures. Collective
bargaining is the normal practice in countries with a high degree of
workers' and employers' organisation, such as Denmark, Norway and
Sweden. It is less important in countries or economic sectors in which
one of the two sides is not well organised, as in Canada, Japan and large
parts of the United States economy. In most countries, the exceptions
being in Scandinavia, less than half of all employees are organised. The
numbers covered by collective agreements are, however, usually larger
than the membership of trade unions, and both are relatively larger for
manual workers in manufacturing, mining and railways than for all
employees in the economy as a whole.
(Minimum) Wage Boards are of considerable importance in the
United Kingdom; statutory national minimum wages are of importance
in France and the United States. Compulsory arbitration is the main
instrument of wage adjustment in Australia.
Collective agreements and other forms of wage regulation fix various
types of rates (straight time-rates, piece-rates and overtime rates, premiums for work on holidays, shift differentials, and various allowances)
and may contain more or less detailed rules of application and administration. For example, there may be rules concerning the average amount
of bonus that piece-workers should earn above basic rates; for the
adjustment of time and piece rates to changes in jobs or working conditions ; for the distribution of overtime ; for the adjustment of wages to
changes in the cost of living, productivity, profits, etc. Rates may be
understood, formally or informally, to be minimum rates, actual rates
1
George W. TAYLOR in George W. TAYLOR and Frank C. PIERSON (eds.): New
Concepts in Wage Determination (New York, McGraw-Hill, 1957), pp. 91 ff. Civil
service salaries are usually in this group although in practice the adjustment of salary
scales is preceded by consultation procedures more or less approaching collective bargaining.

PRICES, WAGES, AND INCOMES POLICIES

being expected to be higher (as in the case of industry-level collective
agreements in several European countries) or they may be standard rates,
actual rates being neither higher nor lower (as in the Netherlands and the
United States).1 Rates and rules of administration, together with actual
conditions of production (which affect overtime, productivity, etc.) and
any freedom that the parties may have to pay or receive higher than
contractual rates, determine the actual amount of wages earned.
Earnings, therefore, may rise and fall over short periods even while rates
remain constant. Discrepancies between formal rates and actual earnings
create major problems in wage policy; they will be discussed further in
later chapters.
Collective Bargaining
The main advantage of fixing terms of employment through collective negotiation is, of course, that the parties concerned agree and feel
committed to them. For the purposes of this study the main question
raised about this method has become whether the movements of labour
costs that it, or certain forms of it, tends to bring about are compatible
with other social objectives—especially with stability in the comprehensive sense given to that term in Chapter I.
It is not necessary to spell out here the major benefits of having
agreed answers to the innumerable questions of organisation and methods
of work, and of the distribution of the proceeds of the work, that have to
be settled in the highly complex production systems of industrial economies. Many of these questions involve some conflict of interest between
management and workers in the sense that a gain for one side entails a
loss to the other.2 Successful negotiation resolves these conflicts by
allowing each side to bring the pressure it can wield to bear upon the
other—but without actually having to apply this power, (basically, ability
to inflict damage) in open industrial warfare. Having exerted such power
as they possess to the best of their negotiating abilities, the parties are
likely to accept the outcome of negotiation with a far greater degree of
1
Both types may occur in any country. For example in Denmark about 40 per
cent, of the wage earners are covered by collective agreements for particular industries or
crafts fixing minimum wages, the rest by agreements specifying standard rates (see
Anders Henrik DAHL in Nationalekonomisk Tidsskrift, 1959, No. 3/4, p. 197 n.).
2
Modern industry also provides, of course, many important opportunities of a
mutually beneficial nature. Cost reduction and technical progress are cases in point.
The Scanlon Plan for workers' participation in the solution of production problems
and the long-range sharing plan for the Kaiser Steel Corporation providing both
employment security and sharing of gains with regard to technical change are two
well-known examples, both from the United States, of a mutually advantageous
approach to problems that have often given rise to conflict.

CAUSES OF INFLATION AND PRICE AND WAGE FIXING

finality than they would do with other methods of settlement. In broad
sectors of most industrial countries, collective bargaining has now
reached a stage of maturity where negotiation normally is successful—as
a result, in large part, of mutual recognition of certain common interests
(for example in the continued existence of a bargaining relationship between the respective organisations) in addition to inevitable divergencies
of interest.
Such a system is a social asset of very high value, but its appraisal
cannot stop at an evaluation of its ability to channel power and to resolve
conflicts in the determination of terms of employment. The resulting
wages and other conditions themselves must be judged with reference to
the main objectives of economic and social policy. And from this angle
the fact that these conditions enjoy the approval and agreement of those
directly concerned is not, in itself, sufficient because the responsibilities
of these private groups do not extend to the general price level, economic
growth and the balance of international payments. These ingredients of
economic stability are the concern of public authorities, whether as part
of some comprehensive plan (as in France) or merely in the way of overall regulation of the monetary and fiscal framework within which a
market economy has to function. The public interest in over-all economic
stability is not a consideration that can be expected to play an important
role in the work of wage negotiators, especially when the bargaining unit
is small — i.e. the individual plant or enterprise (which is the common
bargaining unit in the United States, although some enterprises in that
country are of course very large)—or even in industry-wide bargaining
(as in Denmark, the Federal Republic of Germany, and the United Kingdom, where industrial agreements, however, provide merely a floor for
further bargaining at district and plant levels). Indeed, the formal acceptance by governments of full employment as a major goal of national
policy, and the knowledge that relatively effective means of achieving this
end now exist, may well have weakened considerations of the possible
negative effects of wage and price increases that might otherwise have
restrained parties in their wage settlements.
Trade Union Criteria
A trade union's approach to negotiation is, of course, determined by
the special type of organisation it is. Its leaders have to satisfy the wishes
of the membership, and failure to do so may lead to their replacement or
to the displacement of the union by another one. For this reason, the
achievements of the negotiators of one union often set clear and imperative targets for the negotiators of other unions. The great importance of

PRICES, WAGES, AND INCOMES POLICIES

this fact is that it tends to introduce a certain rigidity in the wages structure.
Wage earners and their representatives are likely to be sharply aware
of the hardships liable to be entailed if levels of living fall, and also of the
dissatisfaction and sense of injustice that are apt to be felt if there is a
deterioration in the relative levels of some groups as compared with
others. It would seem that the following tenets have come to shape, more
or less firmly, wage earners' expectations from wage negotiations:
(1) Except in very dramatic circumstances, for instance when the
only alternative would be the closing down of the plant or firm, basic
rates should never be reduced.
(2) Earnings should generally not be reduced either; for example, if
a period of large overtime earnings is followed by a return to standard
hours, pressures are likely to develop for compensation in the form of
higher basic rates or a shorter standard working week; adjustments of
piece rates to take account of changes in production methods should preferably give the workers not less than the earnings achieved under the old
piece rate.
(3) Earnings should rise in at least the same proportion as the cost of
living, though there may be more or less strong opposition to downward
adjustments even when these are foreseen in a sliding-scale arrangement.
(4) Wage increases for one group of workers create a presumption
that other wage earners should receive a similar increase in either absolute or relative terms; for example, increasing incentive workers' earnings
should sooner or later be followed by similar increases in time wages.
These points may be regarded as standing orders in the briefs of most
union negotiators. A bargainer who disappoints any of these expectations
is likely to be in trouble. But satisfying these claims is merely his minimum task. In addition, he is expected to improve his members' position
further by gaining for them a portion of the firm's profits. High and
rising profits are inevitably the basis of high wage claims.
Management Attitudes
As noted above, the adoption of official full-employment policies may
have weakened some constraints on wage negotiators—the fear of unemployment and falling output as the result of cost increases. In particular, employers' resistance to wage demands may have become weaker.
Furthermore, it appears that modern views on the objectives and tasks of
management have fostered a tendency among personnel and wage administrators to " lean over backwards " in trying to meet demands and to
avoid dissatisfaction. A main reason for this attitude is concern about pro-

CAUSES OF INFLATION AND PRICE AND WAGE FIXING

ductivity, especially when expensive equipment involving new working
methods and job allocations is constantly being installed. " Workers'
dissatisfaction is a marvellously efficient way of ensuring inefficient production. "1
But there may be more behind what seems to be, at least in some
important groups of large enterprises, a management attitude lubricating
the wage-determination machinery. For example, a management objective of gradual expansion of the firm may lead wage administrators to
seek to establish for the enterprise the reputation of being a well-paying
employer; such a reputation helps the firm in its recruitment. High
wages are by no means the most important reason for either new entrants
to the labour market or other job-seekers to accept a particular job;
the interest of the work and parents' wishes seem to be much more
common and important reasons. Nevertheless, " good wages " do play a
part in the choice of jobs and an employer offering such wages is better
placed in attracting good workers than employers offering average or low
wages. This is, of course, an advantage to a firm expecting that it will
grow in the long run or one liable to face sudden increases in demand in
the short run. A well-paying firm has an advantage in being able to recruit
the pick of the labour force.2
So the modern emphasis on good human (and public) relations may
lead employers to adopt an accommodating attitude in matters of wage
determination to the extent that this is compatible with management's
growth objectives. Of course, wage bargaining is often hard and bitter.
Nor is the level of wages itself the only or the main condition determining
the state of labour-management relations in afirmor industry. Again, the
limits to which employers can go in meeting workers' demands depend on
factors beyond their control: the degree of competition in the product
market, the proportion of labour costs in total cost of production, etc.
But there does seem to be a greater desire to meet workers' demands
when possible than there was in the different climate prevailing, say, 50
or even 25 years ago. When a strong union exists anyway, the employer
may not oppose a demand for compulsory membership. When a satisfactory incentive system exists, he may not insist on a reduction in piece
rates when productivity increases somewhat, even though this is due
entirely to better materials or equipment and not to increased effort on
the workers' part. When, due to higher productivity, a choice arises

SCHULTZE, op. cit., p. 68.

Alfred KUHN in Industrial and Labor Relations Review (Ithaca, N.Y.), Vol. 12,
No. 2, Jan. 1959, p. 245, further suggests that managers may be willing to pay relatively high wages as a " conspicuous display of economic and managerial superiority ".

PRICES, WAGES, AND INCOMES POLICIES

between higher wages and lower prices, he may regard the former as
the better alternative.
Furthermore, while such attitudes may be confined to a relatively
small group of progressive employers, when their position in the labour
market is of any consequence at all trade unions will bring pressure to
bear on other employers to follow the example. There are strong pressures
in other firms and industries for wage increases similar to those obtained
in enterprises that, voluntarily or not, are acting as " wage leaders " or
" pattern setters ".
As a result, while industries differ substantially in prosperity, in the
rate at which productivity increases, and in the extent to which they are
able to pass on wage increases to the consumer without causing a fall in
demand, inter-industrial differences in wages and wage increases are far
smaller. And since the pattern setters usually are prosperous industries
with fast productivity growth, this could lead to over-all wage increases
exceeding the over-all rate of productivity increase.
Figures calculated by the United Nations a few years ago suggested
that during the period 1950-56 in several countries, average hourly
manufacturing earnings tended to rise at rates approximately equal to
rates of productivity increase in those individual industries in which
output and productivity grew considerably faster than the average for
the economy as a whole. As the report says—
Even if earnings in the dynamic industries do not rise to the full extent of the
productivity advances in those particular industries, they may still set a target
rate of wage increases for other industries which is well in excess of the productivity growth achieved.x
A more detailed study of average hourly earnings of male wage
earners in 111 British industries over the period 1951-56 showed a high
degree of similarity of wage increases as between these industries, especially within groups related by the similarity of their products and production techniques (i.e. within " two-digit " industry groups in the
terminology of the United Nations International Standard Industrial
Classification of All Economic Activities).2 Afurther international study of
wage increases over the period from around 1950 until around 1960 also
showed a high degree of similarity in wage increases as between various
broader industrial groupings, slightly more so in western Europe than in
North America, more so over longer than over shorter periods. The latter
finding suggests that even when existing wage relativities may be loosened
1

United Nations: World Economic Survey 1957 (New York, 1958), p. 36.
W. B. REDDAWAY in Lloyds Bank Review (London), N.S. No. 54, Oct. 1959,
pp. 32 if.
2

CAUSES OF INFLATION AND PRICE AND WAGE FIXING

during a short period, sooner or later they tend to be restored again.1
These findings of relative rigidity in the structures of wage relativities
correspond to one of the trade union criteria for wage fixing mentioned
in the previous section. They are also compatible with the notion of wage
fixing in individual industries or firms following, to some extent at least,
patterns set by certain key industries or key employers within industries.2
Wages and the Demand for Labour
Most of the factors mentioned above as grounds for wage increases
bear little explicit relation to the demand for labour, itself determined by
the demand for its products. It is certain that wage rates rarely fall when
the demand for labour becomes weaker. In fact, they may rise considerably even when employment is falling. In the British cotton industry,
when recorded unemployment in 1952-53 amounted to 30 per cent., the
unions claimed and obtained wage increases. " It seems that the cotton
unions were more fearful of their members' wage rates falling behind
those of other industries than of the less determinate effect of a wage
increase on employment." 3 Between 1948 and 1952 the number of production workers in the United States bituminous coal-mining industry
fell from 438,000 to 304,000 but during this period the unions claimed
increases in wages and fringe benefits larger than those obtained in the
expanding automobile and steel industries.4 From 1953 to 1955 earnings
in the industry rose by little more than 2 per cent, compared with 6 per
cent, in manufacturing, while employment fell to 199,000, but after an
increase in employment and working hours in 1956 wages resumed their
fast rate of increase. A statistical study of post-war wage movements in
the United Kingdom suggested that, although there seemed to be a relation between the demand for labour and wage increases in the economy as
a whole, economic conditions in individual industries were of much less
importance in explaining industry wage advances.5
It would seem, however, that wages do respond to increases in the
demand for labour. First, earnings rise, of course, when working hours
1
Organisation for Economic Co-operation and Development: Wages and Labour
Mobility (Paris, 1965), table 4, pp. 30-31.
2
Individual firms do not necessarily seek to act as wage leaders; on the contrary,
they may be reluctant to play this part. See, for instance, comments on the position of
the United States Steel Corporation in this regard, as described in Department of
Labor, United States: Collective Bargaining in the Basic Steel Industry (Washington,
1961), pp. 86 ff.
3
H. A. TURNER in The Theory of Wage Determination, op. cit., p. 129.
4
Arthur M. Ross in New Concepts in Wage Determination, op. cit., p. 191.
5
L. A. DICKS-MIREAUX and J. C. R. Dow in Journal of the Royal Statistical
Society (London), Vol. 122, 1959, Part II, pp. 145 ff.

PRICES, WAGES, AND INCOMES POLICIES

increase.1 Secondly, in so far as a high demand for labour in any sector
is accompanied by rising profits it is apt to lead to wage claims supported
by a strong bargaining position for the union. Thirdly, employers may seek
to meet their increasing needs for labour by offering higher wages, perhaps especially in cases in which formal agreements fix minimum rates
only. But any significant increase, especially in formally bargained rates,
is likely to provoke wage advances in other sectors, whether demand
there has increased or not.

1
Just as they fall when overtime is reduced without any other changes; but whereas
this sets up pressures for compensation as noted above (not that such pressures are
always successful), there appears to be no counterpart to these pressures in the case of
a rising demand.

CHAPTER IV
SOME MECHANICS OF INFLATION
INTRODUCTION

The brief review above of price- and wage-fixing methods suggests
that these may in two ways contribute to a situation in which countries
cannot attain or maintain full employment without incurring unacceptable increases in the price level.
First, these methods appear to have created, between and among
prices and wages, a number of fairly rigid relationships which are maintained by upward adjustments only. Thus, increases in consumer prices
make for higher wages and increases in wage costs per unit make for
higher prices. Furthermore, when some firms set higher prices for their
products other enterprises (particularly but not only those that buy the
products) are apt to follow suit, and when the wages of one group of
workers are raised those of other groups also tend to rise. Hence, any
substantial rise in individual prices or wages (for example an increase
in the price of imported raw materials, the reduction of rent controls or
consumer subsidies on food, or wage increases in one industry) may set
in motion a whole series of further increases, which in their turn may give
rise to still further increases, even in the absence of any general excess of
demand. But it is rare for prices, and even rarer for wages, to be reduced
when there are reductions in the cost (whether of living or of production).
In the second place, precisely because of these rigidities, the scope for
management and trade union action to raise prices and wages may well
have increased. When cost-plus pricing is widely accepted (even to the
extent of prices being raised when underutilisation of capacity leads to
higher overhead costs per unit), when prices are co-ordinated as between
firms (through retail price maintenance or by tacit or formal " price
leadership "), and when wage increases in some firms or industries can
be expected normally to be followed by similar increases in other firms
and industries, there is bound to be less hesitation about price and
wage increases than there would have been if price competition had
been a major feature of market economies, and if the structure of wage
differences had been flexible.

PRICES, WAGES, AND INCOMES POLICIES

It seems clear therefore that " cost-push " could occur by itself, or
add significantly to price movements due to strong demand. But for
policy making, since the appropriate measures to cure demand inflation
differ from those to cope with cost inflation, it is important to know
whether and to what extent cost-push has actually occurred. In any
inflationary situation, how can the demand and the cost elements be
separated ?
Analytically, one may say that demand-pull exists when demand is
running at a level that cannot be satisfied at prevailing prices. Then
either prices must rise to choke off excess demand or there will be suppressed inflation (queues, etc.). If wages and other incomes rise, prices
must rise that much more (or queues must get longer). Demand-pull can
be remedied byfiscalor monetary measuies to close the inflationary gap.
Cost-push can occur at any time, whether or not there is excess
demand. It needs to be operated on directly.
The difficulty is to distinguish between the two cases in practice;
this is all the greater because they are liable to be combined in different
proportions in different situations.
It should be recognised from the outset that there is no satisfactory
method of making such a separation in practice. Hence, policy decisions
are bound to involve a substantial element of judgment. But, since a
situation in which full employment cannot be maintained without unacceptable price increases is a serious and acute problem, measures to cope
with it will have to be taken even on a relatively weak analytical basis.
In this chapter the problem of identifying sources of inflation is further
examined in the light of some of the available evidence.
INDICATORS OF DEMAND INFLATION AND COST INFLATION

To distinguish the sources of price increases that can be corrected
only by curbing demand from those that could not be so corrected, indicators are needed that would necessarily accompany demand inflation
but could not occur in the case of cost inflation. This requirement rules
out a mere finding that both prices and wages have been rising—this
would happen in both situations. It also rules out a finding that money
wages have risen faster than productivity—this again would happen
whether the large wage increases were the result of strong trade union
action or of a labour shortage resulting from excess demand. Indeed, no
firm conclusions could necessarily be drawn from a finding that wages
rose first, and prices later. Although this would suggest that the wage
increase had caused the price increase, it might also be the case that the
wage increase was a delayed adjustment to a much earlier price increase.

SOME MECHANICS OF INFLATION

Or wage earnings might have risen simply as the result of labour shortage
due to an increase in demand, while perhaps for such reasons as have been
mentioned above 1, management did not immediately raise prices in
response to the increase in demand.2
It may be doubted whether any general economic indicators (for
instance over-all price and wage levels) could help in separating demand
inflation and cost inflation nearly as effectively as a set of more detailed
indicators (for instance movement of prices, wages, employment and
output in certain individual manufacturing industries and in certain
types of service activities). This more detailed approach would require the
regular collection of new types of data as well as a theoretical framework
for analysing them.
But two over-all indicators of the character of an inflation that may
nevertheless be quite useful are (a) changes in the level of demand for
labour, as an indicator of changes in aggregate demand; and (b) the
relation between profits and wages.
The Demand for Labour
The demand for labour is derived from the demand for goods and
services. There is, however, no full, one-for-one, correspondence between aggregate demand and the demand for labour. To some extent, the
importance of the divergence between the two demands is reduced if
analysis is confined to one particular economy over relatively short
periods of time. We can then assume no significant change in the factors
causing a divergence between aggregate demand and the demand for
labour, and a change in aggregate demand may be expected to result in a
change in the demand for labour, the magnitude of which would depend
upon the hypothetically stable relation between changes in the two
demands.
Assuming a stable relation between changes in aggregate demand and
in demand for labour, we can accept the latter as a reasonable proxy for
the former. However, before we can attempt to select some indicators
of demand for labour we have to solve a conceptual problem: what do
we mean by demand for labour or, for that matter, for a commodity or
service? Strictly speaking, demand is the quantity effectively demanded
at any given price. The relations between quantities and corresponding
prices form a schedule of which only one point corresponds to the actual
situation at any given point of time. Estimation of a schedule of demand
1

See above, pp. 65-66.
Paul A. SAMUELSON and Robert M. SOLOW in American Economic Review, Vol. L,
No. 2, May 1960, pp. 177 ff.
2

PRICES, WAGES, AND INCOMES POLICIES

for labour, however, is not our primary interest in a discussion of inflation. Our principal concern is with the pressure unsatisfied demand
exercises on prices. We are interested in the pressure of demand: in other
words, in a relative concept of demand, a concept which relates total
demand to total available supplies.
The indicators of demand for labour are based on the concept of
pressure of demand, or " excess demand ". Paradoxically, the possibility of measuring demand (through the excess demand concept) depends
on the relative inflexibility of the price mechanism.
If prices were completely flexible, the rise in prices would proceed very
rapidly to the point where excess demand was eliminated, so that it would be
difficult to get a true measure of the motive force which was propelling prices
upward. But.. .prices are notflexibleto this degree; and excess demand in fact
seems capable of being a persistent condition, often lasting years. It seems
possible, then, to treat it as a magnitude capable of at least rough measurement.1
One of the indicators of the pressure of demand for labour is provided by unemployment statistics. In using these as an indicator of the
pressure of demand for labour we face the problem of distinguishing
unemployment due to insufficient demand for labour from other types
of unemployment—specifically from frictional unemployment. There is
no fully satisfactory way of distinguishing between frictional unemployment and unemployment due to insufficient demand for labour. The
standard approach is to take unemployment rates over a period of years
and assume that in the year with the minimum rate all unemployment was
frictional. This frictional unemployment rate is then assumed to remain
stable over a period of years and in each particular year the difference
between it and the actual unemployment rate is assumed to be due to
slackness of demand. There are two basic shortcomings of this approach.
First, the possibility of excess demand for labour is excluded by definition.
The highest point on the scale of demand for labour is the absence of
insufficient demand, and in all other years some slackness of demand is
responsible for unemployment. Second, the choice of the period is by
necessity arbitrary, and consequently so is the year with the lowest unemployment rate. The apportioning of the total unemployment rate between frictional unemployment and demand-deficiency unemployment
in any particular year thus depends on the period within the context of
which we analyse the unemployment data.
These two shortcomings do not, however, invalidate the use of the
unemployment data for the estimation of the pressure of demand for
1
J . C. R. D o w : The Management of the British Economy 1945-1960 (Cambridge,
University Press, 1964), p. 335.

SOME MECHANICS OF INFLATION

labour, if we confine our attention to year-to-year changes in the pressure
of demand.
Statistics of unemployment as an indicator of the pressure of demand
for labour have, however, other deficiencies. The most serious one is
linked with the method by which unemployment statistics are collected.
In a country where the statistics are compiled on the basis of registration
by the unemployed the data may underestimate the slackness of demand
for labour if there is a general feeling that it is useless to register as unemployed since the chances of obtaining a job are small anyhow. This
phenomenon not only results in an underestimation of the magnitude of
unemployment but also obscures the year-by-year changes, for the
degree of underestimation may change greatly, and inversely with changes
in the pressure of demand.
It is sometimes claimed that there is another shortcoming of unemployment statistics as an indicator of the pressure of demand for labour.
This is the fact that employers " hoard " surplus labour in expectation
of a recovery, and thus the unemployment data overestimate the pressure of demand for labour. Strictly speaking, this argument is not correct.
The very fact that employers are prepared to pay the hoarded labour
indicates that they have demand for it, even if they do not use it productively at the moment. They find it more economical to retain than to
dismiss surplus labour and later laboriously to try to recruit and train
new labour. To argue that the hoarding of labour distorts the picture
of the demand for labour is to subscribe to the theory that the driving
force of the market economy is profit maximisation at every point of
time. The more realistic theory of profit maximisation over a period
of time is compatible with the view of hoarded labour being effectively
demanded.1
There are, however, cases of hoarding surplus labour which do distort
unemployment statistics as an indicator of the pressure of demand for
labour. When employers are prevented from dismissing workers by law,
collective contracts or trade customs, hoarded labour cannot be considered as effectively demanded, and unemployment statistics give an
exaggerated impression of the pressure of demand for labour.
Statistics of unemployment, then, present one basic indicator of the
state of demand for labour. Another basic indicator is statistics of unfilled vacancies. Unlike unemployment statistics, which are inversely
linked with the pressure of demand, statistics of unfilled vacancies pro1
This argument, however, should not be understood as denying the distortion,
through hoarding of labour, of productivity statistics. Hoarding of labour may also
diminish the usefulness of indicators of demand for labour as indicators of the state of
aggregate demand.

PRICES, WAGES, AND INCOMES POLICIES

vide a direct measure of excess demand for labour. They, too, have their
shortcomings. On the one hand, they may overstate the excess demand
through employers registering more vacancies than they really have;
on the other hand, many vacancies may be registered not with public
employment services that publish statistics but with private agencies, or
not at all. Nevertheless, statistics of unfilled vacancies provide a useful
check on the first indicator.1
The main use of statistics of unfilled vacancies for measuring the
pressure of demand for labour is, however, in combination with
statistics of unemployment. A simple method of combining the two
indicators is to subtract unemployment (either in absolute or in relative
terms) from the unfilled vacancies. A positive residual indicates an excess
demand, a negative residual a deficient demand for labour. This simple
method may be elaborated by taking into account only demand unemployment (i.e. unemployment data after the deduction of the estimated
frictional unemployment) and data of available vacancies.
Unemployment and vacancy statistics may, however, be combined
in a more elaborate way. The best known indicator for excess demand for
labour in the British economy, the Dow-Dicks-Mireaux index 2 , is
based on these two statistical series. This index takes into account the
possibility that the data of unfilled vacancies understate or overstate the
true number, and an attempt is made, through indirect methods, to
estimate the " statement ratio ", i.e. unfilled vacancies reported to employment offices as a proportion of true unfilled vacancies. The DowDicks-Mireaux index also takes into account the fact that unemployment and unfilled vacancies are not linear functions of excess demand.
As excess demand increases, unemployment falls, but as it cannot fall
below zero a given increase in demand will be associated with progressively smaller reductions in the unemployment rate. Similarly, as
demand dwindles, the number of unfilled vacancies will fall, but as it
cannot drop below zero the vacancies index will become decreasingly
sensitive to falls in demand.
Even an elaborate indicator of the excess demand for labour, such
as the Dow-Dicks-Mireaux index, cannot be considered a fully satisfactory cardinal measure of the pressure of demand. As its authors themselves
point out, the index does not give a precise indication of the zero excess
1
See Dow, op. cit., p. 115. Graphs for the United Kingdom economy between
1946 and 1960 giving the percentage of unfilled vacancies and the percentage of excess
demand for labour (with—on an inverted scale—the percentage of unemployment) show
a remarkable correspondence between the changes in the two indicators.
a
See J. C. R. Dow and L. A. DICKS-MIREAUX in Oxford Economic Papers, New
Series, Vol. 10, No. 1, Feb. 1958, pp. 1 ff.

SOME MECHANICS OF INFLATION

demand point (i.e. the dividing point between deficient and excess demand
for labour), and the estimates of the margin of error in the reported
number of unfilled vacancies are not very precise. Nevertheless, the
index of excess demand for labour may be considered a good ordinal
measure, one showing the direction of year-to-year changes in the pressure
of demand.
In spite of the various shortcomings of the indicators discussed above,
it can be accepted that the best indicator is some index based on data of
unemployment and of unfilled vacancies. To make such an index is,
however, impossible in the case of countries (such as the Federal Republic
of Germany, Italy and the United States) with no information on the
numbers of unfilled vacancies. There the pressure of demand for labour
has to be estimated on the basis of unemployment statistics alone, with
some reference to other conditions which constitute evidence of excess
demand. These conditions provide rather less reliable indicators of the
pressure of demand than the unemployment and unfilled vacancy data,
but the fact that all or most of them usually point in a similar direction at
any one time provides a useful source of information on the changes
in the pressure of demand.
Some supplementary indicators can be derived from employment
statistics. The rate of growth of employment (rather than its absolute
magnitude) may be used as an indicator of the pressure of demand.
Periods with a higher-than-average rate of growth of employment can
be considered as exhibiting excess demand, and vice versa. This method
may be modified in various ways. The data on the rate of growth of
employment may be corrected for increases in the population of working
age and the resulting rate of change in the participation rate may be
used as an indicator of the pressure of demand.1
Another group of indicators of the pressure of demand can be found
in statistics of labour turnover. On the assumption that high labour
turnover is stimulated by a strong demand for labour, the former can
be used as an indicator of the latter. Along similar lines, another proxy
for demand for labour is sought in the " quit rate ", the number of spontaneous departures relatively to the labour force. Several variations are
possible (the availability of data permitting), such as using the number of
spontaneous departures net of those due to retirement, sickness, death,
military service, etc., or comparisons between the number of spontaneous
departures (assumed to reflect excess demand) and the number of dismissals (assumed to reflect, to some extent, insufficient demand).
1
A more sophisticated modification has been suggested by W. A. H. GODLEY and
J. R. SHEPHERD in the National Institute Economic Review (London, National Institute
of Economic and Social Research), No. 29, Aug. 1964, pp. 27-28.

PRICES, WAGES, AND INCOMES POLICIES

A third type of evidence of the pressure of demand for labour can
be derived from wage statistics, through the comparison of negotiated
wages rates and actual earnings. This type of evidence is based on the
assumption that actual earnings depend not only on national wage agreements but on " hundreds of local bargains at the ' work place ' —and
that these are sensitive to demand ".1 The difference between rates and
earnings, called the rate of wage drift, is thus assumed to reflect the pressure of demand. A number of studies have shown that the rate of wage
drift reflects the pressure of demand only imperfectly, but nevertheless
this indicator may be used as an auxiliary piece of evidence about the
demand for labour. In the same category of imperfect but occasionally
useful indicators of the pressure of demand may be included statistics
on the number of hours worked (making allowance for changes in the
standard length of the working-week) and the existence of labour immigration or emigration.
Thus there are a good many indicators of changes in the pressure of
demand for labour. These usually do not show conclusively whether or
not there is excess demand and an " inflationary gap ".2 Starting from a
low level, demand could increase considerably, and continue to do so for
quite a long time before becoming excessive; starting from a very high
level, it could fall quite considerably while still remaining excessive. But
taken in conjunction with information about the level of demand, these
indicators of changes in that level can be useful in diagnosing the nature
of the inflationary pressures prevailing or predominating at a particular
time. Experience shows that prices and wages rise not only at times when
the indicators suggest increasing pressure of demand but often at other
times too, though usually less fast.
In the United Kingdom, for example, excess demand for labour,
measured by the Dow-Dicks-Mireaux index, fell from 1.2 per cent.
in 1956 to 0.1 per cent, in 1957, to —0.9 per cent, in 1958 and to —1.6
per cent, in 1959. At the same time, average wages and salaries per employee increased by 5.6 per cent, in 1957, by 3.9 per cent, in 1958 and by
2.9 per cent, in 1959.3 The consumer price index rose by 5 per cent, in
1956, 4 per cent, in 1957, 3 per cent, in 1958, and remained stable in
1959.

See Dow, op. cit., p. 352.
See above, p. 78.
3
See Dow, op. cit., p. 347. The values of the excess demand index given in this
work are twice the values originally estimated by Dow and Dicks-Mireaux. The index
for each year represents the average of the first three-quarters of the year, and of the
last quarter of the previous year.
2

SOME MECHANICS OF INFLATION

The Relation between Profits and Wages
Profits are the residual item in the incomes of enterprises after other
costs have been met. Thus, one might expect profits to rise more than
wages in a demand inflation, and wages, conversely, to rise more than
profits during a cost inflation. Alternatively, it might be argued that both
market prices and production costs are likely to respond to demand
pressures, so that both profits and wages would rise when demand was
excessive, but that costs, and particularly wages, would rise only after a
short time interval, while profits would tend to rise earlier.
There are, however, considerable difficulties in using published statistics of business profits and wages for the purpose of identifying the nature
of an inflation, and a change in the relation between the two for the economy as a whole, unless it is very pronounced, is likely to be concealed by
numerous other things which could be taken into account satisfactorily
only if national income statistics were presented in much more detail
than they are now. To test any hypothesis involving a " wages lag "
would require statistics on a more frequent basis than those given in
national income accounts.
While statistical analysis of the relationship between wages and profits
for the economy as a whole is therefore very difficult with the present
state of data, various studies of a more restricted scope have been made.
Among these, some inquiries into post-war movements of wages and
profits in a number of British industries showed no significant relationships between the two at all.1 In the United States, quite strong associations were found to exist between the level of profits and percentage
changes in wage rates, but not in a sense that would lend strong support
to the hypothesis that such association was determined by pressures of
demand. For instance, in the case of manufacturing as a whole and in a
number of individual manufacturing industries between 1947 and 1958,
although percentage changes in hourly earnings on a straight-time basis
were strongly related to profits (measured as a rate of return on shares),
no important relationship was found between wage changes and more
direct indicators of the state of demand such as output, and employment
of production workers. Wage increases were also found by one author 2
to be related to the concentration ratio (a measure of the degree of mono1
L. R. KLEIN and R. J. BALL in Economic Journal (London), Vol. LXIX, Sep. 1959,
pp. 465-482 and Richard LIPSEY and M. D . STEUER in Economica (London), New
Series, Vol. XXVIII, No. 110, May 1961, pp. 137-155.
2
Harold M. LEVTNSON: Post-War Movement of Prices and Wages in Manufacturing
Industries, Study Paper No. 21 for the Study of Employment, Growth and Price Level
(Washington, Government Printer, 1960).

PRICES, WAGES, AND INCOMES POLICIES

poly) in the industries concerned. Another author, after finding that
(especially in the post-war period) money wage changes in United States
manufacturing responded but weakly to the level of unemployment,
concluded that a much better explanation of wage changes could be
obtained from a study of the rates of profits and changes in these rates.1
Thus the evidence, such as it is, suggests that profits and rates of
wage increases in the United States have been moving up and down
together, but not in concert with the demand for labour. This does not
accord well with the hypothesis that profits and wages were moving
together on account of excess demand. This finding is, however, not
inconsistent with certain situations of cost inflation—for instance, that
in which wage increases (which may themselves be largely independent
of the state of demand) give rise to profit increases though constant percentage profit mark-ups.2
FACTORS IN INFLATION

While the type of evidence referred to above is not clearly conclusive,
it casts doubt on the hypothesis that inflation is usually a mere matter
of excess demand for goods and labour. Some of the findings seem much
more compatible with models of price and wage fixing (such as those
sketched in the previous chapter) in which cost pressures also play an
important part.
To note this is not the same as saying that modern management or
trade unions or both must be " blamed " for inflation. The policies and
attitudes sketched in the previous chapter may be considered perfectly
natural and reasonable, given the functions that managers and trade
union leaders are expected to perform, and given the fact that both may
nowadays entertain a fairly confident expectation that high levels of
production and employment will be maintained by appropriate measures
of government policy. In particular, it would be incorrect to say that
wages alone were responsible for inflation.
It is often taken for granted that wage increases cause prices to be
higher than they would have been otherwise. Indeed, it may seem obvious that an employer, when he has to incur higher wage costs and if he
is to stay in business, should recover them by raising his prices. On this
1
Rattan J. BHATIA in Economica, New Series, Vol XXVIII, No. I l l , Aug. 1961,
pp. 286-296, and ibid., New Series, Vol. XXIX, No. 115, Aug. 1962, pp. 255-262.
See also William G. BOWEN: Wage Behavior in the Postwar Period (Princeton, N.J., the
University, 1960), Ch. V, where a decline is noted in the influence of inter-industry
differences in profits on inter-industry differences in rates of wage increases as between
earlier and later postwar years.
2
See above, p. 66.

SOME MECHANICS OF INFLATION

basis it might also be asserted that, since wage increases merely lead to
higher prices, they are largely self-defeating, even from the employees'
point of view.
Yet this clearly is not always the case. For mineral and agricultural
products, whose prices are determined by supply and demand in open
markets, the cost of wage increases cannot be passed on to prices by the
employer. Instead, the wage rise is matched by a fall in profits or employment or both.
But in " competitive " manufacturing industries, too, the possibilities
of passing wage increases on to prices may be quite limited, so that either
wages lag behind those in other sectors or wage increases tend to be
accompanied by declining profits or employment, and more or less
constant prices. The former was found to have occurred in " non-concentrated " industries in the United States—in which output is divided
among a fairly large number of medium-or small-sized firms, such as
textiles, clothing and shoe making—and even in large firms subject to
strong competition from other products.1
However, when conditions in the market for goods permit management to fix prices within a fairly wide range of discretion, wage increases
are often promptly followed by price increases. This will be especially the
case when management policy is to maintain constant profit mark-ups
over cost of production.
As was noted in the previous chapter, the price that the buyer pays
often includes a substantial profit margin. Furthermore, this margin is not
always a mere reward necessary to induce employers to continue working
in their business and assuming entrepreneurial risks, but it may also
include provision for expansion of the firm by investment and sales promotion.
In such conditions, to say that wage increases " cause " higher prices
is stating a half-truth. Is a car accident on a slippery road " caused " by
rain, by the driver's failure to change his worn tyres or to slow down on
the curve, or perhaps by the road department's failure to provide a better
road surface ? Clearly all these factors played a causal part. Similarly, the
causation of a price increase may well be divided equally between a
union's successful wage claim and management's decision to maintain or
increase profit margins. A usual trade union argument for wage claims
is that some of the income now going to profits should go to the employees
as higher wages. If efforts to redistribute this income are frustrated by
management price policies, some people may argue that the unions should
stop asking for higher wages. But it is at least as logical to propose that
1

LEVINSON, op. cit., p. 9.

PRICES, WAGES, AND INCOMES POLICIES

managements stop adding constant profit mark-ups to the figures supplied
by their cost-accounting department.
In short, it would be wrong to hold either wage or price fixing solely
responsible for inflation. All that it seems possible to say is that present
methods of wage and price determination are parts of a mechanism
which is making it very difficult to avoid inflation.
However, if it is true that in industrial countries prices and wages are
as vulnerable to upward pressures and as insensitive to downward pressures as is suggested by the above analysis, and if there are so many
ways of starting up this inflationary mechanism, it is somewhat surprising
that the increases were as moderate as they have been. There must evidently
have been some forces protecting the economic system from runaway
inflation. Among them mention may be made of delays in price and wage
adjustments, rising productivity, and the role of stocks and imports.
Though prices of manufactured goods are normally adjusted to cost
increases, the adjustment is not instantaneous. Management may wait
for a psychologically suitable moment; contracts may not permit adjustment of prices before the date of expiry.
As regards wages, while in most countries payment above usual or
contractual rates may be made at any time to individuals or certain
groups of workers, general and formal adjustment of wages to changes in
the cost of living, profits or some other factor cannot normally be made
before collective agreements expire and, even when sliding scales exist,
adjustments are made only at certain time intervals and for changes
above a certain minimum in the cost-of-living index. Collective agreements are usually made for not less than a year, and even when, as in
Belgium and the United Kingdom, they are not negotiated for any particular period, most workers do not have their wages adjusted more frequently than once a year; in some important cases, because of the absence
of formal dates of expiry and lengthy negotiation and arbitration procedures, the period between successive wage adjustments may be rather
longer than a year. Another point is that it takes time for price increases
of, for example, raw materials to reach the consumer price level; in the
interval wage claims cannot be based on increases in the cost of living.1
While there are therefore delays in the passing on of price and wage
increases, productivity can rise in the meantime, reducing some costs and
offsetting other cost increases. It is doubtful whether small gradual
increases in productivity are of much help in this respect. Such improvements may simply accrue to piece workers whose output and bonuses
1
It has been estimated that the period of " throughput " of British imports varies
from two to 14 months, with an average of about eight months : J. C. R. Dow in Oxford
Economic Papers, Vol. 8, No. 3, Sep. 1956, p. 265.

SOME MECHANICS OF INFLATION

increase as a result of better quality of materials, slight improvements in
working methods, and so on. But when productivity rises considerably,
owing to the introduction of new technology or to complete reorganisation of working methods, piece rates will normally be revised and so will
standard costs, on the basis of which prices are calculated. In such cases
rising productivity may well lead to falling prices.
The effects of fluctuations in demand on prices and wages may also
be dampened by variations in stocks. Price stabilisation schemes for raw
materials may involve buifer stocks from which increases in demand can
be met to some extent. Most manufacturers keep stocks of materials, parts
and finished products ; and, of course, one of the main functions of wholesalers and retailers is that they keep stocks of various kinds.
In many cases increases in demand may be met by temporary imports
of both goods and labour. Imports, or the threat of imports, may also
reduce cost pressures ; at the international level, price competition is much
keener than it appears to be within countries, so that producers exposed to
international competition cannot afford to add wage and other cost
increases to their prices as easily as those not exposed to it.
While these factors all tend to limit the inflationary effects of the
modern wage- and price-fixing mechanism, their combined effectiveness
may be somewhat less now than it would have been before the Second
World War. The forces making for inflation may be stronger than before
for several reasons. First, anti-inflationary policies during and after the
war, based on assumptions that in the situation then prevailing price
increases might be limited but could not be prevented, may have been
habit-forming in certain respects. Where price controls were applied,
these were often on a cost-plus basis. When wage increases were generally
forbidden or discouraged, exceptions would be made because of increases
in the cost of living and for the purpose of correcting " manifest inequities
and anomalies ". Thus it may now be considered normal that prices and
wages should rise in these conditions. Secondly, against a background of
high levels of employment in some countries, collective bargaining has
become much more important than it was before the war (for example in
the Netherlands and the United States) or has replaced entirely different
ways of wage fixing (as in the Federal Republic of Germany, Italy and
Japan). More generally, there has been more explicit discussion and policy
making in terms of relative incomes as between wages, profits and farmers'
incomes than was usual before the war, thus encouraging people to
believe that it is normal or just that increases in some incomes should be
followed by similar increases in other incomes. Thirdly, the fact that
almost uninterrupted prosperity has made it possible to maintain costincreasing price- and wage-fixing practices with impunity for about a

PRICES, WAGES, AND INCOMES POLICIES

generation may have made these practices appear to be so normal that it
would be very difficult to change them through relatively small and acceptable reductions in demand and employment levels.
Three kinds of spirals have found a place in the literature of inflation—
first, the familiar wage/price spiral; second, a wage/other incomes spiral;
and third, a wage/wage spiral.
If it is true that a long run of inflation has made wages more sensitive
to price changes (at least upwards), this would obviously make a wage/
price spiral more likely. Other sorts of spiral, however, would take some
time to get going.
The lag before higher rates of pay respond to a relative worsening by
inflation has been especially marked where salaries are concerned.
Salaries are traditionally more stable than wage rates, and, where
salaries are covered by collective agreements, these often run for
longer terms than do wage contracts. The same sort of lag applies to
the appearance of a " whipsaw " relationship between movements in
wages and other sorts of income. Many non-wage incomes are more rigid
than wages. Many contractual payments are or have been tied to fairly
long terms—like some kinds of professional fees, rents and interest. Other
payments of this sort have been subject to legal or institutional controls :
house rents in many countries have for a large part of the period been
regulated by law, and in several cases dividend payments by companies
have been restricted or discouraged by tax discrimination. Thus some
years would elapse, in any case, before incomes of this kind would follow
a general and persistent upward movement of wages and prices. And in
the meantime, wages and profits would gain somewhat at the expense
of such relatively fixed incomes. However, sooner or later a move to
readjust these incomes gets under way, and since this amounts to an
attempt to recover that share of the total national income which may
have been gained by wages during the lag, the result is likely to be a
spiral between wage and non-wage incomes.
For the purposes of this study, the apparently increasing rigidity of
the structure of wage relativities is of particular interest. When wages are
adjusted according to procedures of economy-wide or industry-wide
collective bargaining, or when in a formally decentralised bargaining
system groups of collective agreements move together following certain
" wage leaders ", or when the maintenance of traditional relativities is a
major criterion in sectional wage bargaining, any wage increase that may
be justified in the case of a particular group of workers tends to lead to
very widespread wage increases.
A tendency for the structure of relative wage rates to become more
rigid may be encouraged by certain other effects of prolonged high

SOME MECHANICS OF INFLATION

employment. For instance, the raising of wages, in a number of industries
which were before the war severely depressed, to a level close to that in
other industries may make it more difficult for wages to vary separately
in the two groups thereafter. In fact, a tendency for wage rates in general
to draw rather closer together seems to have been a very widespread
accompaniment to high employment and a long run of general wage
increases, and the cohesiveness of wage structures as a whole has certainly
increased in consequence.
As regards the pattern of occupational wage differentials, in the first
decade of the modern inflation there was a general tendency for lower
wages to be raised proportionately more than higher ones. Certainly
there was a very widespread levelling-up of less skilled and lower-paid
workers' wage rates under the austere conditions of wartime, and in some
cases this trend appears to have been prolonged into the post-war period.
The early stages of this process might have been approved, or at least
accepted, by better-paid groups. Indeed, it is likely that for some while
the effect on relative wage differentials was not substantial enough to have
appeared very significant. But as the process continued a reaction set in
among better-paid workers and in the past few years there have been
many demands for preferential wage or salary increases to offset the
narrowing of differentials.1 To the extent that these demands have been
successful, they have in turn often set off further demands from the now
better-organised lower-paid workers for wage increases to keep them in
line. There is also some evidence in certain situations of a piece-rate/timerate spiral. Increases in piece-rate earnings very commonly occur because
of an accumulation of improvements in products, materials, methods or
proficiencies which are too small individually to justify downward revision of the piece or bonus rates themselves, and this has often led time
workers to demand an increase in the basic wage rates on which their own
earnings largely depend. Piece workers, however, may still feel it to be
unfair if the same increase in basic rates is not extended to them, with the
result that the time workers' lag is only partially made up in relative
terms and is perpetuated in absolute terms. Alternatively, it may be
difficult to compensate time workers for the tendency of their earnings
to lag behind those of piece workers without granting them increases from
time to time which are so large that they provoke consequential demands
for revision of piece rates themselves. In either case, the situation which
led to the first demand from time workers continually repeats itself.

1
For instance, salaries in the United Kingdom and the United States have been
catching up with relative wage movements from about the middle of the 1950s.

CHAPTER V

WAGE POLICY-METHODS AND INSTITUTIONS
In this and in the next chapter some account is given of attempts
in various countries to influence wage adjustments with a view to safeguarding economic stability in the sense given to this term in Chapter I—
relatively steady and uninterrupted growth at a high level of employment and with a fairly stable price level for the economy as a whole. The
present chapter is concerned with methods and procedures that have been
tried or proposed. The following one deals with the criteria according to
which attempts have been made to steer wage and price developments.

ANALYSIS AND INFORMATION

The beginning of any policy is the notion of what it is intended to
achieve. In this sense an incomes policy requires that " the authorities
have a view about the kind of evolution of incomes which is consistent
with their economic objectives, and in particular with price stability ". 1
In the countries with which this study is concerned, for such views to
be effective it is essential, in a matter so important as incomes and involving such basic features of the free enterprise system as private price and
wage fixing, that these views be widely supported by public opinion.
In some cases concern about the effect of price and wage adjustments on inflation has not, therefore, immediately given rise to any
policy properly speaking, but to measures for improving public understanding of the issues involved. For instance, in the United Kingdom in
August 1957, a Council on Prices, Productivity and Incomes (under the
chairmanship of Lord Cohen), consisting of independent experts, was
appointed to —
keep under review changes in prices, productivity and the level of incomes,
and to report thereon from time to time, having regard to the desirability of full
employment and increasing standards of life, based on expanding production
and reasonable stability of prices.
1
Organisation for Economic Co-operation and Development: Policies for Price
Stability (Paris, 1962), para. 19.

WAGE POLICY—METHODS AND INSTITUTIONS

This " Cohen Council " was found to have insufficient influence and
was replaced a few years later by a National Incomes Commission with
more specific and detailed terms of reference. These enabled it to pronounce publicly, when the Government had asked it to do so, on the
merits of particular pay settlements that had been arrived at by collective
bargaining. In making its comments the Commission was required to
take account of certain general guides, including the " desirabiUty of
keeping the rate of increase of aggregate money incomes within the longterm rate of increase of national production ". It was also required to
report from time to time whether there was a need for restraint of aggregate profits to match any restraint of wage incomes, and had the power
to undertake studies related to incomes policy. The refusal of the trade
unions to co-operate with the Commission, as well as the exemption of
arbitration awards from its scope, prevented it from assuming any more
direct role in wage determination. But it provided a mechanism by which
general principles and problems of incomes policy might be subject to
expert post-mortem discussion in relation to key wage and salary settlements, and the effect of the latter assessed with a view to influencing
future wage negotiations.
Early in 1965 the National Incomes Commission was replaced by a
new National Board for Prices and Incomes. A Declaration of Intent,
signed in December 1964 by representatives of both the Trades Union
Congress (T.U.C.) and the national employers' organisations expressed
the agreement of these bodies to co-operate in a " sustained attack on the
obstacles to efficiency " and in giving effective shape to machinery which
the Government proposed to set up to " keep under review the general
movement of prices and of money incomes of all kinds " and to advise
in particular cases " whether or not the behaviour of prices or . . . money
incomes is in the national interest...". The general reviews of price and
incomes movements are carried out by the National Economic Development Council, an existing tripartite body which has its own full-time
office and expert staff and has already published reports both on national
economic growth in general and on particular aspects of that question.
The norms for incomes increases, together with criteria for exceptions to
them, are thereafter handed on to the Prices and Incomes Board by the
Government, which also retains responsibility for references of particular
cases, whether of prices or incomes, to the Board for inquiry. In the
case of wages and salaries the Board investigates not merely settlements
which have actually occurred but claims for increases or other adjustments in terms of employment which have been presented. In the case
of the previous National Incomes Commission it had been found that
examination of particular wage settlements sometimes took many

PRICES, WAGES, AND INCOMES POLICIES

months, and the Government has said that reports of the new Board
should be available within two or three months of reference, or " if
possible even more quickly in the case of claims ". However, " in the
case of both prices and incomes, persuasion and the pressure of public
opinion will be relied upon to ensure that the findings and recommendations of the Board are accepted by the parties concerned ", and
the Government has stated that it would resort to other methods only if
convinced that the voluntary method had failed.1
In 1963 a Council of Experts more similar to the former British
Cohen Council was set up in the Federal Republic of Germany. These
experts are to review from time to time the over-all economic situation
and its foreseeable development, and to examine how a stable price level,
high employment and external balance could be achieved simultaneously
with an appropriate and steady rate of growth. This examination should
include the generation and distribution of incomes and wealth, but after
a lively discussion of this point it was explicitly decided that the Council
was not to make recommendations with regard to specific measures of
economic and social policy.
GOVERNMENT EXHORTATION

A more direct attempt to influence wage and price fixing according
to certain principles occurs when governments publicly appeal to trade
unions and industry (in some cases also to statutory wage-fixing bodies)
for moderation in wage claims and price adjustments, either in general
or in more or less specific terms. Public authorities have issued frequent
warnings against inflation and stressed the responsibilities of industry
and organised labour towards the national community. Statements in
this sense have been made ad hoc on the occasion of particularly important wage negotiations or pricing decisions, as in the United States steel
industry. But in other cases they have become periodical events, as in the
Annual Economic Reports of the President of the United States and of
his Council of Economic Advisers, and in budget messages and government economic surveys in many other countries.
If such exhortation succeeds in winning public opinion over to the
government's view of what are appropriate principles of wage and price
adjustment, in spite of its informality this may become a powerful
weapon. A determined government may then further publicly expose
cases in which its warnings went unheeded, with a good possibility that
1
See United Kingdom: Machinery of Prices and Incomes Policy and Prices and
Incomes Policy, Cmnd. 2577 and 2639 (London, 1965).

WAGE POLICY—METHODS AND INSTITUTIONS

any further pronouncements it might make would be taken more seriously by those concerned. Government appeals may further be strengthened by an indication of steps that would be taken if wage and price
fixers disregarded official views—import duties might be suspended,
taxes raised, credit restricted, and so on. For instance, in January 1964
the President of the United States, after declaring that he saw no warrant
for either inflationary price increases or inflationary wage increases,
stated —
- I shall keep a close watch on price and wage developments, with the aid
of an early warning system which is being set up in the appropriate agencies.
- I shall not hesitate to draw public attention to major actions by either
business or labor that flout the public interest in non-inflationary price and
wage standards.
- And I shall translate into action the view—
a. that antitrust policy must remain keenly alert to illegal price fixing and
other practices that impair competition;
b. that we must resist new steps to legalise price fixing where competition
should prevail.1
Beginning in 1959 the French Government has from time to time
informed employers' organisations of its views regarding the limits
which collectively negotiated wage increases should not exceed, and these
organisations seem to have proved willing to take account of the Government's opinions. This particular instance drew attention to what in
several countries now appears to be a major difficulty in implementing
any kind of policy aimed at guiding collective wage bargaining, namely
the limited control of central organisations over their affiliates. For it
seems that the French Government's statements had no appreciable
effect on wage increases actually conceded by individual employers.2
As pointed out recently by one of the French trade union confederations,
a major difficulty of applying an incomes policy in that country is that
collective wage agreements bear but a weak relation to the actual wage
level.3
CENTRAL CONSULTATION

Opinion as to what kind of evolution of incomes is consistent with
a government's broad objectives, and with other features of foreseeable
1

United States: Economic Report of the President (Washington, 1964), p. 11.
François SELLIER in Droit social (Paris), June 1960, pp. 317 ff. Of course, government control of wage increases for workers in nationalised industries and of subsidies to
farmers and other groups has not proved notably easier to achieve than that of private
wage fixing.
8
Note by the Confédération française des travailleurs chrétiens, in annexes to
MASSÉ, op. cit. (La Documentation française, Recueils et monographies, No. 48),
p. 30.
s

PRICES, WAGES, AND INCOMES POLICIES

economic and social development, depends on information as to what
that development is likely to be. There is an obvious advantage in associating the two sides of industry with the process of forecasting, particularly if, through such an association, those responsible for key decisions
on price and wage fixing can be brought to concert their views about
what is likely to happen in the period ahead, and encouraged to adapt
their policies to this prospective development.
Some years ago a proposal for regular tripartite discussions of the
state and prospects of the national economy was put forward in the
United States.1 It was suggested that, after the Economic Report of the
President had been transmitted to Congress, the Secretary of Labor
should convene a conference with leading representatives of labour and
management. Some top-ranking government representatives would give
their detailed views on economic prospects in the short and in the long
run. These views would then be discussed and the management and
labour participants could contribute their own views and knowledge.
The purposes of these discussions would be—
to develop a consensus of opinion, in so far as possible, or to narrow the
range of views concerning the major problems confronting the economy as a
whole and the expectations of the short-term and longer-term business outlook
by principal sectors. These discussions would not be negotiations nor should
they be designed to prejudice any particular contract negotiations -1
The Government would, however, give imaginative leadership to the
discussion. It would do more than caution and preach. In sketching the
economic problems and climate, on the basis of good statistical and
economic information, it would promote " a more direct exchange of
ideas and information in a society of free men and free collective bargaining ". An application of these ideas may be seen in the convening of
annual White House economic conferences (the first one was in April
1962) at which wages and prices in the United States have been discussed
at length. No resolutions are passed by these conferences, but they perform very useful advisory and information functions. Moreover, the
President's Labor-Management Advisory Committee provides a
standing body through which an exchange of viewpoints on price and wage
policy can take place between government and industrial representatives.
More recently somewhat similar arrangements have been developed
in other countries, although the demarcation between non-committing
discussion and the negotiation of some kind of agreement cannot always
be clearly drawn.
1
John T. DUNLOP in The American Assembly, Columbia University: Wages,
Prices, Profits and Productivity (New York, 1959), pp. 148-149.

WAGE POLICY—METHODS AND INSTITUTIONS

In Denmark an Economic Council was established in 1962, comprising a committee of three independent " wise men " and a further 20
members representing government, employers' and trade union organisations, consumers and others. The Council, which is to meet at least
twice a year, is to review national economic conditions and co-ordinate
the use of various instruments of economic policy. The committee members have a considerably stronger position than the other members of the
Council, because they have to prepare reports to the Government and to
decide to what extent these should be published. The committee's first
report, prompted by a serious balance-of-payments situation, was the
basis for a comprehensive programme of price and wage stabilisation
launched in March 1963 and it is expected that much of the future work
of the Council will also be concerned with questions of incomes policy.
In the Netherlands the Social and Economic Council (S.E.R.)
(comprising independent members appointed by the Government as
well as employers' and trade union representatives) was associated on
several occasions with decisions of national wage policy after its constitution in 1950. But with the introduction of a new system of wage policy
as from 1963 the Council has been given a more formal position in this
context. It is required to follow up the publication of forecasts by the
Government Central Planning Bureau with half-yearly reviews of national
economic conditions. While these reports are not to be limited to problems of wage and price adjustment, they are to serve as the basis for
consultations between the Government and the central employers' and
trade union organisations in the Foundation of Labour with regard to
wage negotiations.
Assessment of economic prospects jointly by government experts and
labour and management representatives, together with voluntary adaptation of public and private investment plans and other projects to this
assessment, with a view to avoiding inconsistencies, characterises postwar planning in France. It might seem surprising that so far (i.e. over
the period 1947 to 1965) the Plans have not included any incomes policy.
The explanation appears to be in part that the trade union movement,
divided along political lines, is largely opposed to any kind of wages policy
which would involve an element of wage restraint calculated to avoid
inflation. Indeed, trade union participation in the Plans has been rather
lukewarm, while the employers also, although participating quite actively,
have not shown any enthusiasm for extending them so as to cover wage
policy.1 From the point of view of the logic of planning, the non-inclusion
of incomes is, however, a serious shortcoming, for the threat of cost
1

See J. DESSAU in British Journal of Industrial Relations, Oct. 1963.

PRICES, WAGES, AND INCOMES POLICIES

inflation to growth (examined in Chapter II above) also interferes with
the implementation of plans. When balance-of-payments difficulties
arise and measures for curbing demand have to be taken, investment,
social welfare and other targets will fail to be achieved.
In these conditions the Incomes Conference, which lasted from
October 1963 to January 1964 resulted in relatively limited, though
by no means insignificant, proposals by its chairman, the Commissaire
général au Plan, M. Massé.1 In particular, they excluded any arrangements
for committing the parties concerned to any particular policy, even by
way of the informal undertakings that have been used in respect of the
investment aspects of French Plans.
However, according to these proposals, the official planning agency
(Commissariat général du Plan) would include in its preparation of the
draft Five-Year Plans indications regarding the future evolution of broad
groups of prices and incomes that would be consistent with both the
production and the social welfare aspects of these Plans. Secondly, in
connection with the annual Economic Budget and after consulting the
Economic and Social Council (which includes representatives of labour
and management) the Government would recommend, in precise and
rather detailed terms, rates of increase in various types of income for the
year ahead. Thirdly, a group of independent experts would be appointed
to form a board for the study and appraisal of incomes, to examine
and indicate in public reports, on the basis of agreements and other
factual material referred to it, the extent to which the Government's
recommendations were actually being followed. This third proposal was
inspired by the example of the National Incomes Commission established
in Britain in 1962.
Bodies representing central employers' and workers' organisations
(with or without other groups) and thus in principle capable of economywide discussion of incomes questions also exist in Belgium, Italy, the
Scandinavian countries, the United Kingdom and the United States.
In the United Kingdom the National Economic Development Council
has stated that it regards tlie solution of the problems involved in implementing a policy for prices and money incomes as a necessary condition
for fulfilling its task.2
In December 1964, as noted above, the British Minister for Economic
Affairs succeeded in getting management and trade unions to sign a
Joint Statement of Intent on Productivity, Prices and Incomes. In return
1

P. MASSÉ, op. cit., pp. 26 ff.

United Kingdom: Conditions Favourable to Faster Growth (London, 1963), paras.
200 ff.

WAGE POLICY—METHODS AND INSTITUTIONS

for wage restraint, management agrees to government fiscal action
to cut down net profits if there is any "excessive growth in aggregate
profits as compared with the growth of total wages and salaries,
after allowing for short-term fluctuations ". 1
The second stage of the negotiations, passed in February 1965, was
the setting up of the machinery required to keep a watch both over
undesirable price increases and over wage increases. The third stage will
be to fix quantitative norms forjudging whether income increases should
be recognised as excessive.
The National Board for Prices and Incomes will depend in the
first instance on voluntary co-operation for the enforcement of its judgments ; it is however envisaged that " the Government would have to
consider giving the Board statutory authority . . . if experience showed
this was necessary." 2 The extent to which the new Board is successful
in practice will depend, of course, on its ability to secure the confidence
of the trade union movement and to persuade employers that it has the
will and can develop sufficient speed to examine wage claims before
they become settlements. For it is clearly much more likely that the Board,
at least while the voluntary system is in operation, will be better able to
influence a firm to roll back a price increase of which it disapproves than
to suggest that wage increases, once granted, should be taken away. The
essence of success, then, under this system is speed of deliberation and of
pronouncement, attended by the greatest possible publicity.
CENTRAL WAGE NEGOTIATION

In some cases joint consideration of wage and price problems by
central organisations of labour and management, with or without government participation, has gone a step further than the exchange of views
and information, and has led to understandings or commitments on the
part of the groups concerned to promote or ensure application of certain
principles of price and wage adjustment by their constituents.
Particularly highly developed and systematic forms of central wage
negotiation exist in Denmark, Norway and Sweden. While there are
differences between the industrial relations systems of these countries
(for example trade union organisation is by skill and craft in Denmark,
by industry in Norway and Sweden; there are also important differences
in the arrangements for the settlement of disputes) they have in common
the central negotiation by employers and trade union federations of
1
2

United Kingdom: Joint Statement of Intent on Productivity, Prices and Incomes.
Idem: Machinery of Prices and Incomes Policy, op. cit., p. 3

100

PRICES, WAGES, AND INCOMES POLICIES

annual or, more commonly, two-yearly and occasionally three-yearly
contracts. These cover primarily manual workers, but salaried employees
have sometimes been directly involved in the negotiations and in any
case the central wage agreements have had a considerable impact on
other incomes, including farmers'. The practice of central wage negotiations has been possible because, in contrast to all other countries, in
Scandinavia there is a high degree of employers' and workers' organisation as well as of control by the central federations over their affiliates.
Although central wage negotiation is older in Denmark and Norway,
it is useful to consider here the case of Sweden, which is entirely free of
any compulsory state intervention and has also been the object of quite
penetrating analysis.1
The central organisations in Sweden are the Confederation of Trade
Unions (L.O.) and the Swedish Employers' Confederation (S.A.F.).
Among the salaried and professional workers the Central Organisation
of Salaried Employees (T.C.O.) is the most important, and it has occasionally participated in the central wage negotiations.
Both L.O. and S.A.F. have a strong interest in central wage bargaining
and in both cases this derives from certain features of modern wage fixing
noted in Chapter IV above, particularly the tendency towards rigidity in
the wage structure. When different rates of increase in productivity,
profits or demand for labour lead to different rates of wage increase in
various activities, this conflicts with the traditional and (as in the other
Scandinavian trade union movements) deeply rooted policy of wage
solidarity which requires L.O. to make special efforts on behalf of lowerpaid workers. Indeed, this policy, first stated at the L.O. Congress of 1922,
was a major reason why in 1941 the member unions agreed to increase
considerably the powers of the central secretariat. Central bargaining is a
means of moderating the development of large inter-industry wage differentials.
On the employers' side, the fact that, because of this solidarity,
wage increases in industries with a large capacity to pay generate strong
pressures for similar increases in activities with much less capacity to pay,
makes it important for the pattern-setting negotiations to be considered
also in the light of their repercussions on other sectors. Indeed, the first
post-war master agreement in 1952 was concluded (after a chaotic bout
1

In Denmark, mediation by state arbitrators and, if this fails, ad hoc legislation to
settle disputes between the central organisations, have played an important part: for
instance in 1956 and again at the 1963 " package deal " referred to in the next chapter.
In Norway the scope of compulsory arbitration was considerably reduced in 1949 and
it was altogether abolished in 1952. But emergency procedures for compulsory arbitration have been applied in later years, for instance in 1958 and again under a special
decision of Parliament in April-May 1964.

WAGE POLICY—METHODS AND INSTITUTIONS

101

of wage increases in 1951) at the initiative of the employers, who made its
acceptance a condition for conceding the cost-of-living clause demanded
at the time by L.O. The next central contract, in 1956, was also still
regarded as an ad hoc measure, after unco-ordinated negotiations in 1955
had given rise to unexpectedly large wage increases, and a major industrial conflict had only just been avoided. But in fact it proved the beginning of an uninterrupted series of master agreements, which henceforth
were to be two-yearly, except that for 1959.
Formally, these agreements have been undertakings on the part of
L.O. and S.A.F. to recommend certain wage adjustments in the contracts
to be negotiated at industry level by their affiliates (the industry contracts
in turn are the basis for negotiations at the plant level). In practice this
means that no smaller increases will be conceded than those centrally
agreed, and that there can be no strikes or lockouts over larger increases
—in fact, agreement on such larger increases is quite common. It is further
the practice for the master agreement to specify increases in average
hourly earnings in the bargaining units (industry orfirm).This determines
increases in total wage payments in each of these units. This is the " cost
framework ", within which are to be accommodated the various kinds of
payment and the various groups of workers by the bargaining partners
themselves.1 Separate figures are given for average earnings of piece
and time workers and it is specified what statistics are to be used in case
the parties cannot agree on the figures applying to their own bargaining
units. Various other matters, such as improvements in fringe benefits,
may also be laid down in the master agreements.
Since the master agreements are recognised to have a major impact
on the economy as a whole, both parties take account of what they
regard as " the scope for wage increases " during the period ahead. Actual
negotiations are preceded by a series of public statements, meetings and
informal consultations, in the course of which both sides form and express
their opinion, while the Government may seek to influence their appraisal
of the economic situation and thus their approach to the negotiations.
Account is also taken by the parties of the extent to which actual wage
developments have deviated from contractual wage increases and may be
likely to do so in the period ahead. These problems of " wage drift " are
further discussed in the next chapter.
Austrian employers are corhpulsorily organised in Provincial Chambers of Commerce and Industry, with a Federal Chamber as the central
body. Workers are organised in 16 trade unions (five of which represent
1
See the text of the Central Agreement for 1960-61 reproduced in T. L. JOHNSTON:
Collective Bargaining in Sweden (London, Allen and Unwin, 1962), p. 349. Similar
principles were embodied in the agreements for 1962-63 and 1964-65.

102

PRICES, WAGES, AND INCOMES POLICIES

salaried employees) affiliated to the Austrian Trade Union Federation.1
Private collective wage agreements are usually negotiated per industry per
province (Bundesland) between the Provincial Chamber of Commerce
and Industry and the trade union concerned.
Throughout the post-war period the Government has consulted and
negotiated with the two central organisations on problems of economic
and social policy. As a result, during the difficult inflation-ridden period
from 1947 to 1951,fivegeneral price and wage agreements were negotiated
as part of comprehensive stabilisation efforts. By the end of 1951
inflation was under control and a period of stable growth set in.
But in 1957 cost inflation was believed to prevail. Wages and prices
were rising although investment was falling. It was thought that monetary
control could put an end to the price increases, but at the cost of stifling
growth. The Federal Chamber and the Trade Union Federation, together
with the central organisation of agricultural producers and the Congress
of Chambers of Workers 2 then reached an agreement with the Government for keeping wages and prices as stable as possible. In particular, proposals for price and wage increases were to be submitted for advice to a
Joint Wage and Price Commission, representing those four organisations
and the government ministries concerned with economic problems. The head
of Government (the Federal Chancellor) became chairman of this new body.
Initially the Commission, whose recommendations have to be unanimous, had no power of applying any sanctions at all. Such authority as it
had was purely moral. But after it had had some initial success in limiting,
or at least delaying, increases in individual prices and wages, inflation
reappeared and it was felt that not only was the Commission becoming
too lenient (in fact it did not have any specific guiding principles for judging wage claims) but that it was also being increasingly ignored.
As a result of an agreement late in 1961 between the leaders of the
Federal Chamber and the Trade Union Federation the competence of the
Commission was extended in two directions. First, it can now in certain
cases recommend thefixingof maximum prices by the Government under
the existing price control legislation (but any such price controls cannot
last longer than six months). Secondly, the Commission's advisory task
is no longer limited to individual price and wage increases but extends to
over-all problems of economic policy, including the formulation of
general principles of wage and price adjustment.
1
See Anton PROKSCH: " The Austrian Joint Wage and Price Council", in International Labour Review, Vol. LXXXIII, No. 3, Mar. 1961, pp. 229 ff.
2
Wage and salary earners are compulsory members of this last organisation, the
main function of which is to promote consumer interests; for instance it has arranged
for retail price surveys of household goods as a contribution to price control.

WAGE POLICY—METHODS AND INSTITUTIONS

103

An Advisory Council (Beirat) for economic and social problems was
established in October 1963. Composed of representatives of the four
economic groupings mentioned above, this body, with the help of the
Central Bank, the Central Statistical Office, and economic experts, is to
make recommendations to the Commission on an objective and factual
basis. Thefirstmajor contribution of the Advisory Council was the preparation of a stabilisation programme in March 1964.
This development corresponds entirely with the aspirations of the
Trade Union Federation, which from the start had looked upon the
Commission as a step towards the creation of a body through which the
workers' organisations would have a say in the formulation of national
economic policy.1 To the Federal Chamber the advantages of joint
consultation and negotiation over problems of wages and economic policy
also seemed attractive enough to lead it to accept the extension of the
Commission's terms of reference. The new arrangement was, however,
criticised on the ground that questions of national interest were to be
decided to a large extent in a body representing mere group interests.2
But, of course, in this respect the new terms of reference of the Austrian
Commission do not really break new ground. In several European countries, organised group interests now exert considerable influence on
government policy through official channels (for instance through social
and economic councils) with a view to supplementing traditional parliamentary systems. Such formal and public channels of communication and
pressure have been found both more useful and less devious than informal
and more obscure forms of lobbying.3
On the. occasion of the reformulation of the tasks of the Commission
in 1962/the idea was put forward in employers' circles of arranging for
the negotiation between the Federal Chamber and the Trade Union
Federation of basic collective agreements rather similar to the Swedish
master agreements! to serve as guidelines for collective wage bargaining
at lower levels. "No effect seems to have been given to this suggestion. But
it appears that in Austria, as in other countries, actual wages have in
recent years considerably exceeded negotiated wages.
CENTRAL GOVERNMENT CONTROL

All governments are directly concerned with thefixingof wages and
salaries for their own employees. In most countries these are quite a
1

PROKSCH, op. cit., p.

244.

See, for example, Gertrud NEUHAUSER in Wirtschaftspolitische Blätter, No. 6/1962,
pp. 468 if.
3
Economic Policy in Our Time, op. cit., Vol. 1, pp. 174 ff.

104

PRICES, WAGES, AND INCOMES POLICIES

large group; for example in the United States one out of about every six
members of the labour force works for federal, state or local government.
In a number of countries governments also control private minimum
wages, whether through fixing a national minimum wage as in France and
the United States, or through a set of statutory industry minimum rates
(as in Belgium and the United Kingdom). In emergencies most or all private wages and salaries are placed under government control. Authoritarian
methods of doing this were applied during the Second World War in
Germany, Italy and the countries occupied by them. During the period
of early post-war recovery in some of the latter (including Belgium,
France and Norway) emergency legislation provided for more democratic methods.
In the United States wage controls were introduced with the Economic Stabilisation Act of 1942, the tripartite National War Labor Board
being made the implementing agency. In 1951, during the Korean war,
the tripartite Wage Stabilisation Board became the main body for administering temporary wage controls. In the Netherlands alone the system
set up under an emergency decree of 1945 outlived the emergency and
was maintained for some considerable time in more or less normal peacetime conditions.
Although there is a large variety of employers' and workers' organisations in the Netherlands, the distinction is mainly between Christian (and
further between Catholic and Protestant) and religiously neutral organisations. There are no basic political divisions within either of the two sides.
And since the organisation of workers is very largely by industrial unions,
few demarcation disputes arise of the type that easily occur with craft
union organisation. Thus it has in general been possible to achieve a high
degree of concerted action among both the employers and the trade union
federations. Furthermore, during the Nazi occupation close contacts between workers' and employers' leaders prepared the ground for post-war
consultation between the two sides, in the celebrated Foundation of
Labour which also became one of the main advisory bodies in matters of
government policy.
The very serious war damage called for many direct government
controls, including controls over wages, with the participation of the
Foundation. But while after a few years many other government controls
were gradually abolished as the economy recovered, wage control has
been maintained. The reason has been that control over the wage level is
considered a necessary instrument of policy for an economy which is
exceptionally dependent on foreign trade and in which a high rate of
increase in the labour force was long believed to raise difficult problems
of maintaining full employment. Furthermore, the trade unions expected

WAGE POLICY—METHODS AND INSTITUTIONS

105

that a system of central wage determination could improve the distribution of the national income.
Until 1959 wage policy in the Netherlands was implemented through a
system in which the Government controlled all the three main phases of
which an effective wage policy consists. First, it determined from time to
time the guiding principles of wage adjustment, after consultation with the
Foundation of Labour and, since 1950, with the Social and Economic
Council. The guiding principles were determined on the basis of an
appraisal of national economic conditions by the government Central
Planning Bureau. Second, the translation of these guiding principles into
actual wage adjustments was made through collective negotiation, but
all collective agreements had to be validated by the Board of State Conciliators, who acted on the basis of instructions from the Government
and also fixed wages and other terms of employment where collective
bargaining did not exist or did not lead to agreement.* Third, actual wage
payments were (and still are) verified by a government wage inspection
service ; infringements of the official wage regulations, if repeated after
warnings, lead to prosecution. It may further be noted that the crucial
guiding principles were used to serve two ends. In thefirstplace they were,
of course, to ensure that the general wage level was consistent with the
basic objectives of full employment, price stability, and external balance.
In the second place they were to bring about a structure of wage differentials which reflected differences in job conditions and requirements, but
not in profits, productivity increases, or the demand for labour.
After economic recovery had been achieved around 1952, pressures
began to develop for both a reduction in government control and a new
approach to the guiding principles that would permit wage differentials to
take account of differences in the economic conditions of firms and industries. But not until thefirstpost-war Government without participation by
the Labour Party had been formed in 1959 did these pressures lead to
action. In particular, wage increases could henceforth be based on a
measure of productivity increases in individual industries. This principle,
however, led to two kinds of difficulties.
First, it turned out that productivity measurement for individual
industries was not amenable to precise administrative control. Wage
increases were negotiated, and had to be approved, on the basis of sometimes very dubious statistical exercises by the parties concerned. Second,
since for technological reasons productivity rises much faster in some
activities than in others, but also since large differences in wage increases
1
Draft collective agreements reached the Board after screening by the wage commission of the Foundation.

106

PRICES, WAGES, AND INCOMES POLICIES

are neither acceptable politically nor justified economically, the new
principle not only aroused discontent but also had to be qualified by
several complex additional rules to avoid excessive wage differentiation.
Since these rules were issued by the Government itself, those bodies
(especially the Christian trade union and employers'organisations) who
wanted a lessening of government control became dissatisfied. At about
the same time the neutral Netherlands Federation of Trade Unions
(N.V.V.), which had been the staunchest advocate of the original system,
began to take a different view, partly because on balance experience was
thought to have been disappointing in terms of income distribution.
Furthermore, labour shortages became so acute as to make any kind of
wage control increasingly difficult to enforce, while wage restraint in the
Netherlands was increasingly embarrassed by the higher wages in neighbouring countries, to which a certain movement of labour could occur.
In these conditions S.E.R., asked by the Government for advice regarding the guiding principles, instead recommended a drastic change in the
system itself, which would involve a considerable reduction in government control.1 Most of the recommendations were accepted by the
Government and the new procedure was introduced on 1 January 1963.
As we have noted above 2 S.E.R. reports on the national economic
situation twice a year—in February and September—on the basis of an
analysis by the Central Planning Bureau. Normally, the September report
is to be the basis of consultations between the Government and the Foundation for determining the scope for wage increases in the year ahead.
Unless irreconciliable divergencies of opinion prove to exist between
these two, it is for the various federations on both sides to consult their
own affiliates with a view to matching the need for maintaining an overall balance with the requirements and conditions of individual industries.
In fact this " internal co-ordination " consists of determining how the total
scope for wage increases cculd be distributed among the individual industries. It is normally to be preceded by consultations between the employers'
and trade union federations, and followed by the actual industry negotiations. The results of these negotiations are to be validated by the
Foundation. At the same time the employers are to report to the Foundation
any effects that the proposed wage agreement would have on their prices.
Thus, normally, the Government, and especially the Board of Conciliators, should play but a limited formal role in the new system. But in case
of conflicts, both are to be given quite extensive powers. For instance, if
the Government and the Foundation disagree about the conclusions to be
1
2

Advies inzake net Systeem van Loonvorming (The Hague, 6 July 1962).
See p. 97.

WAGE POLICY—METHODS AND INSTITUTIONS

107

drawn from the S.E.R. report, the Government may decree a wage freeze
or, as under the old system, issue guiding principles for implementation
by the Board. At the same time the questions at issue between the
Government and the Foundation would be referred to S.E.R. for advice.
Furthermore, the Board may notify the Foundation that the approval
of a certain proposed agreement would lead it to advise the Government to
quash the contract as inconsistent with the central agreement. Thirdly,
if the Foundation itself is unable to reach agreement on a proposed
contract, the latter will be referred to the Board for validation. Similarly
the Board may impose a regulation in case of conflict between the Foundation and an individual industry.
The crucial innovation in the new system is the replacement of the
Board by the Foundation as the normal agency for approving collective
agreements. This means that proposals submitted by affiliated organisations now have to be approved by the central federations that constitute
the Foundation. The moral authority of the federations replaces the
formal legal powers of the Board. Doubts as to whether this would be
possible made the neutral Central Employers' Federation dissent from
the 1962 report of the S.E.R. But the first major difficulty encountered
with the new system was of a slightly different nature. In the second
half of 1963, at a time of serious labour shortage, actual wages rose further and further above official wages, thus undermining the position of
the employers' and workers' organisations that had negotiated the official rates and appeared to have been more impressed by the Government's wishes than attentive to the interests of their members. In these
conditions a government decision to quash some contracts approved by
the Foundation but involving fairly large wage increases was badly
received. Furthermore, two large employers publicly announced that in
order to stay in business they had to pay " black wages ", official wages
being inadequate for them to retain their workers. In spite of these indications that the official wage level had become untenable, the September
report of the Central Planning Bureau calculated that there was only a
very modest scope for wage increases in 1964. Not surprisingly, the
Bureau's predictions were ignored. After difficult negotiations by-passing
S.E.R., the parties represented in the Foundation agreed that increases
roughly twice as large as those recommended by the Planning Bureau
were in order—and the Government had little choice but to accept this
finding.
CENTRAL ARBITRATION

Where economy-wide wage bargaining exists, disputes may call for
economy-wide arbitration. For instance in Denmark a State Arbitrator

108

PRICES, WAGES, AND INCOMES POLICIES

may submit proposals for the settlement of disputes between the central
organisations. Such proposals, when rejected by the parties, have on
several occasions been made enforceable by special legislation. But in
this and other cases (for instance in Norway) negotiation is the normal,
and arbitration the exceptional, procedure of wage fixing. In Australia
and New Zealand the reverse is true; in these countries arbitration itself
has developed into a method of pursuing a national wage policy.
In Australia the Commonwealth Conciliation and Arbitration Commission may, in order to prevent or to settle an industrial dispute extending beyond the limits of any one state in the country, make an award
fixing " the basic wage ".1 This is the lowest wage that may be paid to an
unskilled adult worker within the Commission's jurisdiction, no matter
to what industry or occupation he belongs (there are separate male and
female basic wage rates). Furthermore, " margins " may be fixed by the
Commission—minimum amounts awarded above the basic wage for
specific occupations, among which that offitterin the metal trades plays a
central part in that many other skill differentials move with it to take
account of requirements of skill, and of the arduousness of the work.
The basic wage plus any margin constitute the minimum wage for the
occupation concerned.
Conciliation and arbitration are predominantly compulsory, and in
practice awards of the Commission have become the normal method of
wage determination for workers within the Commission's jurisdiction
(about 40 per cent, of the country's wage-earning labour force). From
time to time the Commission undertakes an inquiry that may result in an
award changing the basic wage. Many of the statutory wage-fixing bodies
in individual states tend to follow the Commission's awards (prior to
1953 several of them were legally required to do so), so that the actual
impact of the awards is very wide. Furthermore, although the formal purposes of the system are to avoid industrial unrest and to dispense justice,
the awards are made with explicit reference to the country's economic
and social conditions. The indicators used for assessing these conditions
include, for example, data relating to employment, investment, productivity and foreign trade. In this sense the system is one of a national wage
policy.
From its beginnings early in this century until 1956 the arbitrating
agency was a legal court, and even the Commission can make awards
1
Information on the legislation governing, and a survey of principal awards made
under, this system are published regularly in Commonwealth Bureau of Census and
Statistics : Labour Report (Canberra). For recent developments see " Wage Determination in Australia: Basic Wage and Total Wage Inquiries, 1964", in International
Labour Review, Vol. 92, No. 2, Aug. 1965, pp. 128 ff.

WAGE POLICY—METHODS AND INSTITUTIONS

109

only when in presidential session; that is, awards can be made only by
those members who have been solicitors or barristers of certain high
courts, or judges of the earlier Arbitration Court. The proceedings are
similar to those of a court of law. Parties are represented by counsel;
economic information is submitted as " evidence " by " witnesses " who
may be cross-examined. The Government has, of course, no power to give
the Commission any instructions about the nature of the award it should
make, though the Attorney-General may appear before it on behalf of the
public interest (and did so, for example, in the 1960 basic wage case).
Strikes about matters settled in an award may be treated and punished
as contempt of court and a trade union responsible for an illegal strike
may be fined and deregistered, which bars it from access to the Commission. In the course of its long history the system has developed a voluminous jurisprudence which, in fact, comprises a number of principles of
national wage policy to which further reference is made below.
In New Zealand a registered trade union may compel, or be compelled
by, the employers of its members to accept action under the Industrial
Conciliation and Arbitration Act,firstenacted in 1894.1 As a result, either
party to a dispute may be obliged to accept conciliation and arbitration,
but a large proportion of the disputes are in fact settled at the conciliation stage.
The Court of Arbitration has, however, an important additional
power—it may, from time to time, make general wage orders by which
rates of remuneration fixed in all awards and industrial agreements in
operation may be changed. The court may do this of its own accord or on
the application of a trade union or employers' association; such an
application is not subject to compulsory conciliation procedures. This
power does not rest on the Industrial Conciliation and Arbitration Act
but was first introduced under wartime emergency legislation, and continued under the Economic Stabilisation Act of 1948. The orders affect
wages throughout the economy and must be based on a number of
national economic and social indicators that are specified in the Act.
So this system, too, is a form of national wage policy in the sense in
which that term is used in this study.
DIFFERENCES OF FORM AND SUBSTANCE

Leaving aside the somewhat indeterminate forms of exhortation and
consultation, national wage policies have thus been carried out through
1
See Sir Arthur TYNDALL: " The New Zealand System of Industrial Conciliation
and Arbitration ", in International Labour Review, Vol. LXXXII, No. 2, Aug. 1960,
pp. 138-162.

110

PRICES, WAGES, AND INCOMES POLICIES

bargaining between private bodies, control by the government, and
arbitration by agencies having powers comparable to those of courts of
law. But though these may seem sharply differing kinds of arrangements,
in actual practice the distinctions are rather less clearcut.
On the one hand there is Scandinavia where, though the Swedish
Government is not a party to any negotiations between L.O. and S.A.F.,
it is not wholly without influence on the process of wage determination.
It may feel impelled to state its views on wages (as in January 1955,
when the Minister of Finance issued a strong warning against an overall wage increase exceeding 3 per cent.) and these may well be heeded.1
The crisis of 1955 was averted after the Government had convened a special conciliation commission. On earlier occasions government policy
and private negotiation clearly supplemented each other: for example in
1948 a government appeal for wage restraint was accompanied by
measures to limit dividends and to increase profits taxes, and in 1950
increasing subsidies helped to facilitate wage restraint.
In Norway the Government's influence is somewhat greater on otherwise private wage negotiations rather similar to the Swedish system.
There is a close association between the Norwegian L.O. and the social
democratic party—the trade union movement expects a government
controlled by that party to be sympathetic to the workers and tends to
take account of such a government's problems. On some occasions, as
during the settlements of 1956 and 1963, the Government was heavily
involved through its agricultural price and subsidy policies, and in 1958
and 1964 it had to refer a major dispute to a public wage board for arbitration.
In Denmark, finally, highly centralised collective bargaining between
the Danish L.O. and the employers' federation (D.A.F.) is supplemented
by arrangements for compulsory mediation by the State Arbitrator,
whose proposals, if rejected by either side, may be made law by Parliament.
On the other hand, in certain countries where the government has
power to control wages directly, the actual implementation of policy
may be largely through processes involving a high degree of negotiation
with private groups. As already noted, the United States National War
Labor Board and the Wage Stabilisation Board were both tripartite
bodies. The latter, for example, comprised three (later six) representatives
of each of three groups—the trade unions., the employers, and the public.
1
Indeed, in 1952, the Government deliberately refrained from estimating private
consumption in its economic survey on the ground that such estimates " are readily
considered as an authoritative pronouncement of an economically correct development
of wages ".

WAGE POLICY—METHODS AND INSTITUTIONS

111

The Board, incidentally, could also be asked by the President to recommend terms for the settlement of disputes threatening to interrupt defence
production; but its power to make binding awards was limited to disputes
where both parties agreed to ask for such an award.
Attention has also been drawn above to the part played by employers
and trade union organisations in the system of government wage control
in the Netherlands. Similarly, though statutory minimum wage fixing
in the United Kingdom is of course formally an act of government, enforced by the state-maintained wages inspectorate and by the possibility
of prosecution and conviction in court, there is a tradition of self-determination in the wages council system which stems from the unwillingness of
Parliament to give the Minister of Labour the power to determine wages.
An outstanding feature of the British system of statutory wage regulation
has always been the restriction of the Minister's power over wages to a limited
right of veto. A Wages Council decides, by voting if necessary, what legal minimum wages to propose; the Minister has no power to alter these proposals. He
can either ratify them or refer them back to the Council with his comments for
their reconsideration. In practice the Minister very seldom refers back any
Council's proposals— On almost all occasions the Ministerfindsthat the same
proposals come forward after his " reference back " and, as a rule, he then
accepts them.1
Finally, the awards of the Australian Commonwealth Arbitration
Commission (the purest instance of wage policy by central compulsory
arbitration) are not made in the abstract but are influenced on the one
hand by evidence submitted on behalf of trade unions, employers and the
Government and on the other hand—presumably—by the Commission's estimate of what government fiscal and monetary action, or
what degree of industrial unrest, is likely to follow from certain awards.

1
F. J. BAYLISS: "British Wages Councils and Full Employment", in International
Labour Review, Vol. LXXX, No. 5, Nov. 1959, p. 425.

CHAPTER VI
CRITERIA OF WAGE ADJUSTMENT IN
NATIONAL WAGE POLICY
When an incomes policy includes a national wage policy, two types
of decisions have to be taken. First, the permissible or desirable increase
in total wage payments has to be determined. Second, the distribution of
this totalincrease among various groups of wage earners must be decided.
Not all attempts at central guidance of wage adjustments have been
based on explicit and consistent rules for settling these two problems.
Many of these attempts have been ad hoc measures to deal with specific
emergencies (such as an acute balance-of-payments crisis), but when
wage policies have been pursued systematically, certain fairly common
criteria have emerged. Some of these relate directly to the purposes that
the wage policy is to serve—i.e. mainly the avoidance of inflation. Other
criteria have been adopted in order to make the wage policy acceptable
to those affected by it: for instance, the adjustment of money wages to
increases in the cost of living. Both types of criteria are briefly discussed
below.
ADJUSTMENT OF THE OVER-ALL WAGE LEVEL

To the extent that the purpose of incomes policies is to avoid price
increases, the implication seems clear: the rate of increase of total
money incomes (i.e. costs including profit margins) must be kept within
the rate of increase of national output. It does not necessarily follow
that every main category of income (wages, salaries, professional incomes,
profits, dividends) should rise at the same rate. This would mean that
the current distribution of the national income between these categories
and, as a result, many of the existing inequalities in personal incomes,
would be maintained; and there is no reason to think that the existing
distribution is necessarily fair in any meaningful sense, or even economically efficient.
Nevertheless, for several reasons it is often considered appropriate
in the short run for incomes policies to aim at approximately equal
rates of increase for various broad income categories, particularly wages

CRITERIA OF WAGE ADJUSTMENT IN WAGE POLICY

113

and profits. In the first place, even if it is desired to bring about a change
in the distribution of incomes between these broad categories, incomes
policy may not be considered the best instrument for this purpose. In
the short run governments may prefer to use measures of fiscal policy—
for instance more effective death duties, taxation of wealth and capital
gains, and less generous treatment of expense accounts. While some
changes could certainly be made also in the primary (pre-tax) distribution of incomes, the results that could be achieved in the short run
would probably be fairly small, while the cost of achieving them might
be very heavy. For instance, strict price controls according to accurate
cost-accounting methods might well be capable of reducing profits, but
the administrative costs would be very large.1 In the longer run, emphasis
may have to be given to efforts to bring about changes in certain important aspects of a country's economic structure that have a major bearing
on the existing income distribution. These include particularly inequalities
in educational opportunities and in skills, which are an important source
of income inequalities.
The Productivity Criterion
In those cases, then, in which the existing primary income distribution is broadly accepted2, the appropriate rate of increase in the wage
level has often been defined as the rate of increase in over-all productivity.
For instance, the Council of Economic Advisers in the United States
formulated in its 1962 report, and reiterated in following years, a " guidepost " for wage adjustment as follows :
The general guide for non-inflationary wage behavior is that the rate of
increase in wage rates (including fringe benefits) in each industry be equal to
the trend rate of over-all productivity increase. General acceptance of this
guide would maintain stability of labor cost per unit of output for the economy
as a whole—though not, of course, for individual industries.
Similarly the Government of the United Kingdom in 1962 declared
that—
In recent years national production per head has risen by about 2 to 21/2 per
cent, a year. We ought to be able to do better than this but on present trends it
seems likely to increase at about this rate in 1962. It is accordingly necessary
1
It should, however, not be assumed that such a scheme would necessarily cause
the pricing system to function less satisfactorily than without price control. As noted
in Ch. Ill, modern price-fixing practices in private business may differ considerably
from theoretical models of the market economy; for instance, prices may not be
reduced when costs have fallen.
2
Equity in the distribution of income is further considered in the next chapter.

114

PRICES, WAGES, AND INCOMES POLICIES

that the increase of wages and salaries, 1as of other incomes, should be kept
within thisfigureduring the next phase.
Again, the Australian Commonwealth Arbitration Commission considered in its basic wage inquiry of 1961 that, apart from annual adjustments to changes in the cost of living, the basic wage should reflect
" a review of the economy generally and, in particular, productivity
increases " which it decided to undertake every three or four years (thus
abandoning its practice since 1956 of making annual reviews).
Mention may also be made of some private agreements for the
systematic adjustment of wages to the rate of productivity growth in the
economy as a whole, though they are outside the framework of a national
incomes policy. In the United States as early as May 1948 a provision
of this nature was included in a two-year agreement (followed by a fiveyear contract in 1950 and by three-year agreements in 1955, 1958 and
1961) between the General Motors Corporation and the United Automobile Workers. A few years later similar provisions began to be made also
in several other important collective contracts. In addition to automatic
adjustments to changes in the cost of living, these long-term agreements
provide for an " improvement factor ", i.e. annual wage increases agreed
in advance. The purpose of this arrangement is to make wages rise in
concord with productivity without annual negotiations being necessary.
In the General Motors contract the size of the improvement factor has
deliberately been set at, and in most of the other agreements it in practice
roughly corresponds to, the estimated long-term annual increases in
productivity of the United States economy as a whole, and not of the
firm or industry concerned. Indeed, in the main industries concerned,
the rate of productivity increase has probably been rather higher than the
average, and larger than their improvement factor.2
Where the rate of increase in national productivity is accepted as the
basis for determining the appropriate rate of wage increases it is not always concluded that both should rise at the same rate. Thus, at one time
the French Government considered that, when over-all productivity rose
by about 5 per cent, a year, the wage level should increase by only about
4 per cent. " so as to leave room for investments or price reductions ".3
1
United Kingdom: Incomes Policy. The Next Step, Cmnd. 1626 (London, 1962),
para. 5.
2
Joseph W. GARBARINO: Wage Policy and Long-term Contracts (New York,
Brookings Institution, 1962), p. 106. Somewhat similar collective agreements have been
negotiated in France but there, for instance, in the contracts concluded in 1955 and
1958 for the Renault works, the improvement factor was based on expected productivity increases in the firm itself; see " Works Agreements of the ' Renault Type ' ", in
International Labour Review, Vol. LXXXI, No. 3, Mar. 1960, pp. 205-232.
3
Statement by the Prime Minister in the National Assembly (Journal Officiel de la
République française, Débats parlementaires, No. 60 A.N., 4 Oct. 1961, p. 2345).

CRITERIA OF WAGE ADJUSTMENT IN WAGE POLICY

115

In the United Kingdom the Council on Prices, Productivity and Incomes
recommended in 1959 that a particular increase in productivity expected
to take place at that time be used to bring about price reductions rather
than wage increases.1
Some Problems of the Productivity Criterion
When wages per worker (i.e. the general wage level) rise at the same
rate as output per worker (i.e. over-all labour productivity) total wages
increase at the same pace as total output. Also wage-cost per unit of
output remains constant, so that the over-all price level need not rise on
account of wage increases. If all other incomes taken together also rise
at the same rate as total output, the over-all price level can then be kept
constant. However, the mere fact that money wages rise faster than productivity does not imply that wages are a cause of inflation.2 When there
is excess demand, prices will tend to rise and so will money wages—the
wage increase will be an effect of inflation, not its cause.
The productivity criterion raises certain problems of measurement.
There are many concepts of labour productivity for the economy as a
whole. On the output side one may consider gross or net physical output
per man-hour; one may include or exclude production for purely military
purposes. On the input side, labour may be measured in terms of unweighted man-hours (which take no account of differences in quality
or skill), or in terms of weighted man-hours (that is, allowing for differences in individual earnings taken as an indicator of differences in
skill). The two methods may give quite different results. For example,
over the period 1889 to 1953 the United States gross physical " peacetime " output per weighted man-hour rose at an annual average rate of
1.6 per cent. But over the same period gross physical output (according
to the definition of the Department of Commerce) per unweighted manhour rose by 2.2 per cent, per annum—or nearly half as much again. In
France, over the period 1949-58, productivity rose at an annual rate of
4.5 per cent, if man-hours are weighted, but 4.8 per cent, if they are not.
Weighting the labour input by wages in calculating global productivity
indices means that, in adding up man-hours worked in the different
sectors of the economy, those in industries and occupations with high
average earnings are counted as more hours than the same number
worked in industries and occupations with a lower wage level. An hour
of better paid work is, as it were, regarded as a multiple of an hour of
1
United Kingdom, Council on Prices, Productivity and Incomes: Third Report
(London, 1959), p. 26.
2
See above, p . 78.

116

PRICES, WAGES, AND INCOMES POLICIES

low-paid work, on the assumption that, generally speaking, it is more
highly skilled, and therefore represents a larger input of labour. As a
result, when between two dates there is a shift in employment from lowwage to high-wage industries or occupations, total labour input is shown
as having risen, even though the number of hours actually worked did
not increase at all or increased less than the apparent rise in labour
input. It is because a weighted labour input index thus tends to rise
faster than an unweighted one that a productivity index using weights
rises less than an unweighted one. But this is appropriate for an index
used for the productivity criterion of wage policy because, while structural
shifts in manpower will normally result in a rise in the total output index,
there is no corresponding scope for wage increases, the shift of labour
to better-paying industries or occupations itself having already resulted
in a rise in the wage level. So raising wage rates in the same proportion
as a national productivity index calculated with unweighted labour input
would tend to lead to increases in unit labour costs. The point is of some
importance because in certain cases structural shifts in employment have
been quite substantial. For example it has been estimated that almost
one-third (7.9 per cent, out of a total of 26.5 per cent.) of the rise of
productivity in Austrian manufacturing between 1937 and 1956 was due
to such shifts.1
Productivity does not necessarily rise without any interruption. There
may be periods in which it actually falls. But it is now nearly inconceivable
that wage rates in industrialised countries would be reduced; in practice
it will thus be impossible to have a very precise adjustment of the general
wage level to short-term movements in a productivity index. Short-term
fluctuations in productivity are due mainly to short-term fluctuations
in the volume of output : for instance because employers " hoard "
labour in periods of low demand. So in principle it would be best to
base a wage policy on average rates of productivity growth over a period
long enough to comprise upward and downward movements in demand
and output. 2
1

J. STEINDL, quoted in Österreichisches Institut für Wirtschaftsforschung: Möglichkeiten und Grenzen einer produktivitätsorientierten Lohnpolitik (Vienna, 1960), p. 33.
Structural employment shifts in western Europe since 1953 are discussed in : United
Nations Economic Commission for Europe : Economic Survey of Europe in 1958
(Geneva, 1959), Ch. II, p. 31. In Japan, as a result of large shifts during the 1960s, productivity in the economy as a whole is expected to rise at a considerably faster rate than
in any of the main sectors—6.6 per cent, a year compared with 5.6 per cent, in primary
industry, and 5.5 per cent, in secondary and tertiary industry, transport and communications (New Long-Range Economic Plan of Japan 1961-70 (Tokyo, 1961), p. 25).
Output per man-hour is not the only productivity index that could be used for
purposes of national wage policy. Another index might be what has been called " global
factor productivity " (See SELLIER, op. cit., pp. 317 ff).
2
Möglichkeiten und Grenzen, op. cit., p. 50.

CRITERIA OF WAGE ADJUSTMENT IN WAGE POLICY

117

The Scope for Wage Increases
As a first approximation the productivity criterion is a solid basis
for determining appropriate adjustments of the general wage level when
the objective is to maintain over-all price stability. There is much in
favour of this criterion in any case because over long periods real wages
have in fact been rising at about the same pace as over-all productivity.1
It should be possible to continue this trend. But, as already noted, in
the short run it may sometimes be convenient or necessary to deviate
from the productivity rule. In a particular year, wages and other incomes
may have to rise at a lower rate because of balance-of-payments deficits,
or a strategic opportunity may occur for reducing prices rather than raising money incomes.2 Again, private incomes and consumption may have
to be limited in order that public consumption may rise faster. For
instance the 1962-65 long-term programme for Norway provided for a
rise between 1961 and 1965 in private consumption of 15 per cent., and
in public consumption of 21 per cent., education and research being
largerly responsible for the latter increase.3
Conversely, at certain times it may be necessary or convenient to
allow money incomes to rise faster than the rate of over-all productivity;
for instance because there is demand inflation or because of favourable
balances of payments. Thus, in 1954 and again in 1963 wages and prices
in the Netherlands were allowed to rise, partly with a view to their alignment with those in neighbouring countries.4 It is also conceivable that
public consumption may sometimes fall, so that private incomes and
consumption could rise faster than productivity. Disarmament would be
a case in point, and it has also been suggested that with rising incomes
the useful scope of compulsory social security schemes may become
more limited than it was at lower income levels. Or again, it may be a
policy objective to alter the distribution of income as between wages and
other categories. For instance, as long ago as 1937 the Australian Commonwealth Arbitration Court awarded " prosperity loadings " to increase
1
This has been one of the main management reasons for adopting the improvement factor together with provision for wage adjustment to changes in the cost of
living referred to above.
2
United Kingdom, Council on Prices, Productivity and Incomes: Third Report,
loc. cit., p. 26.
3
In such cases the necessary increase in public revenue may, of course, also be
obtained by higher tax rates, while primary incomes are still permitted to rise in accordance with the productivity criterion.
4
Alternatively, the currency could have been up-valued, and this was in fact
done in 1960 (when the Federal Republic of Germany also revalued its monetary
unit).

118

PRICES, WAGES, AND INCOMES POLICIES

the basic wage in accordance with the view that the latter should be
" the highest that the community can afford ".
For reasons such as these it has often been found convenient to
broaden the economic basis of wage adjustment so as to include several
other factors in addition to productivity growth. In Sweden and the
Netherlands the term "scope for wage increases" has been coined to
describe this broader approach.
The periodical reports of the S.E.R. in the Netherlands and the studies
and consultations preceding the negotiation of master agreements in
Sweden constitute procedures for determining what scope there is for wage
increases. The annual national economic budgets in Norway and the
type of reports issued in the United Kingdom by the former Council on
Prices, Productivity and Incomes also serve this purpose, and so would
the five-yearly and annual reviews that were proposed for France in the
Massé Report. 1 During most of the 1950s the Australian Commonwealth Arbitration Court and Commission applied a series of economic
indicators to determine the national community's capacity to pay a basic
wage (the indicators included statistics on foreign exchange reserves,
rural and non-rural production, investment and profits, employment, etc.).
Determination of the scope for wage increases consists basically in
comparing the resources expected to be available during the period
concerned with the various claims that will be made on them.2 The former
is a matter of economic forecasting comparable to the predictions that
should underlie any country's annual budget estimates. Determining
claims on the resources involves specification of economic and social
policy goals during the period (as regards, for instance, the level of
agricultural incomes, social security, defence and investment) and of the
measures to be taken for achieving them. The process of comparison is
greatly facilitated by the use of national accounting systems as in the
case, for instance, of the national economic budgets of the Netherlands
and Norway.
Since determining policy goals and instruments is a problem of political decision, the scope for wage increases clearly cannot be regarded as
an objectively given quantity, but must be determined by negotiation as
well as calculation. It may also be noted that increases in output (or
" resources ") due to rising employment and productivity do not become
available in a way such that the authors of an incomes policy could
freely allocate them among diiferent purposes, but always accrue in the
1

See above, p. 98.
For a more formal statement of the principles and problems involved see Bent
HANSEN: The Economic Theory of Fiscal Policy (London, 1958), pp. 330 ff. and 377 ff.
2

CRITERIA OF WAGE ADJUSTMENT IN WAGE POLICY

119

first instance to the owners or managers and workers of the undertakings
where they take place. Thus in practice there is but limited room for
manœuvre in determining and distributing the scope for wage increases.
Maintenance of Real Wages
Provision for adjustment of wages to changes in the cost of living
during the life of wage agreements has, of course, been common in
periods of rapidly rising prices. Even in periods of stable or slowly
rising price levels such provision has often been made in long-term wage
agreements (i.e. longer than a year), to guard against the risk of future
increases. Indeed this safeguard has often been a condition of trade
union acceptance of long-term wage agreements such as have become
common in Scandinavia and the United States.
Provision may be made for adjustments specified in advance ("escalator
clauses " or sliding scales—e.g. so many cents per hour for each
rise in the price index of so many points) or for the opening of negotiations to determine the size of the adjustment (" reopening clauses "—
e.g. wage rates to be reviewed when the price index rises above a certain
point). Escalator clauses are quite common in collective agreements in
Belgium and Denmark.1 Reopening clauses were included in the central
agreements of 1954, 1956 and 1958 in Norway and in the Swedish twoyear agreement for 1957 and 1958, but not in that for 1960 and 1961.
Since 1952 the French legal national minimum wage (S.M.I.G.)
has been linked to a retail price index.
Cost-of-living clauses have often been held responsible for the accentuation of inflation by maintaining a price/wage spiral. This criticism
assumes that wages would normally rise more slowly in the absence of
such clauses: but probably if no escalator or reopening clause were
negotiated, trade unions would usually not be willing to sign long-term
contracts. For instance, in the United Kingdom escalator clauses apply
to only a minority of workers and often effect only a partial compensation in earnings for changes in the cost of living, but collective agreements in that country are almost invariably of indefinite duration, so
that wage discussions can usually be reopened at any time.
However, very frequent adjustment has the disadvantage that many
accidental temporary price increases (for example those due to late or
bad harvests) tend to lead to permanent increases in costs and prices.3
1
Also in Italy; for a discussion of wage indexing in that country and in Belgium,
Luxembourg and France, see Jean Pierre GERN: L'Indexation des salaires (Neuchâtel,
la Baconnière, 1961).
2
The Problem of Rising Prices, op. cit., p. 52.

120

PRICES, WAGES, AND INCOMES POLICIES

Thus, until 1957 escalator clauses in Italian industry provided for adjustment every two months. In that year it was recognised that the clause
itself constituted an inflationary factor and the interval of adjustment
was lengthened to three months.
On the whole, escalator (and reopening) clauses tend to bring about,
as their net direct effect, increases in average wages rather smaller than
the percentage increase in the cost of living. One reason is that adjustments are made to wage rates and not to total earnings, which may be
considerably higher. Another reason is that cost-of-living allowances
have often been fixed as a standard casli sum for each specified rise in
the cost-of-living index, irrespective of the wage of the workers concerned. And where the allowance was fixed as a percentage increase
of existing wage rates, it was often a percentage of minimum rates. As
a result, while wage rates of the lower-paid workers have often increased
in the same proportion as the cost of living, those of more highly-paid
workers have increased relatively less. But, as noted in an earlier chapter,
the very fact that relative wage differences according to skill were reduced
in this way has on some occasions led to formal or informal adjustments
of the higher wages in order to restore these differences. For example,
in Italy again, the wage structure was distorted in the early post-war
years so seriously that in the period 1947-49 a few special inter-confederal
agreements were made to raise the wages of skilled workers again. In
other cases such readjustments were made in a less formal, but not less
real, manner by way of a wage drift 1 of irxreases conceded by individual
enterprises.
THE DISTRIBUTION OF WAGE INCREASES

After determining the permissible or desirable over-all increase in
wage payments, the persons responsible for a national wage policy must
decide how this total sum is to be distributed among various categories
of wages and workers, through negotiation or otherwise. This distribution can be used to increase, reduce, or eliminate existing inequalities, to
restore former ones or create new ones. It therefore raises the questions
of the functions and the appropriate size of various types of wage differentials.
The most important aspects of this problem are listed in the terms
of reference given to the National Incomes Commission in the United
Kingdom as regards the criteria by which it was required to judge
" matters of importance relating to incomes " in specific cases that were
referred to it, namely—
1

See below, Ch. VII.

CRITERIA OF WAGE ADJUSTMENT IN WAGE POLICY

121

(a) the desirability of paying a fair reward for the work concerned;
(b) the manpower needs of the case, taking account of regional differences in such needs, and the importance of securing efficient deployment of resources;
(c) practices in the service, industry or employment concerned in
pricing, profit margins, dividends, efficient use of manpower and
equipment, and organisation;
(d) repercussions that the settlement of the case may have in other employments. 1
Briefly, as far as wages are concerned, these criteria relate to (a) the
comparative criterion, or the nature of the work, (b) the demand for and
supply of different kinds of labour, (c) productivity, and (d) possible
repercussions elsewhere in the economy. All these aspects have regularly
presented themselves in proposals and projects regarding national wage
policies. The instruction to the Prices and Incomes Board in the United
Kingdom (which replaced the National Incomes Commission in 1965)
that it might approve exceptional pay increases " where there is general
recognition that existing wage and salary levels are too low to maintain
a reasonable standard of living " apparently adds another criterion to
this list. However, this may in practice be considered a special case of
the comparative criterion. Although there are many instances in which
national minimum wage systems have been based on a specified family
budget, representing what was considered to be a necessary minimum
standard of subsistence for a typical wage-earning family, there are in
fact no scientific or objective criteria by which such a budget may be
determined—as is demonstrated by the fact that wage-minima under
such systems vary substantially between countries. This variation appears
to have some relation to differences in average living standards in the
countries concerned, which illustrates the point that the criterion of
minimum needs is essentially a comparative one. It may, however, be
so in two senses. First, it may be felt that the lower wages (for instance
of unskilled workers at large) are generally too low in relation to the
average living standards of the community concerned. This amounts to a
belief that the distribution of income, and in particular the distribution
of wages and salaries, is too unequal, and that a general levelling-up of
lower wages ought to occur. There are, again, no objective standards
by which it may be determined whether or not this is so, provided that

1
United Kingdom, National Incomes Commission: Report No. 2, Report on the
Agreements of February-March 1963 ... Cmnd. 2098 (London, H.M.S.O., 1963).

122

PRICES, WAGES, AND INCOMES POLICIES

the demands for labour of different types are being met under the current
pay structure without particular shortages or surpluses. So that in this
case the wage-policy authority is really required to interpret and apply
predominant social sentiment on the matter. Secondly, however, it may
be held that certain wage and salary levels are too low to maintain a
standard of living which is " reasonable " when measured against that
of comparable groups of workers in the economy concerned. In this case
the criterion really represents an application of the " nature of the work "
criterion to the lower levels of the national pay structure.
The Comparative Criterion or the Nature of the Work
The oldest and most consistent application of the first criterion has
been the minimum wage structure of the Australian Commonwealth
arbitration system. As we have seen 1 this consists of a basic wage which
is uniform except for a sex differential, plus a structure of margins
defined as—
minimum amounts awarded above the basic wage to particular classifications of employees for the features attaching to their work which justify
payments above the basic wage, whether those features are the skill or experience required for the performance of that work, its particularly laborious
nature, or the disabilities attached to its performance.8
The principle of basing wage differences wholly or largely on differences in the nature of the job has also been characteristic of government
wage controls, for instance during the Second World War in Germany
and the United States, during the Korean war, again in the United States,
and, in peacetime conditions, until 1959 in the Netherlands. In this last
case the national wages policy was essentially based, in a manner somewhat similar to the Australian system, on uniform general wage rounds in
accordance with changes in the cost of living and with the scope for wage
increases, supplemented by such selective increases as could be demonstrated with the aid of a national standard system of job evaluation to be
necessary for correcting manifest anomalies in relative wages as compared
with relative job conditions.
The Scandinavian concept of wage solidarity already referred to 3
has also tended towards a " job evaluation point of view " with the nature
of job requirements and conditions as an important criterion in wage
adjustments.
1

See above, p. 108.
Australia, Commonwealth Court of Conciliation and Arbitration : Commonwealth Arbitration Reports, Vol. 80, 1954-55, p. 24.
3
See above, p. 100.
2

CRITERIA OF WAGE ADJUSTMENT IN WAGE POLICY

123

Outside the framework of a national wage policy, the comparative
criterion has also been applied in settling important wage and salary
claims. In the United Kingdom, for instance, in 1960, a committee of
inquiry chaired by Mr. C. W. Guillebaud made detailed comparisons,
for a number of jobs, of rates of pay and other conditions of employment as between British Railways and other nationalised industries, public services and private undertakings. It was established on this basis
that railway wages were about 8 per cent, and railway salaries about 10
per cent, below what they should have been if terms of employment were
to be roughly equivalent to those in comparable outside occupations, and
in fact wage and salary increases of this order were later negotiated.
The principle has an obvious appeal of fairness. Equal work in the
sense of equally demanding work tends to be rewarded equally, while
manifest differences in the skill and other job requirements (including
willingness to accept adverse working conditions) tend to be compensated
by différences in remuneration.
From the point of view of national wage policy, exclusive application
of this criterion as the basis for wage differences is quite attractive. This
is so because, compared with most other criteria for adjusting the wages
structure, it provides a reasonably clear and consistent, and a relatively
uncontroversial, standard for wage differentiation. As noted in earlier
chapters, wage relativities have tended to become rather rigid because
of a widespread tendency to maintain differentials for no better reason
than that they exist. The nature-of-the-work criterion provides a basis for
separating those existing wage differentials that seem rational from those
that do not. It should, however, be noted that job evaluation is not by
any means an objective—let alone a scientific—method of determining
appropriate wage relativities. Though it provides a method for systematic
study of relative job requirements and relative remuneration, it is based
on essentially subjective value judgments. Furthermore, relatively little
progress has been made in applying it to non-manual, and especially
professional, work.
The question is whether the criterion is sufficient to its purpose and,
especially, whether it does justice to the function of wage differentials
as a means of attracting labour to occupations, firms and industries in
which there is a shortage. A partial reply to this question would seem
to be that in most methods of appraising job requirements and conditions those aspects tend to be rated highest for which the necessary qualifications are scarcest, such as advanced skills, willingness to assume heavy
responsibilities or to accept unpleasant working conditions. Thus, there
should be at least a long-term correlation between categories of labour that
are in short supply and their relative remuneration. In order to cope also

124

PRICES, WAGES, AND INCOMES POLICIES

with short-term scarcities, the best approach is of course the direct one of
increasing the geographical and occupational mobility of labour by means
of an active labour market policy. However, some special allowance
could presumably also be made in wage policy for the existence of acute
shortages if there were good reasons to believe that widening the wage
differentials to encourage labour mobility could achieve what an active
labour market policy could not bring about. In present conditions, however, wage differentials often have but little effect in eliminating labour
shortages and surpluses, precisely because the pressures making for
rigidity in the wage structure tend to wipe out " n e w " wage differentials
in favour of certain categories of labour before these differentials can
have the desired result of attracting a sufficient number of workers to
jobs with a shortage of labour. 1 Moreover, labour mobility appears
insensitive to comparatively small differentials in wages, so it may often
be that the differentials required to bring about any substantial transfer
of workers would be too large to be acceptable on other grounds, such as
those of general equity or of cost.
Demand for and Supply of Different Types of Labour
However, it has often been felt that the wages structure should
reflect short-term partial shortages and surpluses of labour by shortterm variations in wage differentials. As early as 1951 a procedure was
suggested for doing this in the framework of a wage policy in conditions
of full employment.2 According to this scheme general wage adjustments
would be made fairly frequently on the basis of the estimated rate of
over-all productivity growth, money wages in individual occupations
being varied in accordance with " indices of relative attractiveness ".
Such indices would be calculated as the ratios between the number of
people qualified and ready to work in various occupations but unable
to get a job there and the number of people actually employed in these
occupations. The wage rates in relatively more attractive occupations
would then be reduced as compared with those for less attractive jobs
according to some formula for the distribution of the share in increasing
productivity going to labour, by raising the absolute rates less than those
for the less attractive occupations. In order to make this mechanism
work effectively every possible restriction or hindrance in the way of
movement from one labour market to another would, of course, have to
be eliminated.2
1

See I.L.O.: Job Evaluation, Studies and Reports, New Series, No. 56 (Geneva,
I960), pp. 58 ff.
2
A. P. LERNER: Economics of Employment (New York, McGraw-Hill, 1951),
pp. 213 ff.

CRITERIA OF WAGE ADJUSTMENT IN WAGE POLICY

125

The " guidepost " for non-inflationary wage increases proposed by
the United States Council of Economic Advisers resembles this suggestion. The productivity rule for over-all wage increases already quoted 1
is to be qualified by modifications for wage adjustments in individual
industries according to their manpower position (but also in order to
reflect the nature-of-the-work criterion) as follows :
(1) Wage-rate increases would exceed the general guide-rate in an industry
which would otherwise be unable to attract sufficient labor; or in which wage
rates are exceptionally low compared with the range of wages earned elsewhere
by similar labor, because the bargaining position of workers has been weak in
particular local labor markets.
(2) Wage-rate increases would fall short of the general guide-rate in an
industry which could not provide jobs for its entire labor force even in times of
generally full employment; or in which wage rates are exceptionally high
compared with the range of wages earned elsewhere by similar labor, because
the bargaining position of workers has been especially strong.8
In 1955 the Netherlands S.E.R. also suggested that inter-industrial
differences in the manpower situation be made a basis for differential
wage increases. However, no effect was given to this suggestion either
at that time or in 1959, when new guiding principles were introduced.
Productivity Growth in Particular Industries
As noted above, the productivity criterion is primarily of interest for
determining the size of increases in the over-all wage level. If wages in
individual firms or industries were raised in proportion to their respective
rates of productivity growth the wages structure would be grossly distorted. Nevertheless, suggestions have been made for reflecting partial productivity indicators in industry wage adjustments, at least to some extent.
Thus an " ideal productivity-based wage policy " has been suggested,
under which sectoral wage increases would be a weighted average of the
rate of increase in productivity for the economy as a whole and that in
the sector (e.g. industry) concerned. For example, if productivity in the
whole economy rose at an annual rate of 2.5 per cent., in agriculture by
4.5 per cent., and in manufacturing by 2.0 per cent., and if the weights
given to the general and the sectoral indices were 4 and 1 respectively,
agricultural wages would rise by 2.9 per cent, and industrial wages by
2.4 per cent.3
The justification given for this proposal is that in an industry in which
productivity rises fast, prices ought to fall (if over-all price stability is to
1

See above, p. 113.
United States : Economic Report of the President... 1962, together with the Annual
Report of the CouncilofEconomic Advisers (Washington, 1962), p. 189.
3
Möglichkeiten und Grenzen, op. cit. p. 39 ff.
2

126

PRICES, WAGES, AND INCOMES POLICIES

be maintained) so that demand and output will increase and the industry
will need additional labour. The opposite would be true of an industry
in which productivity rose slowly. The " ideal " formula, by making
wages in the former industry rise faster than in the latter, would facilitate
this movement of labour. But it may be objected that, if wage differences
are needed to make labour move, they are better fixed with the aid of
some direct indicator of labour surpluses and shortages. For example, in
recent years labour productivity in western Europe seems to have risen
faster in agriculture than in industry; but employment in agriculture, far
from increasing, has been falling rapidly. And there are several instances
of industries in which a sustained rise in productivity coincided, not with
prosperity and capacity to pay high wages, but with depressed conditions
(for example the textile industry in the United States). Indeed, the proposal seems to be open to the charge of distorting the wage structure,
even though it does so more slowly than if sectional wage increases were
made simply porportional to sectional productivity changes.
From the middle of 1959 a system of wage increases based on sectional
productivity growth was applied in the Netherlands. Industry wage rates
were allowed to rise by a percentage determined by the rise in labour
productivity indices for the industries concerned over a stipulated period.
But for obvious reasons this general rule had to be qualified in various
respects. On the one hand, in cases where productivity was demonstrated
to have risen very fast, wages would not be allowed to rise by the full
percentage, so as not to upset inter-industry wage differences too much,
and employers were encouraged to reduce prices. On the other hand,
in cases where productivity was found to have risen little or not at all
(as in road transport and the civil service) wages were allowed to rise
substantially.
In 1961 a formal rule for calculating wage increases in individual
industries was issued which was of the weighted-average type to which
reference has been made above. The permissible rate of increase was to be
calculated as the average annual percentage rate of productivity growth
in the industry concerned multiplied by three, with the addition of two
percentage points representing average annual percentage growth of
productivity in the economy as a whole, the total to be divided by four.
But, as mentioned above, this system was found to raise considerable
difficulties and it was soon abandoned.
Sectoral productivity changes may serve as a criterion for differential
wage adjustments in a different sense to that discussed above. It may be
accepted that differential rates of productivity growth are largely the
result of technological and economic circumstances, that they therefore
do not on the whole reflect differences in the contribution of employees

CRITERIA OF WAGE ADJUSTMENT IN WAGE POLICY

127

to production and so ought not in general to involve different rates of
pay increase. Exceptionally, however, a preferential pay rise may be
justified where the workers concerned have made some special contribution to productivity growth, or where such an increase is desirable in
order to induce workers to accept new methods or techniques. Thus
the circumstances under which the United Kingdom Prices and Incomes
Board may approve pay increases exceptional to the norm include—
where the employees concerned, for example by accepting more exacting
work or a major change in working practices, make a direct contribution
towards increasing productivity in the particular firm or industry.
This provision for productivity bargaining as part of wages policy
has obvious practical recommendations. However, it also has certain
disadvantages. For instance, the room for such concessions on the workers' part is obviously the greater the more inefficient the existing utilisation of labour, so that workers in sectors which already have a high
degree of efficiency in the use of labour, and who may themselves have
in the past accepted modifications in working practices to facilitate that,
can get only the norm of wage increases: in effect, the arrangement
puts a certain premium on present inefficiency. Where, moreover, this
inefficiency is the result of deliberate restrictive practices on the employees' part (such as an excessive limitation on the admission of
recruits to skilled jobs by the unions concerned), it is likely also to be
found that those who are best placed to make concessions, and thus gain
a preferential wage increase, are likely to be already relatively highly
paid. It may therefore become necessary at some stage to offer some
compensation to groups of employees who are unable to benefit from
this exceptional circumstance—such as a selective pay increase granted
them on the grounds that their pay has fallen behind that of other workers
in comparable jobs. And the instruction given to the Prices and
Incomes Board provides that, even in the exceptional case referred
to in the quotation above, " some of the benefit should accrue to the
community as a whole in the form of lower prices ", thus by implication
limiting the departure from the norm which may be approved under this
head.
Repercussions Elsewhere in the Economy
The last of the criteria we mentioned for judging wage settlements
on behalf of particular groups of workers concerns the repercussions
such settlements may have on wage movements for other groups. The
basic problem here is, of course, the increasing rigidity of the structure of
wage relativities mentioned in Chapters III and IV above. For instance,
wage increases negotiated or awarded by arbitrators in certain key indus-

128

PRICES, WAGES, AND INCOMES POLICIES

tries (the metal trades, or certain parts thereof, are examples of such
industries in several countries) tend to provoke wage increases of equal
size in other industries. While the former might not in themselves have
seriously threatened price stability, the combined increases of wage
leaders and followers could well do so.
This is, of course, a rather unsatisfactory standard for appraising
sectional wage adjustments. It leads to a negative view of settlements that
would be justified in their own right. The appropriate way of tackling the
problem would be to break up irrational links between various kinds of
wage changes. This is in fact done when it is decided to base wage increases
only on one or more of the criteria we have been considering—the nature
of the work, differences in the state of supply of and demand for different
kinds of labour, and sectional productivity increases. But as long as no
such criteria can be applied effectively (and the Netherlands'experience
shows that they may be discarded even after they have been accepted
for fairly long periods) the opportunistic consideration of anticipated
spreading of a particular wage increase will continue to be advanced
from time to time. Moreover, it must be remarked that the three criteria
referred to are by no means always readily compatible with each other.
In particular, a wage increase granted to one occupation on the grounds
that it is experiencing a special shortage of recruits or has made an unusual
contribution to productivity growth may give rise to apparently reasonable arguments from other occupations that the nature of their work is
closely comparable and that the same increase ought to be extended
to them.

CHAPTER VII

SOME PROBLEMS OF NATIONAL WAGE POLICY

This chapter briefly reviews some of the difficulties which practical
implementation of a national wage policy would raise. First, there is the
problem which emerges from the previous chapter, of determining the
criteria or guiding principles for the adjustment of wages, which involves
both the general wage level and the structure of wage differences. These
problems are discussed in the first two sections below, and it is pointed
out that they are likely to be much more intractable in an economy that
is growing only slowly than in one that is growing fast. Secondly, assuming that agreement can be reached on broad criteria, the question arises of
how actual wage changes at the plant level can be made to conform to
such centrally established criteria—this includes the problem of wage drift
as well as the question of the extent to which wage and salary adjustments
are in fact determined by collective bargaining. This leads to the question
of the organisational requirements of a national wage policy. Finally,
reference is made to the need, in addition to appropriate institutions and
procedures, for some basic agreement that there should be any wage
policy at all and about its broad contents. As far as trade unions are
concerned, such consensus clearly depends on the development of a
policy for restraining also incomes other than wages ; some aspects of the
latter question are briefly discussed later.1
EQUITY IN THE DISTRIBUTION OF INCOME

The criteria for adjusting over-all wage levels discussed in the previous
chapter are essentially short-term, in the sense that the existing structure of
income distribution is broadly accepted as the framework within which
wage adjustments may be made.
It may seem curious that this should be so. For the sectional wage
bargaining (necessarily concerned rather with sectional than with national
economic and social considerations) which national wage policy seeks to
correct is ostensibly concerned with redistributing the proceeds of the
1

See below, Ch. VIII.

130

PRICES, WAGES, AND INCOMES POLICIES

industries and firms concerned. If there is a problem, in conditions of
high employment, of reconciling decentralised wage bargaining with price
stability, it arises at least in part because trade unions do not accept the
existing income distribution. It may be true that trade union wage policies
have been much less effective in redistributing income than in provoking
price increases. But if for this reason the general interest required the
replacement of sectional bargaining by, or its subordination in some
degree to, central decision making, it would be reasonable to expect that
the new procedures should provide not only a remedy against inflation
but also a new approach to the problem of bringing about a more generally acceptable income distribution. Furthermore, it might seem that the
concept of the scope for wage increases is bound to raise the problem of
income distribution. For there is as much scope for wage increases as
remains available from the total income after provision has been made
for other income claims. So inevitably the question arises : which other
claims should precede, and which should come after, wage increases?
In many cases central decisions on wage policy have been taken ad
hoc, for the purpose of coping with emergencies of war and reconstruction, or with acute balance-of-payments problems. Such decisions might
be regarded as truces in sectional bargaining rather than as its replacement. There is then little need or scope for a basic policy in respect of
income distribution. However, in Australia, where central wage decisions
by arbitration have been regarded as a more or less permanent procedure,
the court did evolve a principle of distributive justice (albeit a vague one)
that the basic wage should be the highest that the capacity of the community as a whole can sustain. In the Netherlands, when in 1955 the
Social and Economic Council discussed the conditions of a long-term
wage policy, it concluded that such a policy ought to take account of
social justice—though the principle it formulated was much vaguer even
than that of the Australian court: " the wage level must be socially justified ". 1 And in France, although no specific principles were formulated,
the functions of incomes policy have been defined as including
" the realisation of social progress in the framework of economic
equilibrium ". 2
In general, however, such principles have mainly been used either to
support the raising of the lowest wages to a level thought to be more
consistent with current concepts of an adequate minimum standard or to
justify general wage increases exceeding the current growth of productivity in cases where it seemed that wages at large had fallen behind
1
Advies insake het Vraagstuk
Sep. 1955).
2

MASSÉ, op. cit.,

24.

van de Toekomstige

Loonpolitiek

(The Hague,

SOME PROBLEMS OF NATIONAL WAGE POLICY

131

the growth of other incomes. Otherwise, national wage policies in the industrial countries have been broadly neutral to the existing distribution of
income between labour and other revenues—as is indeed implicit in the
customary postulate that the current average percentage rate of productivity growth indicates also the appropriate proportion by which average
wages and salaries might, in normal circumstances, be increased.
While this avoidance of the problem of income distribution has probably been dictated by a desire to avoid prejudicing policies of economic
stabilisation by precipitating a social conflict, it may also be thought to
have a certain rationale. Historical experience suggests that increases in
wages generally do little more than secure to wage earners their preestablished share of the annual increase in the economy's product so that,
despite rising average earnings, the share of labour in total income tends
to remain relatively stable over long periods. Exceptions to this rule
occur when either the employers are unable to increase profits in the same
proportion as wages or there are other sections of the community whose
incomes cannot be rapidly increased in line with profits and wages
together. In the first of these cases, wage increases disproportionate to
productivity growth may lead to unemployment if they involve a reduction
in profit rates below the level at which private employers are willing to
maintain the existing volume of activity—and this may in any event be
the consequence if wages and profit margins together are involved in a
competitive spiral which ultimately compels the government to attempt
to restrain inflation by measures which halt or restrict economic growth.
So that in this case the increase in labour's relative share of the total product may be realised only at the cost of a decline in its absolute or real
share.1 In the second case the improvement in labour's relative position
may be achieved at the expense of groups whom the labour movement
itself might rather wish to protect—retired workers, independent artisans,
etc.
Historically, social measures such as the use of redistributive taxation
to finance services to lower-income groups appear to have been a much
more effective instrument of reallocating income in favour of wage earners
than have direct wage increases. Some trade union movements may have
felt it worth while to accept some limitation on their power to pursue
redistribution by wage claims alone—in effect, to accept substantially
the existing distribution of income before tax—in exchange for firmer
guarantees of the maintenance of employment and growth than would
1
" Some tend to reason that adapting the claims of various groups to the requirements of over-all balance is a sacrifice they would be asked to make. But this means
overlooking that an excessive growth of incomes is self-defeating." (MASSÉ, op. cit.,
p. 23).

132

PRICES, WAGES, AND INCOMES POLICIES

otherwise be possible, and for specific measures to improve the position
of workers and poorer people generally by social action. However, the
possibility of such an exchange will depend substantially on the attitude
of the government itself in such matters as the use of taxation for redistributive purposes and the social responsibility of the State, and this may
well imply that it is at least easier to achieve and maintain a national
incomes policy when the political complexions of the government and the
trade unions are broadly similar.
This suggestion is possibly reinforced by another consideration. If a
national wage policy does not in itself aim at a redistribution of income
in favour of labour, it certainly seems a condition of its effective continuation that the acceptance of limits or "targets" for the rise of wage incomes
does not lead (or appear to lead) to a disproportionate increase in profits
and property incomes. This involves questions of the possibility of controlling prices and profit margins, or of offsetting disproportionate
increases in profits and private wealth by fiscal or other measures; such
questions are in any case raised by the general problems of regulating
inflation, and are sufficiently complex to discuss as a separate issue.1 It
also involves questions of the possible participation of workers in profits
and investments, which are discussed later.2 But it emphasises too the
possibility that a national wage policy is more likely to be acceptable
if the character of the government gives the unions some confidence that
it is unlikely to be disproportionately considerate of property interests.
SECTIONAL WAGE INCREASES

The second problem of principle in a national wage policy concerns the
detailed structure of wages and salaries, and the priorities to be observed
in the allocation of wage increases between different groups of workers.
Except perhaps as a rather short-term expedient, such a policy seems
impracticable to operate by means of standard general wage advances
alone. In practice, some groups of wage earners will have to be given
some preference over others on grounds such as those mentioned in the
previous chapter with regard to the distribution of wage increases.
The fact that a realistic policy must involve some " discrimination "
between various types of wage claims raises at least three difficult problems. First, there must be some agreement regarding the grounds that
can, and those that cannot, be accepted as valid for preferential wage
claims (for instance the possibility that some groups were left behind in
1
2

See below, Ch. VIII.
See below, Ch. IX.

SOME PROBLEMS OF NATIONAL WAGE POLICY

133

previous wage movements, or that their wages do not reflect the skill and
other requirements of the occupations concerned, or that incentives are
needed to attract labour into certain jobs or to raise productivity, etc.).
Second, it must be decided what weight is to be given to such criteria as
may be accepted in principle. For instance, what above-average percentage increase in wages is appropriate in an occupation where unfilled
vacancies are found to add up to 5 per cent, of total possible employment? This leads to a third group of questions, concerning the measurement of indicators such as actual wage payments in various industries,
regions or occupations (should one use wage îates or earnings, straighttime hourly earnings or total weekly earnings ? how are various kinds of
fringe benefits to be reckoned?), labour shortage and surplus, changes
in productivity, and so on.
The settlement of these and similar questions is particularly difficult
because each group will advocate decisions favourable to its own sectional interest. And most of the time this interest will be to the effect that,
while wage increases everywhere else may well be restrained, in one's
own section they should not. For instance, in the Netherlands, when past
productivity growth was accepted as a criterion for sectional (industry)
wage increases, the seemingly factual measurement of this growth became
a negotiable issue instead—past productivity growth was what employers
and trade unions agreed they wanted it to have been. It seems quite
obvious that, for example, any concept of labour shortage as a criterion
for sectional wage increases would lend itself to similar manipulation.
It might be thought that the grading of wage claims (which often
appears as a conflict between equity and incentive) as a basis fcr policy is
not quite so important or difficult as might at first sight appear. For instance, if it were decided that wages in different industries should reflect to
some extent differing rates of productivity growth, it might not be necessary to lay down formal rules to this effect, because some such differentiation would almost inevitably occur in actual earnings through wage drift
(particularly through rising piece-work earnings). Similarly, short-run
changes in the demand and supply position of labour of particular kinds,
or in particular places and industries, will almost inevitably be to some
extent reflected in relative earnings (for instance through overtime). To
that extent, it would not be necessary for official increases in wage rates
to take account of the position of demand and supply. Of course, relying
on wage drift to solve problems that are too difficult for formal policy
decisions may mean that the policy itself is increasingly circumvented and
undermined. But it is at least possible that some forms of wage drift (and
certain kinds of fringe benefits, such as the provision of dwellings by
employers in order to attract labour) do not spread from one group of

134

PRICES, WAGES, AND INCOMES POLICIES

workers to others in the way in which formally negotiated wage increases
clearly do.
It seems in any case necessary to provide for a degree of flexibility in
the rules for preferential wage increases. For instance it should be possible
to deal with cases in which productivity growth may rise substantially
as a result of particular concessions on the workers' part, such as the
relaxation of established trade union practices and the readjustment of
workloads. Formal rules in terms of expected or promised productivity
increases would of course be even more difficult to lay down and enforce
than criteria based on actual past productivity growth. But such difficulties should not stand in the way of what has come to be known in Britain
and the United States as " productivity bargaining ", in which major
wage increases have been exchanged for very substantial workers' concessions.1
In general it seems inadvisable to expect too much consistency from
national wage policy. Where relative wages are concerned, any attempt at
central guidance does not start with a clean sheet but has to deal with a
structure established by past economic, social, and institutional factors
(including a large measure of mere accident) and still subject to pressures
of the same kind. Wage policy can therefore only attempt to adapt this
structure, by preferential wage advances or other methods, in a way
which will prevent its major aim (the achievement of a greater concordance between global income movements and productivity growth) from
being frustrated by conñicts and tensions between wage rates, while
encouraging such other trends in relative wages as will further that aim's
fulfilment.
THE NEED FOR PRODUCTIVITY GROWTH

The question of consistency becomes especially important when there
is not enough scope in the total wage increase available to satisfy the
requirements both of eliminating anomalies and of productivity bargaining—not to mention another requirement: that the average wage advance
must not merely allow for the rectification of manifest inequities 2 , and
1

For instance, in July 1960 the management of the Esso refinery near Southampton, England, conceded a series of wage increases of the order of 40 per cent, in return
for various changes in restrictive practices, including some relaxation of job demarcations, the redeployment of certain kinds of workers on other jobs, and increased
shift work (See Allan FLANDERS: The Fawley Productivity Agreements (London, Faber,
1964). For an example in the United States see Charles C. KILLINGSWORTH in Industrial
and Labor Relations Review (Ithaca, N.Y.), Vol. 15, No. 3, Apr. 1962, pp. 295 ff.
2
There may be limits to the egalitarian levelling-up of wages which will be tolerated because of its effect on traditional wage differentials, and this may be to some

SOME PROBLEMS OF NATIONAL WAGE POLICY

135

for special concessions where these will clearly induce an accelerated
productivity growth; it must also, if the policy is to command continued
support, give everybody (or nearly everybody) something, not too small
and at not too infrequent intervals. This suggests two things : first, that
it is unlikely to be effective if its requirements in terms of price stability are
so tightly set that, at the current rate of productivity growth, they involve
a pace of average wage increase too small to encompass these three
requirements; second, that incomes policy is likely to have a greater prospect of success if it is not considered primarily as a method of restraint
or restriction but rather as part of a general programme to accelerate economic and social advance (so that its chances of success may again be
influenced by such political circumstances as the government's broad attitude towards planning).
Moreover, the pace of average wage increase relevant in this context is
almost certainly the visible increase in wage rates, not the largely invisible
increase in earnings that goes by the name of wage drift. Thus when the
rate of productivity growth is only marginally higher than the rate of
wage drift there is virtually no room for an incomes policy aiming at
stability of prices ; the most that can be hoped for is some success in limiting the rate of increase in prices. The greater the rate of growth in relation
to the rate of wage drift, the wider the scope for an acceptable incomes
policy. Experience, whether of temporary central " wage-stop " agreements or of continuing attempts at central wage regulation, shows that
the limit of a temporary reduction in real wages seems unlikely to exceed
5 per cent, and the period of a wage freeze about one year, at least where
there are independent trade unions.
Thus, although it seems to be an advantage of co-ordinated wage
policies that their existence may help an economy to meet temporary
crises with less disturbance to employment and growth than would otherwise result, their use merely as an instrument of passing expediency—to
secure a temporary wage-stop—involves certain dangers. Experience
(for instance in the United Kingdom after the 1948-49 " restraint ", in
Sweden during the early 1950s, and to some degree also in the Netherlands
during the 1960s) suggests that a wage freeze or severe restraint may be
terminated, not by a gradual wage thaw but by an explosive " dam break "
in which wages rise very much faster than before the stop. The risks of

extent true also of the application of the " rate for the job " principle. But it seems
usually possible to placate dissatisfaction with the effects of levelling-up on higher-paid
workers by only partial concessions to the latter. Difficulties arise mainly when the
amount of economic room available for such concessions is limited because there is
pressure at the same time from lower-paid workers for wage increases to maintain their
standard of living.

136

PRICES, WAGES, AND INCOMES POLICIES

setting-off a new inflationary spiral would seem much reduced if the
temporary restraint represents only a deviation from a longer-term policy
to raise real wages, so that the distortions and inequities in wage differentials which may result from the freeze can be corrected by subsequent
adjustments in the distribution of wage increases, and any disproportionate sacrifice by the workers in general which may have resulted also
be compensated.
WAGE DRIFT

One of the major practical difficulties of wage policy is the extent to
which central wage decisions may fail to control the actual movement
of wages. The relation between nominal wage rates and actual wages
is a complicated one, of which there is insufficient study in most countries to permit very detailed analysis and which is the subject of much
specialist controversy where studies exist.1 But it has certainly involved
most difficult practical problems during attempts at a central wage policy
and there is at least some statistical evidence that attempts at central
regulation may exercise an only partial or even, in critical circumstances,
an inadequate effect. This evidence consists mainly in a tendency for
actual earnings to increase faster than is envisaged by central agreements
or decisions.
In this divergence there are several elements, the relative importance
of which in individual countries it is not easy to evaluate. Thus, changes
in the composition of the labour force—as between men and women or
adults and juveniles—may be of considerable importance in some statistical series of earnings. Average earnings naturally tend to rise faster
than average wage rates, because of the movement of workers to better
paid trades or districts and because of the general tendency for the
proportion of skilled workers in the labour force to increase. Such things
do not indicate any breakdown or ineffectiveness of central control, but
other elements in the disparity may be of a different order.2
Thus, one factor in a rise of average earnings in several countries is a
tendency, if not for actual working hours to increase, for the proportion of
overtime payment to grow larger. This conflicts with the usual inference
of collective agreements that overtime is of the nature of a temporary
expedient only, and there may well be particular cases where working
1

See H. PHELPS BROWN in Economica, Vol. XXIX, No 116, Nov. 1962, pp. 355 ff.

for a full list of references. Subsequent studies include those by Lars AARVIG in British
Journal of Industrial Relations, Vol. II, No 2, July 1964, pp. 182 ff. and by H. A. TURNER
in Manchester School, June 1964, pp. 155 ff.
2
See the discussion of this problem, in a slightly different context, on pp. 115-116,
above.

SOME PROBLEMS OF NATIONAL WAGE POLICY

137

hours themselves have risen to a degree undesirable for health and productivity. One reason for this trend could be that where wage rates are
controlled, employers may offer overtime working as an inducement to
labour recruitment, particularly for lower-paid occupations. In so far as
the opportunity for or the possibility of overtime working is necessarily
conditioned to some extent by the technical circumstances of different
occupations, this creates considerable inequalities in the earnings of
similar classes of worker. On the other hand, the possibility of manipulating overtime conditions presents employers with an opportunity of
evading at least the intention of central wage awards.
Another factor is the upgrading of individual operatives to work with
new equipment. This would be a normal consequence of technical development, but since the pace of technical change is necessarily uneven as
between different industries, earnings may rise independently of wage
rates, but at a different relative pace. However, there is the possibility
that the upgrading of workers may be used by management as a method
of increasing actual wages without formally infringing central agreements.
In some cases, however, the effect of genuine technical change is to produce ranges of new jobs, rates for which are not laid down by the normal
collective contracts, and which are therefore at least temporarily exempt
from formal collective regulation.
A factor which appears particularly important in several countries—
indeed some students have thought it the most important element in wage
drift in general—is the effect of piece-work earnings, including incentive
systems of payment generally. Piece-work wages naturally tend to rise
with increasing productivity, but since productivity normally grows at
widely different rates the effect on actual wages is necessarily uneven.
In some cases, moreover, piece-work rates are fixed by local or workplace bargaining, which it is difficult to bring under any form of central
control, while incentive payment can obviously be used by individual
employers as a method of increasing wages independently of either central
decisions or of changing productivity.
In the Netherlands, piece-work bonuses have been limited under
wage regulations, but where earnings rise above set maxima it is hard
to determine whether this is due to an unforeseen (but welcome) rise in
productivity or to a " loosening " of rates ; and rate revisions may then be
inhibited by fear of disincentive effects.
In each of these three cases quite normal phenomena of wage movements may thus produce considerable disparities in the relation between
formal wage rates and actual wages. They involve, in any case, difficulties
of control, in the sense of reconciling their effect with the intention of
central wage decisions, and they present opportunities of evading the

138

PRICES, WAGES, AND INCOMES POLICIES

latter. Yet another element in a divergence of actual earnings from formal
wages rates, however, may be the actual payment by individual employers
of rates exceeding the latter. In the past, the general problem of enforcing
collective decisions about wage rates has been that of preventing underpayment. This, of course, is usually ensured by some procedure for the
legal enforcement of collective agreements or of wage rates fixed by
legal wage boards or by compulsory arbitration. More generally, in
western industrial countries, however, enforcement of agreed wage rates
as minima has depended on the extent to which workers are unionised,
and employers organised and disciplined in their observance of agreements. In several European countries, legal procedures exist for extending the terms of collective agreements to unorganised employers and
workers. These various methods have usually appeared at least sufficiently
effective to permit the collective fixing of wages to proceed on the assumption that collective decisions will have a realistic impact.
The problem, however, becomes rather different when the purpose of a
centralised wage-fixing system is to include wages in a general system of income regulation. To prevent individual employers from paying rates above
those envisaged by the wage policy is considerably more complicated
than preventing underpayment, because in the latter case trade union
resistance can be assumed—whereas trade unions, although in principle
they may even be in favour of the observance of agreed wage increases
as maxima as well as minima, will rarely in practice initiate action against
overpayment when their own members are benefiting. In some cases the
extent of overpayment by individual employers will be limited by the
effectiveness of employers' associations or by traditional discipline
amongst employers in relation to collectively-agreed wage rates. Thus in
the Federal Republic of Germany, employers' organisation is not only
very extensive but seems often to exercise a very strong restraint upon
departures from collective agreements by its members. Similarly, several
employers' federations in the United Kingdom possess a substantial
control over the wage policy of their individual affiliates. But even in
such cases the controls of employers' associations cannot extend to nonmembers, among whom there may be very large independent concerns
which are able to resist pressure, as well as the much more numerous small
firms who may evade regulations. In addition, many employers' organisations have no such control over their members, nor any strong tradition
of uniformity in actual wage rates.
This question of employers' cohesion is of special interest in Europe
because of the importance there of industry-wide bargaining. The traditional function of industry agreements, for example in Germany, the
United Kingdom and (to some extent) Denmark, is to fix minimum wage

SOME PROBLEMS OF NATIONAL WAGE POLICY

139

rates keyed to the capacity to pay of the weakest firms in the industry,
so as to protect the workers against the hazards of competition in the
markets for products and labour.1 In keeping with this pattern of bargaining, trade union organisations in western Europe tend to be strong at the
national level and much weaker at the plant level. In Great Britain the
unions extend to the plant level through the shop steward system, but
in many cases the effective control of the unions over the stewards is
weak. It was predicted some time ago that for this reason " there isn't
going to be an effective wage policy—now, or if and when the Labour
Party gets into power ".2 For even if the general secretaries of the unions
were willing to accept wage restraint, their bargaining functions might
well be taken over by the shop steward committees. In Italy the "internal
commissions", in the Federal Republic of Germany the " works councils"
and in France the " enterprise committees " are but weakly controlled by the
national unions.3 Thus the unions could easily find themselves in an
embarrassing position if they sought to counter wage increases that could
be obtained by the works committees.
In some countries the enforcement of legal, arbitrated, or agreed
wage rates has therefore been extended from the legal prohibition of
underpayment to that of overpayment. However, this has usually been a
temporary measure. Several such measures—like the grant to the Australian and New Zealand arbitration courts of the power to fix maximum
as well as minimum wages—have applied only in wartime. A legal
" wage stop " was enforced in Norway in 1947, but was not continued.
Only in the case of the Netherlands has any systematic and continuing
control over maximum wages been attempted. It seems probable, in
fact, that any attempt at legal control is likely to be ineffective unless there
is already a high degree of collective discipline on both sides of the labour
market, and unless the normal coverage of collective agreements or legal
provisions is so wide that it includes the many other things which may
be offered as an alternative to straight wage increases (and which may
thus represent a concealed wage drift) such as fringe benefits, welfare
provisions, and so on.
In general the extent to which the increase in actual wages has exceeded that provided by collective agreements seems to have been fairly
1
In the Netherlands, official rates are also actual rates, since the concept of simple
minimum rates would be incompatible with government wage control. In the United
States, actual wage rates are usually equal to those officially negotiated.
2
Peter DUNN in The Observer (London), 8 Sep. 1963, p. 7.
3
Theoretically, French unions have sub-locals in each plant, but the functions of
these bodies are very restricted. Some of the other councils and committees are not
legally allowed to bargain over wages.

140

PRICES, WAGES, AND INCOMES POLICIES

moderate in recent years1 and may be partially accounted for by changes
in the distribution of the labour force, and by upgrading in wage-scale
ratings due to technical progress. However, even where the gain of earnings over wage rates is accounted for entirely by these factors, the incidence is necessarily rather uneven and may lead to inequities between
workers. Inequities of this sort are particularly likely to appear where
such factors as rising piece-work earnings, additional payments by employers, or gains from supplementary collective bargaining are significant.
And the accumulation of anomalous conditions arising from such
sources may create a dissatisfaction which could make it difficult to
maintain the process of central agreement or the effectiveness of arbitration or official awards, even where the total excess over agreed increases
was not sufficient in itself to jeopardise the general policy of restraint.
These things may be treated as problems of wage structure reform:
for instance, where industry-wide collective agreements exist, an attempt
may be made to reduce established inequities by the rather elaborate
process of absorbing existing workplace wage differentials into a new
industry wage scale. On the occasion of the 1963 wage explosion in the
Netherlands, individual firms were allowed, as a special case, to increase
their wages by up to 4 per cent, over and above the level fixed for their
industry, for the specific purpose of thus legalising payments that had
been in excess of official wages and therefore unlawful. Similarly, Swedish
master agreements 2 have taken account of wage drift during the period
preceding negotiations, in part by specifying separate rates of increase
for piece and time workers so as to enable the latter to catch up with the
former. In the Netherlands again, wage drift has been taken into account
in the process of determining the scope for wage increases, by deducting from the theoretical amount a percentage of " autonomous wage
increases " representing the rise which experience had shown to occur
quite independently of any rise in basic rates.
Another approach to the problem of wage drift arising from piece
work would be to improve methods of piece-rate fixing and their adjustment to rising productivity. In the Netherlands piece earnings based on
bona fide work measurement have been permitted to rise further above
basic rates than those based on more casual methods of rate-fixing. In
Norway,firmsattempting to keep wage drift within limits have succeeded
in doing so to some extent by delaying piece-rate fixing until they have
had a few weeks of experience with the jobs concerned; firms with
1
2

See above, table V, p. 20.
See above, p. 101.

SOME PROBLEMS OF NATIONAL WAGE POLICY

141

much wage drift were found to have encouraged this in part by deliberate
generosity in piece-rate fixing.1
However, some of the problems involved—such as the maintenance of
an equitable relationship between piece-workers' and time-workers'
earnings—are difficult technically. (Indeed, proposals have sometimes
been made for the extreme measure of abolishing piece-work and incentive payment because of its potentially embarrassing consequences for
wage policy.) There is, too, the further difficulty that the elimination of
anomalies from such sources would require some workers to sacrifice
gains which they might otherwise have automatically anticipated, and
others to restrain themselves from making compensatory demands while
the rather gradual process of wage structure rationalisation is carried
through.
Generally, however, it is clearly necessary, if incomes policy is to be
effective, that the influence of central agreements or awards should
extend beyond the simplefixingof general wage advances. In the Netherlands case, wage policy first envisaged the ultimate adjustment of all
sectional wage rates to a national scale of wage differentials. In Sweden
policy has attempted to meet this problem rather by way of sporadic
correctives, through discrimination (in the wage increases permitted or
recommended by central agreements) in favour of groups (such as time
workers in manufacturing) who had " drifted " behind.
The difference between the movement of nominal wage rates and of
actual wages may, however, be of considerably greater importance in certain
circumstances. In particular, a sharp boom, involving a general upsurge of
profits, is likely to increase the difficulty of maintaining a controlled
relationship between the two. Thus the boom due to the Korean war
would probably in any case have made it impossible to maintain wage
restraint in Sweden and the United Kingdom beyond the end of 1950.
Similarly, a persistently high level of profits in one sector or industry,
coupled with a labour shortage, is very likely to induce a differential rise
in wages in that sector. In several countries, for instance, the post-war
shortage of housing, and the very considerable demands on the construction industries arising from population growth and movement, industrial
reconstruction and new investment, led to a particular labour shortage in
the building trades and the appearance of a black market in building
labour. This was the most important reason for the legal wage stop of
1947 in Norway, and has always been a substantial problem for wage
policy in the Netherlands. Similarly, the post-war prosperity of the timber
and pulp trades in Sweden was a continuing embarrassment to central
AARVIG, op. cit., pp. 185 and

186.

142

PRICES, WAGES, AND INCOMES POLICIES

attempts at wage restraint. And finally, the wage structures of certain
countries may make even the normal gain of actual earnings over wage
rates so substantial as to constitute a continuing problem. This is probably to a large extent an affair of the special character of their statistical
indices. In Norway and Sweden the wage drift seems remarkably high,
even under conditions in which employment and profits are relatively
depressed.
The evidence suggests, therefore, that the movement of actual wages
is amenable to a continuing central policy only under two conditions.
First that there is no state of pronounced excess demand, either generally
or in any substantial sector of the economy, so that if for no other reason
an attempt to control inflation by wage policy alone seems certain to be
ineffective. The second condition is a willingness on the part of individual
trade unions, groups of workers, and employers to sacrifice some autonomy in matters of wagefixingunder circumstances which may be particularly favourable or conducive to their exercise of it.
INSTITUTIONAL PROBLEMS

Another group of problems that attempts at national incomes policies
may encounter is caused by the fact that the structure of labour market
institutions may be ill-adapted to the requirements of central
agreement or control. For instance, one background factor in the wage
drift in several countries may be a traditional difference between industries in the function of widely-based collective agreements (that is,
agreements the scope of which is not confined to the employees of a
single firm but extends to all the members of a particular occupation,
or to all thefirmsin a district or industry). In some industries, and particularly in public services, such an agreement may determine absolutely
the rates to be paid for all occupations and all establishments within its
scope. In other industries the general agreement is treated only as a base
on which further agreements may be made for particular occupations,
districts or individual firms. In the latter cases, the basic agreement is
itself often a rather simple affair specifying only general conditions, such
as normal working hours or holidays, etc., and containing only a few key
rates, while the detail of the wage structure isfilledout in local or workplace bargaining. It may be the accepted thing for such agreements to be
regarded as settling only minima, and for average agreed wage rates to
be rather higher. Where both standard and minimum types of general
agreement exist together—as in Scandinavia and the United Kingdom—
this creates a problem for co-ordinated policy, since whereas in one case
a centrally agreed wage advance would determine the actual increase in

SOME PROBLEMS OF NATIONAL WAGE POLICY

143

wage rates for the industry, in the other it would normally represent
only a minimal advance to be followed by supplementary negotiations.
Institutional problems, however, are still more acutely raised by the
trade union structure in many countries. In all those with which we are
concerned the unions are primarily sectional associations whose principal
object is the forwarding of their own members' interests. They are also
democratic organisations, in which the existing leadership's efficiency
is likely to be assessed very largely in terms of the gains secured to the
members by its policies. But the basically sectional character of trade
unionism may be either modified or intensified by the particular structure
of national labour movements. This involves such questions as the number of separate trade unions that exist, the extent to which they are
organised on the same principles or are in competition with each other
for membership, the extent to which they are federated, the power of
such federal bodies in relation to their affiliates, and the general degree
of unity on broad policy in the trade union movement as a whole.
There are clearly cases in which the structure of the movement itself
would make any systematic or effective process of central bargaining
impracticable, and where only a comprehensive system of compulsory
arbitration or legal restraint would supply an effective mechanism of
general wage regulation. But even where unions are disciplined in the
sense that they can usually guarantee their members' acceptance of agreed
policies, and where the federal organisations are comprehensive and
powerful in relation to their affiliates, it is by no means always possible to
secure a general agreement. Thus it may be extremely difficult to persuade
a union operating in an industry where profits are persistently high to
limit its wage demands to those considered tolerable by the economy as
a whole, or to enforce any such rigid principle or uniformity of treatment
in actual wage packets. And in practice it has usually been found necessary to make some concession from the general principles of wage policy
to accommodate such circumstances. In Sweden this has been done by
intermittent relaxations of central restraints on sectional bargaining. In
the Netherlands, apart from the general wage agreement of 1956, which
permitted actual wage increases to vary within a range to be determined
by negotiation in particular industries (though little variation in fact
emerged), it seems to have been achieved rather more indirectly by
introducing a certain practical elasticity into the actual application of the
formal wage structure.
It seems possible that any over-prolonged or rigid restraint of wage
advances would imply too rapid a transformation in the character of trade
unions to be tolerable. For example, the very rigid system of wage control
originally envisaged under post-war Norwegian arrangements seems to

144

PRICES, WAGES, AND INCOMES POLICIES

have involved strains which, in fact, led to a subsequent relaxation to a
degree of centralisation considerably less than that of the more flexible
Swedish procedure (to which it later began again to approximate). In the
Netherlands, fairly strict controls could be maintained much longer, but
from about 1962 these were relaxed, as regards both procedures and
guiding principles for wage adjustment, [t may be, in fact, that to maintain the general acceptance by trade unions of a long-run system of
central wage guidance, a periodic relaxation of central control is
desirable.1
It seems, in any case, that the degree of wage stability that can be
hoped for from a system of central wage regulation in a mixed economy
with free trade unions is limited. A restraint on wages which is excessive,
either in duration or in relation to the general possibilities of increasing
real incomes, is likely on the one hand to lead to evasion by private
employers and thus to both social and economic distortions in relative
wages, and on the other hand to intolerable tensions in the trade union
movement. Thus an attempt to go too far too fast may prejudice the
growth of the " consensus " and collective discipline on which any
system of centrally agreed wage direction must rest, and provoke compensatory outbursts of competitive wage inflation.
It may also be true, however, that several attempts at central wage
policy have had to be made under circumstances of particular difficulty.
In several cases, systems of wage restraint have been identified with
stringent wartime controls in general, or with the severe restraints
which many countries found it necessary to impose on all sorts of individual choices and economic activities in the period of post-war reconstruction. The subsequent drive to relax such restrictions inevitably affected
to some extent the acceptability of wage controls. In particular, central
wage policies have had to deal with a number of rather abnormal pressures. In the period up to 1950, at least, several of the economies in which
they were attempted had not only the exceptional post-war demand to
deal with but, inevitably, sectors (like building) in which demand was
particularly intense and labour shortage disproportionately severe.
Moreover, it was also necessary during the period to take into consideration a very strong movement—which had both economic and social
sources—to raise the standards of lower-paid workers of all kinds.
1
Just as it would be possible to use demand restraints on inflation only as intermittent checks to prevent cost inflation which may develop in association with a period
of economic growth from becoming habitual and to engineer desirable transfers of
resources by occasional sharp reductions in demand, so a possible long-run wage policy
might consist of an alternation of periods of relatively free sectional bargaining with
intervals of wage restraint in which accumulated anomalies in the wage structure were
also reduced by enforced priorities in the limited wage increases.

SOME PROBLEMS OF NATIONAL WAGE POLICY

145

Then in 1950-51 there was the abnormal impact of the Korean boom and
the consequent very sharp rise in the prices of food and raw materials.
Under these circumstances it may well be that the demands made upon
wage policy were quite exceptional. With the rate of productivity
growth which now seems practicable in most industrial countries the
degree of actual restraint on wage movements implied in a system of
central wage co-ordination in future is likely to be considerably less.
One further institutional factor determining the possibility and effectiveness of a national wage policy is the extent to which contractual
payments (wages, salaries, and certain similar payments) actually are
the object of collective negotiations of a kind that could be fitted into a
national policy. In several countries, notably including the United States,
a fairly large proportion of manual workers are unorganised and their
wages arefixednot after negotiation but as a result of unilateral management decisions. But while in such cases wage adjustments may in practice
follow closely those resulting from collective bargaining in other sectors
or firms (and thus could also be affected by any central guidance that
might be given with regard to wage negotiations) this is perhaps rather less
likely for salaries of non-manual workers. The trade union movements of
most countries in western Europe and North America are still substantially manual workers' organisations. Although some groups of nonmanual employees have for long been effectively organised, by and large
the proportion of trade unionists among salaried workers is still much
smaller than it is among manual workers, and the coverage of salaries by
collective bargaining is much less than for wages. In 1960 non-manual
workers constituted over 42 per cent, of the labour force in the United
States (compared with 37 per cent, in 1950 and 31 per cent, in 1940), and
this percentage is still rising. In other industrialised countries the proportion is lower, but in all of them the date is coming within sight when the
number of white-collar workers will exceed that of blue-collar workers,
and salaries already rival wages as an element in labour costs in many
countries. As long as the non-manual employees were but weakly
organised, the movements of their salaries gave rise to no inflationary
concern. Indeed, over an extensive period, clerical salaries have been
found to decline compared with the wages of manual workers. But in
recent years professional workers in various countries have shown a tendency to organise in professional associations as well as to growing militancy, and increases in salaries are now clearly becoming an important
factor in rising costs and prices. Yet in the Netherlands wages and
salaries, which (at the moment) exceed about £800 a year, in practice
escape wage control, while in Britain, doctors and dentists, together
with higher civil servants and executives in the nationalised industries,

146

PRICES, WAGES, AND INCOMES POLICIES

were excluded (with other groups which had their own private arbitration
procedure for pay determination) from the terms of reference of the
National Incomes Commission.
It would appear that an effective incomes policy should cover
salaries of non-manual, and especially salaries and fees of professional,
workers as well as wages of manual workers. To the extent that an increasing proportion of all salaried pay is determined by collective agreements, this represents no greater problem than in the case of wages—
indeed, since such potential elements of wage drift as overtime payment
and output bonuses are less common for salaried workers, the regulation
of salary movements might be more effective. Moreover, a large proportion of all salaried workers are employed by public authorities, or by
large semi-public institutions such as banks, and to these it is clearly
easier to apply any public wage policy that is developed. However, there
will remain the problem of ensuring that movements in salaries in private
employment where these are not regulated by collective agreements
conform to the requirements of national incomes policy, and this may
be particularly difficult for certain higher salaries which are substantially
self-determined, like those of company directors.

THE NEED FOR A CONSENSUS

That a broad degree of agreement on the desirability and the main
lines of incomes policy is a precondition for its operation is illustrated by
much of the foregoing analysis, and by other things.
For instance, where there is, as in the United Kingdom, both a welldeveloped conciliation system and a tradition that outside intervention in
wage settlements will be limited to the provision of mediatory services
or of voluntary arbitration, any attempt to enforce an over-rigid restraint
may prejudice the future acceptance of such conciliatory services and
increase the likelihood of industrial stoppages. There is probably a
tendency for general economic and social considerations to exercise an
increasing weight with arbitrating and conciliation agencies, and this is
probably on the whole to the good. But if it exceeds by too wide a margin
the extent to which such considerations are also accepted by unions and
employers, the result may be not merely to sacrifice a potentially valuable
instrument of co-ordination between wage settlements but also to increase the chances of industrial dislocation because such mediatory
services may be regarded as no longer treating issues in dispute " on
their merits " to the parties concerned.
A second illustration concerns the special position of public employees
in a national incomes policy. It has been a quite common rule for deter-

SOME PROBLEMS OF NATIONAL WAGE POLICY

147

mining wages and salaries in public employments that these should be
broadly equal to those payable in comparable occupations in the private
sector. Where the government has an incomes policy, however, it would
seem inconsistent for it not to apply this to its own employees, and difficulty will certainly arise if the policy's broad lines are not also accepted
by (or otherwise imposed on) unions and employers in private employment. The policy may then appear as an act of discrimination against
public servants and lead to consequent unrest among them (as recently in
France, and in the United Kingdom under the " pay pause " of 1961).
Thirdly, the usefulness of government exhortation and consultation
in regard to particular wage settlements or on a more-or-less permanent
basis obviously depends greatly upon how far some basic consensus
exists between trade unions, employers, and the government concerning
the need for and the basic principles of an incomes policy. The same is
true of the usefulness of appointing groups of independent experts to
provide information and other general guidance to wage negotiators.1 If
such groups disagree too much with the government their mandates are
not likely to be renewed. If their conclusions differ widely from those of
trade unions or employers they are likely to be treated as partial and propagandiste. And if the experts try to avoid this, when there is strong
disagreement among government, trade unions and employers, they may
have to confine themselves to statements that may hardly be worth making.
Finally, unless such a broad consensus exists, it is clear that a centralised wage-fixing system may involve a number of political risks. In a
sense, the purpose of such a system is to transfer economic tensions which
would otherwise lead to inflation into the sphere of political negotiation.
This involves a risk of industrial conflict on political decisions, or political
crises on industrial issues. Where the political tensions are less tolerable
than the economic, the result is likely to be an attempted withdrawal
from central wagefixing—asin the case of France in the early 1950s, and
perhaps in the Netherlands since 1960. Previously, the Netherlands presumably found the political risks of direct government intervention in
wage fixing smaller than its advantages. The Swedes have attempted to
compromise by a system which avoids either direct government participation in negotiation or legal control of wages, and makes flexible the
actual degree of centralisation or public influence in the wage-fixing
process. The risks are especially pronounced, however, where an important section of the trade union movement—as in some countries where
there are separate communist union federations—is opposed to any form
of restraint on wage bargaining.
1

See above, Ch. V.

148

PRICES, WAGES, AND INCOMES POLICIES

Indeed, it would seem that acceptance of a national wage policy by the
trade union movement to a large extent still depends on the political
colour of the national government of the day. For instance the trade
unions in Norway and Sweden, where social democrat governments have
been in power for very long periods, have accepted to conduct, or participate in, a wage policy taking account of national economic conditions.
In the Federal Republic of Germany the Social Democrat Party has
never been in power and the trade unions there have always emphasised
the need for autonomy in collective bargaining, while looking very
sceptically upon proposals for a national wage policy. In the United Kingdom the unions accepted a wage policy from 1948 to 1950 under a Labour
government. At the 1963 Trades Union Congress the sponsor of the
resolution (adopted) declaring the Congress's " complete rejection of any
form of wage restraint " later explained that this statement applied only
as long as there was a Conservative government, and indeed in 1965 an
agreement of principle regarding incomes policy was reached under a
Labour government. This agreement involved the acceptance of broad
limits to wage increases which were not very different from those the
unions had previously rejected when they were proposed by a Conservative government. In the Netherlands the support by the Netherlands
Federation of Trade Unions for a national wage policy involving government control rapidly weakened after the Labour Party had withdrawn
from the Government at the end of 1958.

CHAPTER Vili
POLICY FOR NON-WAGE INCOMES
A question which has only recently begun to be generally discussed, as
a larger number of countries move towards attempts to restrain inflation
by direct influence on costs (and therefore incomes), is the problem
involved in converting a " national wage policy " into a " national
incomes policy "} This extension, it has already been noted, is necessary
if the co-operation of unions and workers in wage planning is to be
secured. It is desirable not merely on grounds of equity; although prices
and profits in the industrial economies are generally much less institutionally determined or administered than are wages and salaries, there
are also in most such economies sectors where this is not so (besides such
general phenomena as downward rigidity of prices, etc.); and (as has also
been already noted) these things may make an independent contribution
to cost-inflation. Moreover, it will clearly be difficult to hold to any
general agreement on wage policy if price levels rise sharply despite it
or to an extent not envisaged by it.
Indeed, since inflation has been defined as a persistent tendency for
prices to rise, the most direct attack on it might seem to be a comprehensive system of price control. However, while this may be practicable
for a short period—for instance, for a temporary price freeze such as
adopted by Denmark and France in 1963—it clearly involves considerable
difficulties if the period is prolonged and the regulation is restricted to
prices alone. Without a system of administration and enforcement so
cast-iron as usually to be impracticably elaborate and expensive, it will
not necessarily restrain the growth of incomes and therefore costs, and
where it fails to do this the almost inevitable result is a black market.
Where it succeeds, however, the result may be unemployment. Wages
may rise to a certain extent at the expense of profits, and in some cases
this may have no effect on the demand for labour. But the room for
reductions in profits necessarily varies from sector to sector, and there are
likely to be some where a reduction in output will result.
1
See Organisation for Economic Co-operation and Development: Policies for
Prices, Profits and Other Non-Wage Incomes (Paris, 1964).

150

PRICES, WAGES, AND INCOMES POLICIES

A system of price control is in any case virtually bound to be incomplete (even if it is effective in the sectors where it operates) because of the
difficulty, for instance, of controlling the prices of many kinds of services.
Thus it will tend to produce distortions in the structure of relative prices,
even in the controlled sector, unless controlled prices are varied in relation to the development of costs per unit, because different industries
naturally have different rates of productivity growth which would
normally lead to varying price trends as between one industry and another. However, it is peculiarly difficult so to regulate individual prices
that appropriate reductions in fact occur in the industries where productivity is rising faster than wages, or to control prices effectively where
quality and design of products change quickly.
The difficulties of a permanent and general system of direct profit
regulation are likely to be of much the same kind as those already described for simple price control—indeed, since such a regulation can
operate only on profit margins, it is rather difficult to control profits
directly without erecting a price control system. But another general
problem in attempts to regulate profits directly is that the rate of profit
is by no means the sole determinant of the amount of profits, and it is the
latter which is usually most relevant if profit restraint is required to
facilitate restraint on the part of other incomes. The amount of profits is
determined by other factors too—such as changes in the value of stock
and work in progress as a result of external economic fluctuations, and
particularly the level of economic activity itself. The easiest way to
restrain profits in the sense in which they seem most often thought relevant to wage negotiations, for instance, is to reduce demand: which is
apt to create unemployment.
In fact, of course, general systems of price and profit-margin control
have been largely identified with war and post-war situations of extreme
economic stringency, and the relaxation of price controls which followed
has often left the public sector as the only substantial area in which the
government is able to influence prices directly. The danger of economic
distortion is perhaps particularly present where price regulation is limited
to (or can only be made effective for) nationalised industries, public utilities, etc. The subjection of public enterprises to price restraints in an
attempt to offset price pressures in other sectors has often created substantial difficulties for these undertakings, limiting their development by
making it impossible for them to finance their own investment, or even
requiring their subsidisation by the government.
The risk of distortion in the price structure by selective, but rigid, control is perhaps also illustrated by the case of rents, which in most countries
were subjected to legal controls during the early post-war years. These

POLICY FOR NON-WAGE INCOMES

151

controls were mainly intended to deal with the housing shortage created
by wartime cessation of housebuilding, or by the physical destruction
and decay of housing. Their maintenance into the post-war period in
several cases distorted the accommodation market by restricting supply
to, and concentrating demand on, forms of accommodation which were
naturally difficult to control. In the area where controls have been effective, they have tended to be carried to the point where the supply of private
housing for rent has substantially declined (and its quality deteriorated),
and have sometimes discouraged the movement of population and labour
because people were reluctant to give up controlled accommodation. A
general relaxation has therefore occurred.
Nevertheless, despite these and other difficulties in direct price and
profit controls, several countries have found it useful to maintain continuing regulation or supervision of prices in association with measures of
income planning. The most extensive examples appear to be Norway
(where the general power of the Government to intervene in price fixing
was retained after the relaxation of post-war controls, and where detailed
price regulation was extended as part of comprehensive central wage and
income settlements in 1963), and the Netherlands, where official price
supervision has been regarded as an essential complement to national
wage policy—and where, too, despite the loosening of the latter that has
occurred recently, price-control powers were actually strengthened in
1963. However, in the case of Norway, although specific controls apply to
a large proportion of the goods that are widely bought, it seems recognised that a too-rigid regulation may be harmful, and price-control policy
is flexible not merely in this sense but also in the sense that the goods
chosen for control may be changed from time to time.
In the Netherlands, price policy has proceeded rather by price discussions between government and business representatives, at which rules for
price behaviour have been formulated, than by frequent direct regulation
(though the Government's power to fix prices in case of non-observance
of the rules has naturally provided an incentive for business to adhere
to agreed policies). Such rules have sometimes been attached to national
wage settlements. It is clearly easier for public policy to influence
pricefixingwhere business itself is well organised in trade associations or
employers' associations, or in industries where there is a high degree of
concentration. However, in most countries there are sectors characterised
by many small firms where price control by voluntary agreement could
hardly be effective; these sometimes include trades which have presented
special difficulty to post-war stabilisation policies, such as retailing and
house construction.

152

PRICES, WAGES, AND INCOMES POLICIES

Several other countries have varying degrees of price supervision.
Comprehensive powers of price regulation are still available to the French
Government. Prior notification of price increases is required in Austria
and Belgium and in the former country (where also some direct controls
have been retained) the Joint Price and Wage Commission may request
the Government to enforce a general control for up to six months. In the
United States the President has occasionally issued a public warning
against the consequences of specific price advances planned by business.
In the United Kingdom the Prices and Incomes Board established in
1965 "to keep a continuous watch on the general movement of prices and
of money incomes of all kinds " may have particular prices referred to it
by the Government for detailed inquiry to determine whether or not their
level is justified, and several commodities and services have so far been
thus referred. Price supervision of this kind avoids the problems of general price control, but may still exercise a general restraint on price
movements by exposing to public attention and censure cases in which
price increases exceed what may be reasonably attributed to advances in
costs, or in which prices have been maintained at an unjustifiably high
level. It may also call attention to areas where selective government action
might be fruitful of general results.
A temporary general price freeze (as opposed to permanent regulation) has sometimes been applied, not merely to meet an urgent danger of
intensified inflation but to provide a starting point for a more satisfactory
system of income restraint. Perhaps it should be noted here that there are
also other methods which have helped to provide general conditions in
which agreements about wages have been reached. The cost of living can,
of course, be stabilised by direct subsidies. This is rather dangerous as a
long-term policy because it may lead to distortion in the allocation of
resources as between goods which are subsidised and goods which are
not; but as a short-term measure it may be justifiable. A better longterm alternative perhaps is to protect those people who are most likely to
suffer from an increased cost of living by direct personal subsidies
(family allowances, etc.).
Direct regulation of non-wage incomes themselves has a rather limited
area of possible application, since their major component, profits, is
determined as a residual after other costs (i.e. mainly incomes) have been
deducted from total sales. In several countries farm incomes are largely
dependent on government regulation by price support, subsidy, etc.,
so that in such countries total farm income may be already very much a
matter of central negotiation or public determination.
Dividend distributions by companies have occasionally been subjected
to legal restraints (for instance in Norway) and seem in any case quite

POLICY FOR NON-WAGE INCOMES

153

susceptible to governmental pressure which falls short of actual legal
limitation. (A voluntary " dividend restraint " operated quite effectively
in Britain during 1948-50, and some tax systems discriminate against
distributed profits.) One problem is that the level of dividends at any one
time is a compound of the separate dividends which are paid by many
individual enterprises and which often reflect rather temporary positions.
A mere dividend freeze thus naturally tends to be rather arbitrary in its
impact on different concerns and their shareholders. A wider problem is
that generally restraint on dividends increases the accumulated assets or
reserves of firms. On the one hand this may make it possible for shareholders to offset direct restraint on their incomes by capital gains derived
from a consequent appreciation of share values ; on the other it may add
to the difficulties of setting up which new firms are likely in any case to
experience from the mere fact of dividend restraint, since less new capital
can be secured from personal investors. It may thus increase both inflexibility and monopoly in the economy.
In addition to general or selective price control by regulation and the
less formal supervision or discussion of price increases, there are two other
major routes to a regulation of non-wage incomes. One amounts to
attempting to increase the competitive efficiency of markets for goods and
services. This includes a whole gamut of possible devices, among them
being measures (which have been strengthened in several European countries recently) to prevent or control monopoly and restrictive business
agreements and practices ; incentives (by way of tax arrangements and
loan facilities) to modernise businesses or to start new firms ; the encouragement of freer international trade (including the selective use of tariff
reductions against industries whose pricing policies are thought objectionable); the reorganisation of out-of-date industries under public pressure
or with public support; the introduction of public (or publicly supported
or induced) competition into particular trades. Each of these techniques
is a subject in itself and cannot be discussed in detail here. The other major
route to a general incomes policy is the use of the government's fiscal
powers to correct trends—either in particular categories of income or
in income distribution in general—which seem inconsistent with such a
policy.
These techniques are, of course, not mutually exclusive ; but the use
of taxation to correct disproportionate increases in non-wage incomes
does have the advantages that it operates via a normal function of government, and does not involve a separate administration like price control,
and that its effect can be clearly stated and measured, unlike that of
measures to increase price competition. This may be especially advantageous where control of profits is particularly desired to facilitate accept-

154

PRICES, WAGES, AND INCOMES POLICIES

ance of co-ordinated wage policy. Moreover, it is possible to envisage the
use of fiscal measures not merely to restrain profit incomes as a quid pro
quo for wage restraint (as seems implied by the National Incomes Commission's instruction to advise the United Kingdom Government periodically " whether there is need for restraint of aggregate profits, resulting
from restraint in wage and salary incomes, by fiscal or other appropriate
means ") but to promote social justice as well as economic aims as a
part of a general approach embracing all categories of income and all
stages in its distribution. This view of incomes policy appears projected
by M. Massé 's introductory report to the discussions initiated by the
French Conference on Incomes in October 1963.
However, profit taxation is not without its problems. A market economy relies on the profit motive to promote growth and necessary adjustments to structural changes. It is often said that growth and the speed
and appropriateness of adjustments may be adversely affected by taxes
on profits, particularly if they are high and progressive. However, the
fact that high rates of taxation and high rates of economic growth have
often gone together casts some doubt on the correctness or importance
of this view.
Another point is that since the level of profits fluctuates generally
with economic activity, it is necessary for the other parties to income
settlements to accept a fairly long-run view of the effectiveness of profit
restraints; this may not be easy when what is at issue is a year-to-year
guidance of wages (which perhaps supports the view that wage and income policy are matters for rather more long-run perspectives). Profits
also vary, particularly with the efficiency of individual firms (and also,
in some cases, with the moderateness or otherwise of their pricing policies).
So it is also desirable that governments should avoid methods of profit
taxation which may penalise more efficient concerns—and it is not easy
for tax regulations to distinguish between profits which are high because
of relative efficiency and those which are high because of private pricing
policies. Moreover, some types of profit taxation—for instance, a toosteeply progressive scale—may militate against other forms of income
restraint by making it very cheap for firms to grant concessions to their
employees and suppliers.
It has been suggested that some of the difficulties of dividend limitation or direct taxation of profits could be met by imposing instead a personal tax on income from dividends, levied at a rate that would be determined with retroactive effect at the end of each year, in such a way as to
make the average rate of increase in incomes from dividends equal to
the average rate of increase in labour incomes. The tax would be based
upon an index of dividends, recording the amount by which dividends

POLICY FOR NON-WAGE INCOMES

155

had changed on a wide range of representative shares, and an index of
earnings per hour covering salaries as well as wages. When the dividend
index increased faster than the earnings index, individual incomes from
dividends would be taxed at a rate determined by the cumulative difference between the two indices since a base date, in such a way as to keep
the increase in post-tax dividend incomes in line with the recorded increase in labour incomes.1 This would offer a guarantee to trade unions
that, to whatever degree they could reduce the rate of increase of earnings
per hour, an equal reduction would occur in the rate of increase of dividends in the hands of theirfinalrecipients. Such an " incomes equalisation
tax " could, of course, be supplemented by other measures such as taxation of capital gains or of wealth.
Wage and salary policy in the industrial countries appears to have
paid little attention to the pay of directors and related higher management—although, given the managerial revolution in company control,
this may be more freely self-determined than almost any other type of
income. Yet directors' fees and emoluments represent a quite substantial
ratio to (for instance) ordinary shareholders' dividends in some economies, and very high directoral pay and perquisites may be a visible disincentive to wage restraint by workers. Steep rates of personal taxation and
a tight supervision of tax-free benefits are not necessarily a complete or
satisfactory answer to this problem: since the higher management élite is
becoming organised collectively in some countries (an example is the
British Institute of Directors)—or is often at least in a large degree identifiable personally—the inducement to some overt act of " collective selfdenial " on its part, to facilitate broader agreement on incomes, does not
seem quite impracticable. A private members' Bill was recently introduced
into the British Parliament to require directors' salaries and other emoluments to be declared, in the hope that this exposure to public notice would
itself act as a means of restraint ; the proposal however has not so far been
proceeded with.
1
The Economist (London), 28 Sep. 1963, p. 1135, and 30 Nov. 1963, p. 946. The
scheme proposed would include a remission of tax to recipients of dividends in years in
which dividend incomes increased more slowly than earned incomes. But the principle
of using personal taxes on incomes from dividends to establish a relationship between the rate of increase of dividend incomes and that of labour incomes seems capable
of other applications than its suggested use to ensure equality between these two rates.
Equality in rates of increase is the appropriate relationship if it is felt that the distribution of income between labour earnings and profits at the base date is optimal. If,
however, incomes policy were conceived as a measure not only against inflation but
also to bring about gradually a greater equality of income distribution, consideration
might be given to utilising the kind of tax proposed in The Economist to maintain a
slower rate of increase in post-tax dividend income than in labour income; or it might
be decided that remission of taxes to dividend receivers in years in which dividends
increased more slowly than labour incomes would not be appropriate.

156

PRICES, WAGES, AND INCOMES POLICIES

The distinction between private income and the accumulation of
private wealth is largely a formal one. It has been noted that dividend
limitation (direct or by tax discrimination) may be offset by a resultant
appreciation in shares, and profits taxation or price control may similarly
induce a self-investment by firms which would otherwise not occur, and
so increase their owners' wealth. These things may also stimulate productivity growth, but where wage restraint is urged in order to permit
(inter alia) a high level of investment, it may be difficult to induce the
continued co-operation of wage earners if that also produces a sharper
growth in private wealth, which is already much more unequally distributed than is income. Effective capital gains taxes operate in some countries—for instance the United States. In the Federal Republic of Germany and the Netherlands proposals have been canvassed for schemes of
" collective profit-sharing " and " investment wages " by which a part of
the accretion of business assets would be allocated in the form of shares to
social funds, or distributed to employees (with restriction on subsequent
sale—for instance, not before retirement). Some of the gain in wealth
resulting from incomes restraint (and the faster economic growth that it
might facilitate) would, it is thus argued, visibly accrue directly to popular welfare. This subject is discussed in somewhat greater detail in the next
chapter.

CHAPTER IX

STEPS TOWARDS STABILITY—IMPROVING
WAGE SYSTEMS
The problems of formulating and implementing a wage policy 1
and the equally great difficulties arising with policies in regard to other
incomes, make it probable that, although an increasing number of
countries may well experiment in the near future with methods of direct
control or guidance of wage and other income movements, progress
towards fully comprehensive and effective price and incomes policies will
be gradual. Although the inflationary problems which such a policy
would seek to solve have been troublesome, they have not been catastrophic. This opens the question whether measures could be devised
which, without providing a complete and once-for-all solution, would
reduce the inflationary effect of modern methods of price and wage
fixing, or prepare the way for more basic improvements, or do both.
In this concluding chapter some possibilities of devising such measures are briefly examined. The purpose is to stimulate discussion of
certain questions that are difficult but need not be beyond answering,
rather than to suggest any firm conclusions. The points to be considered
would have a bearing on four problems. First, how could general agreement by trade unions to moderate their demands in collective wage
bargaining be secured? Second, how could centrally agreed wage adjustments be translated into appropiate wage decisions at the plant level?
Third, how could some of the inflationary features of certain systems of
wage payment be weakened? And fourth, in what ways could government participation in wage determination procedures help in promoting
stability? The approaches mentioned below are meant to be illustrative
rather than exhaustive.
GREATER SECURITY OF WAGES AND WAGE ADJUSTMENTS

One possible way of somewhat reducing the size of negotiated wage
increases might be to offer the unions more security, of both wages and
1

See Ch. VII.

158

PRICES, WAGES, AND INCOMES POLICIES

wage adjustments. This could be done through long-term agreements with
provision for automatic annual (or more frequent) " improvement "
increases and for automatic adjustment to increases in the cost of living,
of the kind already mentioned.1 Such agreements might further include
wage guarantees, in the sense of a minimum number of hours of work
per man per year (or some other suitable reference period). Trade unions
might well be prepared to accept a smaller wage increase in return for
guarantees as to its periodicity and the stability of the total wage
package. Long-term contracts, too, may increase the possibib'ty of a
general co-ordination of wage movements. If the main agreements could
be negotiated at about the same time, the need for adapting wage adjustments to over-all economic prospects would become fairly obvious. Of
course, recognition of this need by one side of industry might lead to the
intrusion of political influences in wage bargaining and disputes, rather
than to the adoption of a national wage policy, if either unions or
management could plausibly argue that their position was in accordance
with the requirements of national economic prospects and should therefore logically receive government support.
Guaranteed wage increases make wage costs less flexible downward
in case of recession, but the fact that in the United States the initiative to
introduce automatic annual advances came from the employers suggests
that they expected the results to be beneficial from the point of view of
wage costs. At the same time the principal trade union involved took the
initiative to spread this type of contract to other firms and industries,
which suggests that from the union point of view, too, the arrangement
was attractive.2
The net effect of such provisions, of course, would be to make wages
much more like salaries, and it is relevant here that salaries in general are
more stable than wages in both directions, up and down. There is an
interesting prospect opened up by the abolition by some firms in the
United Kingdom and the United States of traditional forms of wage
payment. Thesefirmshave put all their employees on a salaried basis, and
the possibility of others also doing this will obviously increase with
mechanisation, automation, etc. One other feature of salaried employment, however, has not usually been applied to wages, at least for adult
workers, and may also be worth taking over. This is the incremental scale
principle. One reason for the less demanding character of salaried workers
collectively may very well be that as individuals many of them get
automatic annual increments.
1

See above, p. 114.

GARBARINO, op. cit., pp. 26 ff.

STEPS TOWARDS STABILITY

159

Yet another possibility is that of trading economic concessions for
non-economic ones. It seems probable that the willingness of unions
in the United States to accept the wage control system during the Second
World War, for instance, was at least partly attributable to their desire
for union recognition and security. Similarly, the trade unions in the
Federal Republic of Germany apparently preferred (for a while, at least)
to concentrate on securing representation in management rather than
making straightforward economic demands. The extent to which concessions in such directions will be acceptable as an alternative to immediate
material improvement obviously depends on the outlook of individual
trade union movements and their members.
One further possibility of reducing wage demands by offering greater
security instead lies in improved industrial pensions, supplementary
unemployment benefits, and generous redundancy compensation. Such
schemes might be administered by trade unions, and in some circumstances the benefits might be limited to their members. In that case trade
unions in some countries might be enabled to regain some of the ground
which, through the growth of public social insurance schemes established
by law, they have lost as providers of important services to their members.
Schemes of this sort would be less costly to employers only if workers
were prepared to accept employers' contributions to insurance and
compensation funds somewhat smaller than the amount they would
otherwise have demanded as a straight wage increase. But supplementary
unemployment benefits and provisions for larger severance pay would be
less inflationary than equivalent wage advances, since they constitute
savings when employers' contributions are paid in, while providing a
valuable boost to demand during recessions when benefits are paid out.
Improved pension schemes share the former feature, but not the latter,
so that their net anti-inflationary effect is more doubtful. Furthermore,
decentralised private pension schemes may reduce labour mobility.

WORKERS' PARTICIPATION IN PROFITS AND INVESTMENTS

The main and obvious retort by trade unions to appeals for moderation in their wage demands is of course that there is no guarantee of
equivalent restraint being applied to other incomes, quite apart from the
fact that (without any obvious justification) the existing distribution of
incomes is grossly unequal to begin with. Furthermore, precisely in
periods of high employment, when the pressures for trade union moderation are strongest, profits are at their highest and may indeed increase
further by restraint on the unions' part.

160

PRICES, WAGES, AND INCOMES POLICIES

As noted above1 these high profits are, of course, used to a large
extent for investment and thus contribute to a faster rise in average living
standards. But this is not an adequate answer, because the process of
reinvesting undistributed profits entails an increase in private wealth on
the part of business owners and managers (but not of workers) which in
several countries has been very fast indeed during the post-war period.
In so far as the justification for wage restraint and high profits lies in the
need for providing investable resources, it should be possible to achieve
this result without the ownership of all the new capital accruing to
shareholders, managers and directors.
On this basis, proposals have been put forward for systems of " collective profit-sharing " whereby a proportion of business profits would be
allocated, in a form that would not allow this share to be immediately
consumed, not to the employees of undertakings as such, but to general
pools for the benefit of the workers at large. For instance, in the early
1950s a scheme to this effect was suggested by the Netherlands Federation
of Trade Unions.2 Recently a rather similar proposal was announced on
behalf of the Trade Union Federation (D.G.B.) of the Federal Republic
of Germany. According to the latter plan,firmsabove a certain size would
be required by law to transfer the equivalent of a substantial portion (of
the order of 50 per cent.) of annual increases in their capital into regional
investment funds from which workers covered by the scheme would
receive certificates. These certificates would be negotiable only when their
nominal value had reached a fairly large amount; so as to ensure that the
system would effectively lead to a redistribution of wealth, inducements
would be offered for investment. The fund would be managed by a body
comprising representatives of employers, workers and the Government.3
In the summer of 1964 a somewhat similar system was again proposed
in the Netherlands, but this time by the three national trade union
federations, and with the intention that iheir proposals would become
the object of concrete negotiation in the near future.
These new approaches to profit sharing obviously raise a great many
problems of an administrative as well as of a general economic nature.
For instance profit sharing would resemble profits taxes in respect of any
effects it might have on incentives to enterprise and risk bearing. Profits
would have to be defined and the conditions in which workers' shares
might be sold for current consumption would have to be determined.
1

See above, p. 64.
A. VERMEULEN: "Collective Profit Sharing", in International Labour Review,
Vol. LXVII, No. 6, June 1953, pp. 495 ff.
3
Heinz MARKMANN in British Journal of Industrial Relations, Vol. II, No. 3, Nov.
1964, pp. 322 ff.
2

STEPS TOWARDS STABILITY

161

There is the question of how schemes of this nature could be made to
cover workers in sectors that are not expected to make profits, such as
the civil service. And since profits are earned by fixing prices above costs
and thus have been " contributed " by all consumers of the products
concerned, it is legitimate to ask why, or to what extent, workers as
distinct from other consumers should have a special claim to a substantial
part of them. On the other hand, distribution of profits among both sides
of industry would appear less inequitable than their appropriation by
one side alone. The existing distribution of income is by no means selfevidently just, and it would seem doubtful that any lasting system of wage
policy could be acceptable to the trade union movement if it did not come
to grips—whether through collective profit sharing or otherwise—with
the grosser existing inequalities.
Any form of collective profit sharing requires of course a fairly high
degree of solidarity among the trade unions concerned. More individualist
unions would presumably prefer to reserve profits made in a particular
firm or industry for the workers directly concerned.
One other possibility of compensating wage earners for exceptional
profits is the device adopted by the United Kingdom cotton industry in
the boom after the First World War, and again in the Korean period. This
was a collectively agreed (and thus industry-wide) bonus on the year's
earnings. Since it was a once-for-all payment, it did not raise wage rates
permanently as a result of a temporary boom. For this reason, and because
it carried the implication that the textile workers might not be so well
off when the boom had passed, it seems not to have had the same effects
on wage rates outside the industry as a normal wage advance would have
done; and much of the bonus appears to have been saved by the cotton
operatives. It provided no justification for price increases, since it
explicitly represented a sharing-out of the gain from prices already thought
exceptionally high.
General once-for-all payments have occasionally been awarded in the
Netherlands. On the occasion of the negotiations in October 1963 the
trade union federations claimed a substantia] payment of this kind to
compensate workers for the fact that previous wage increases had been
based on an estimate of the scope for wage increases which subsequent
developments had shown to be unduly low. However, the claim was not
conceded.
COMBINED WAGE AND PRICE NEGOTIATIONS

Still another possibility of persuading trade unions to moderate their
wage demands (especially when the existence of high profits would make

162

PRICES, WAGES, AND INCOMES POLICIES

restraint seem incomprehensible to their members) would be that of
including prices in the process of wage negotiation. There are three good
reasons for doing this. First, a union which is able to negotiate price
reductions at the same time as wage increases might in some cases be
prepared to moderate its wage demands to secure the resultant public
goodwill. Second, profits as well as wages would become a subject for
collective negotiation, and union pressure might become a lever against
monopoly profit. And third, synchronisation of a wage increase and a
consequential price increase would prevent a merely temporary boost to
real wages provoking subsequent wage claims to maintain it.
Where central wage negotiation or control has been applied, it has
not been uncommon to include price agreements or directives. From 1956
central wage decisions for the Netherlands have been accompanied by
more or less clear indications as to whether, or to what extent, wage
increases can be passed on to prices, and employers are now required to
notify the agency that approves collective wage agreements of the effects
that the proposed contracts would have on prices. The Danish stabilisation measures taken early in 1963 were accompanied by legislation to
freeze prices, profit margins and agricultural support programmes (as
well as dividends and similar incomes).1
A straightforward proposal for a moderation of trade union wage
demands in exchange for price reductions by the employers was made
a few years ago by Mr. Walter Reuther, President of the United Automobile Workers of America, to the United States automobile manufacturers. A reduction in prices, of course, would be a rather logical complement of the type of agreement existing in that industry, under which
wages are explicitly made to rise at the same rate as over-all productivity
in the economy as a whole, while productivity in the industry itself rises
faster than this average.2 The importance of explicitly negotiated price
reductions in this case would be the greater since the industry is a patternsetter for collective agreements in other important industries.
INCREASING CONTROL OVER PLANT BARGAINING

Trade union consent to a wage policy is necessary, but trade union
control over actual wagefixingis limited.3 As has already been noted,
many wages are the result of what might be called a two-level wage1

In Sweden wage negotiations have always besn linked with farm incomes and in
1959 a six-year agreement was concluded providing that industrial wage earnings were
to be used as a direct guide for fixing these incomes.
2
See above, p. 114.
3
See above, p. 138.

STEPS TOWARDS STABILITY

163

fixing system, in which general agreements are made on an industrial,
occupational or regional basis, but a good deal of supplementary bargaining—sometimes individual, sometimes collective—occurs in the
workplace. Even when it is collective, bargaining at this second level may
be largely beyond trade union control because it takes place in bodies
(such as various kinds of legal works councils) on which the unions as
such are not represented, or because union plant representatives act
largely on their own initiative, independently of the central union organisation (as do the shop stewards in parts of British industry). Thus, in
Austria, France, the Federal Republic of Germany and the United
Kingdom most workers' actual wages are far above the rates negotiated
by the unions at thefirstlevel of bargaining. Negotiations at that level do
influence actual wages, which are likely to be raised by the negotiated
increases ; or wage relationships may be restored that had been upset as
a result of uneven wage drift. But further increases may be negotiated at
the second level, weakening the effectiveness of any wage policy that may
be applied at the first. Furthermore, the remoteness of most individual
wages from the amounts that are the subject of trade union bargaining
has been regarded in some countries as undermining the position of the
unions with their members.1
Aligning the two types of wagefixingwould strengthen the effectiveness of wage policies as well as the position of trade unions.
In order to find means of doing this, it is useful to recall the origin of
this dual system. It is traditional in some industries, particularly those in
which piece rates are usual but in which no method for central rate
determination exists. In other cases, the system is the result of obsolescent
central bargaining at thefirstlevel.2 In several countries the main structure
of collective agreements is still remarkably similar to what it was in
the inter-war period. Thus industry-wide agreements are still the norm
in Norway and Sweden, and occupational agreements in Denmark, in
spite of periods of economy-wide control and negotiation. In broad sectors
of industry in the Federal Republic of Germany collective wage negotiation seems to have reverted to what it broadly was under the Weimar
Republic.
On the other hand, the industries covered by such agreements have
often changed immensely—this is particularly true of the engineering and
metalworking trades—while the problems which confront negotiators
and the general economic and social context of their proceedings are
clearly quite unlike those which existed when this structure formed. Thus,
1

See above, p. 139, and MARKMANN. op. cit.

See above, p. 138.

164

PRICES, WAGES, AND INCOMES POLICIES

many agreements may have much too wide a coverage to be effective in
contemporary circumstances. For instance the national agreement for
the engineering trade in the United Kingdom now embraces such a wide
range of specialised industries and so many new occupations that it is
quite impossible for their particular problems to be dealt with by global
bargaining.
Similar difficulties have arisen in other countries in connection with
collective agreements in the metal and other trades, thus making the
development of supplementary bargaining inevitable. The problem then
seems to be one of rationalising the collective bargaining system and
adapting it to modern conditions.
An attempt to rationalise the system should probably include efforts
both to define the area of wide agreements realistically in relation to the
actual divisions of the contemporary industrial structure and to define
more closely the relationship between these wide agreements and workplace bargaining. For instance the wide agreement might confine itself to
determining only key rates in the wage structure in detail, but might also
include a statement of principles for the determination of other rates at
the local level and a system of supervision. This would be something
rather like the established system of wage fixing in the United Kingdom
cotton manufacturing industry. This is, in one sense, the most rigidly
centralised of all British bargaining systems for private industry, in that
its central agreements are extremely comprehensive and are backed by a
high collective discipline; they have also gone farthest in the attempt to
relate individual wage rates to scientific methods of job evaluation,
workload assessment, etc. On the other hand, they give an immense scope
for local negotiation; this arises both in the adaptation of central agreements to special conditions and contingencies which affect earnings, and
in the fixing of wages and working conditions under new technical
methods which have not been so widely adopted as to make a firm basis
for central agreement possible. In addition, some sections have used
the scope for local negotiation in another way; that is to adjust central
agreements to the conditions of narrowly local labour markets.
More recently, a series of agreements in Italian industry established
definite and complementary roles for economy-wide, industry-wide and
workplace bargaining. And in September 1963 the I.L.O. Iron and Steel
Committee suggested a similar approach to the problems of multi-level
bargaining in the following terms:
Where collective agreements are negotiated at different levels the parties
would be well advised to make sure that there is harmonisation between collective negotiations at different levels.... One method by which such harmonisation
may be achieved might be the conclusion... of a general agreement setting the

STEPS TOWARDS STABILITY

165

framework for negotiations to be conducted at a lower level. Another method
might be the conclusion at different levels of separate agreements dealing with
different matters.1
At the time of writing the Metal Workers' Union in the Federal
Republic of Germany was seeking a complete overhaul of wage bargaining methods, so that industry-level agreements would provide for
such things as adequate job classification on the basis of job evaluation,
and application of sound methods of rate fixing where payment-byresults systems are used. Much of the actual job evaluation and rate
fixing would continue to be undertaken at the workplace, but within a
framework and according to rules set by industry-wide agreements.
Such adjustments in the structure of collective agreements to the
technical and economic changes that have taken place during the last few
decades would help in increasing the effectiveness of wage policies. In
particular, they would make trade union restraint more meaningful. In
present conditions the twofold system of wage fixing means in effect that
any such restraint at the first level creates scope for unintended wage
increases at the second level, thus defeating the object of the restraint.
In some European countries such adjustments would involve a
greater role of trade unions in plant-level negotiation (which in certain
cases might raise problems with regard to existing legal workers' councils)
as well as a restriction of the freedom to undertake wage fixing at the plant
level. These two implications might be viewed with disfavour by some
employers 2 as well as by members of works councils. But they might find
compensating advantages in a more rationalised system of wage fixing.
Piece work has come to be regarded as a major source of wage drift 3
and its advantages in terms of productivity improvement are now viewed
with some reservation. It is not obviously true—indeed it may be largely
untrue—that the pace of work in plants or industries where payment-byresults is applied exceeds that where time rates are the rule. Piece workers
1

See I.L.O. : Official Bulletin, Vol. XLVI, No. 4, Oct. 1963, pp. 543-544. In the steel
industries of some European countries such multi-level bargaining exists already and
has been found beneficial also from the point of view of industrial peace: " it is interesting to note that the bargaining machinery in steel in several of the low strike experience
countries minimises the potential for conflict growing out of all-encompassing top-side
negotiations. The ' hierarchy' of bargaining levels involving the co-ordination of national
or industry-wide, district or regional, and local terms and conditions of employment
tends to reserve for each bargaining layer those issues about which the bargaining parties are best equipped to negotiate. The apparent prerequisites for such split-level bargaining involve either strong unions with effective internal discipline and co-ordinated
bargaining with equally effective employer associations and individual employers, or
splintered and relatively ineffectual local bargaining units which tend generally to be
dominated by strong employers and employer groups. " (See Collective Bargaining in the
Basic Steel Industry, op. cit., p. 317.)
a

MARKMANN, op.

See above, p. 137.

cit.

166

PRICES, WAGES, AND INCOMES POLICIES

have often been found to limit their rate of output to what is necessary to
maintain a certain level of earnings by working at a normal level of effort
rather than to aim at maximum earnings by making especially great
efforts. Also, piece work may be a source: of resistance to improved work
methods when workers fear that a new job at a new rate may not yield
the same earnings at the customary pace of work; efforts to overcome
such resistance then tend to take the form of generous rate fixing and
thus to cause further wage drift. But in any case changing technology has,
in several industries, reduced the workers' control over the rate of output,
thus undermining the whole basis of piece work.
In spite of this, piece work tends to be maintained (although since
1955 it has, of course, been drastically reduced in eastern Europe), often
on a fairly unsystematic basis of work measurement. Therefore,
rationalisation of the wages structure may well call for abandoning piece
work in many cases and for much more systematic and accurate work
study and rate fixing in others. In some industries and occupations a
possible compromise between piece work and time work which
retains some of the incentive features of the former whilst reducing its
inflationary potential may be found appropriate. This is the system of
" work-measured time rates ", under which the workers concerned may
choose between two or three different levels of effort or output, for each
of which there is a specified time wage.
PUBLIC INFLUENCE OVER WAGE FIXING

Three points about the influence of the public on wagefixingmay be
mentioned: first, the possibilities of improving statistical and other data
so as to provide a sound factual basis for policy decisions; second, the
possibilities of using government policy towards the incomes of public
employees as a pattern-setting instrument; and third, the importance of
timely initiatives in exploring areas of agreement.
The improvement of statistical and other basic data would not exert
any deliberate public influence over wage fixing, but its importance
should not be underrated. To establish facts beyond dispute may help
greatly in securing agreement on policies. Discussions may much more
easily break down when there is room for legitimate doubt and disagreement on such factual questions as whether wages in a certain industry or
occupation have risen more or less rapidly than other wages or than the
cost of living, or how far profits in the industry have risen or fallen, or
how they compare with other profits.
In many countries the available statistics, even of a general kind, are
very inadequate; and even where general statistics exist these are often

STEPS TOWARDS STABILITY

167

liable to ambiguous interpretation, and are frequently in the form of
general averages only. If one is negotiating on the basis of living costs,
what is relevant is not an average change in retail prices but the change in
consumption prices for workers at the wage level actually concerned.
Similarly, arguments referring to wage relativities often have to be
conducted in relation to either a general average (so that they proceed
on the implied basis that nobody should get either more or less) or
scattered individual wage rates or earnings, the representative quality of
which is dubious and the relevance to the wages actually under negotiation indeterminate. Some countries do not have statistics for both formal
wage rates and earnings and, for those which do, the information
available is quite inadequate to determine which is the more realistic
reference, or the connection between the two. Indices of profits are
particularly ambiguous; there is probably a need for some standard
system of accounting and publication which could be applied to the
profits of individual industries and concerns, and certainly a need for
•clarification of the alternative profit indices available for bargaining
purposes. The confusion of physical and economic productivity is a
continual barrier to sound bargaining reference.
Even in the United States, a country with highly developed labour
statistics—
with the possible exception of basic steel, there are probably no major industries... for which reliable and commonly accepted industry data on hourly
•employment costs exist (and, in view of the apparent variation in hourly nonwage costs among competing firms, the necessary task of obtaining such data
from employers will not be easy). Even where such data might be accepted, it
is frequently difficult to achieve consensus on estimates of the impact of proposed changes in wage rates on average hourly earnings and non-wage benefits,
•or as in the 1962 steel negotiations, of changes in benefit levels and programs
on employment costs.1
Apart from efforts to make more and better statistics available on a
•continuing basis, there may also be considerable value in having tripartite
•councils or independent expert committees examine detailed problems
of wage fixing and wage policy—for instance, problems of wage and
price structure, relations between productivity and wages in different
industries and under different systems of wage payment, questions of job
evaluation, and so on. In the United Kingdom the National Incomes
Commission has on some occasions developed explicit principles of wage
1
Lloyd ULMAN: The Labour Policy of the Kennedy Administration [mimeographed],
1962, pp. 16 and 17. See also MASSÉ: Rapport, op. cit., p. 25, recommending improved
-working relations between the statistical services and industry with a view to providing
;a sound factual basis for policy discussions.

168

PRICES, WAGES, AND INCOMES POLICIES

and salary determination, notably in its report on remuneration for university teachers.1 In the United States the tripartite Advisory Committee
on Labor-Management Policy, and in various European countries tripartite social and economic councils or committees thereof, might also be in
a position to undertake useful studies of this kind.
Another possibib'ty of direct public intervention short of general wage
control is that of using the wages of public employees as a pattern-setting
instrument. The most general formal attitude of governments to wages
in public services appears to have been that these should follow those paid
in private employment. The reasons for this are fairly straightforward
—that since public authorities make no profits there is no genuine market
determination of capacity to pay, that this policy preserves governmental
neutrality in industrial issues and prevents public pay becoming a votecatching device.
In several countries there seems already, however, a difficulty in
holding this position in face of the extent to which the government, local
authorities, public corporations or other public bodies are responsible
for employment. This is particularly a problem where the public has
become the owner of industries which were formerly private and have
strong trade unions and independent traditions of wage settlement. There
are also certain new industries, like atomic energy development, which
are products of public enterprise from the start, but which employ
workers with whom no direct private comparison is possible. In several
countries, at any rate, public servants of one sort or another have played
a critical role in recent wage movements, and the government has been
obliged to take some responsibility for the result.
If civil servants or other public employees are made the subject of a
pay pause they may be doubly dissatisfied if private industry fails to toe
the line and they fall behind other workers.
Nevertheless, if the government really believes that the national
economy will be endangered if wages rise more than a certain amount,
to demand that it should not apply its belief to the sectors for which it
has a direct responsibility seems incongruous.
One possible approach would be for the government to announce
some basic figure of economically tolerable increases in the public sector's
total wage bill during a reference period, say one year. If this was, for
example, 4 per cent, it might be reasonable to offer an increase averaging
perhaps 3 per cent, to most public servants, the distribution of this to
be negotiated between the government and the civil service unions,
1
United Kingdom: Remuneration of Academic Staffiti Universities and Colleges of
Advanced Technology, Cmnd. 2317 (London, 1964).

STEPS TOWARDS STABILITY

169

1 per cent, being kept back for special pay increases to certain named
categories of workers in short supply.1
As argued earlier, it will be difficult to extend to the private sector the
use of criteria for wage advances enployed in the public sector, unless
there is a consensus that such criteria are generally acceptable. However,
few if any recognised pattern-setters in wagefixingin the private sector
in any economy have as many employees as the government, and the effects
of an example set by it may be considerable. This may be particularly so
when more detailed problems of wage structure or the exploration of
new forms and systems of wage determination are concerned. For
instance, in France the agreements of 1955 in the nationalised Renault
works, which contained several innovations as regards remuneration and
fringe benefits, were followed in several other French firms, in the metal
trades as well as in other industries.2 In the United Kingdom the Treasury's
Civil Service Pay Research Unit, which undertakes investigations with
a view to comparing representative civil service jobs with more or less
equivalent jobs in industry, has been rendering services to public wage
negotiators which might well be copied by private bargainers if systematic
comparison of jobs came to be recognised as a useful basic principle of
wage determination.
Provision (varying from country to country) is often made for the
submission of declared or threatened disputes in the private sector to
voluntary arbitration in which the government plays some part. Attention has already been drawn to the danger of increasing industrial
disputes and unrest if such machinery is used in a way that undermines
faith in the independence of arbitrators and their freedom to make
awards on the merits of the cases before them. But it would not seem
unreasonable that the reports of any arbitration courts in whose establishment the government plays any part should be obliged to lay the plain
truth before the public about (1) the percentage increase in average
earnings that their awards would bring about and (2) the length of time
for which a new award would have to last if the average increase per
annum granted under it was not to exceed the average wage increase
currently being granted in the government's own public service wage
awards, and to make (3) some estimate of the current level of unfilled
vacancies in the industry concerned, both by grades and by regions. If it
became mandatory or customary for public arbitration awards to contain
these details, there would no doubt be a demand that the awards of
private industrial aribtrators should include similar details.3
1
2
3

See The Economist, 21 Oct. 1961, p. 205.
" Works Agreements of the ' Renault Type ' ", op. cit., pp. 219 ff.
See The Economist, loc. cit.

170

PRICES, WAGES, AND INCOMES POLICIES

Finally, if there can be no real incomes policy without a consensus
that this is desirable, and a willingness to explore the possibilities of
making it work, such a consensus, in turn, is likely to depend on timely
and imaginative government initiatives, leading to careful exploration
of areas in which agreement and co-operation may be possible, and
backed, it may be, by clear statements of the measures, fiscal or other,
that the government proposes to take if it is unable to secure the agreement and co-operation needed for a common approach to the problems.
Possibilities of securing a consensus are also likely to depend on the
rate of growth of the economy in relation to the rate of wage drift. If the
rate of growth does not exceed, say, 3 per cent, and wage drift occurs at
the rate of 2 per cent, per annum, then unless prices are to rise or wages
to increase faster than other incomes there will be room for an average
increase of only about 1 per cent, in wage rates. It is very difficult to
imagine that a trade union movement could agree to limit wage demands
to this extent except in conditions of national emergency. But with a
growth rate of 4 per cent, or more and a wage drift of iy2 per cent, or
less there would evidently be more scope for agreement. True, the rate
of wage drift may not be entirely independent of the rate of growth.
In any case, however, a major difficulty of a national wage and incomes
policy lies in reconciling the various pressures that are likely to be put
upon the agencies responsible for the policy's formulation and implementation. Clearly, the bigger the annual increment to the total of resources
available, the more room there will be for such conflicts to be resolved.
Thus problems of incomes policy, like so many other social problems,
are likely to be more easily resolved in a rapidly growing economy, and
incomes policy itself is likely to be the more effective the more it is
designed to foster that condition.

INDEX
A
Alexander, Sydney, 65«
Ando, Albert, 39«
Australia, 108, 122
Austria, 101, 102
B
Bach, G. L., 39«
Ball, J. R., 85«
Baumol, W. J., 62«
Bayliss, F. J., Ill«
Bhatia, R. J., 86«
Bowen, W. G., 86«
Bronfenbrenner, M., 57«
Browne, M. H., 50«
Brownlee, O., 40«
C
Collective bargaining at plant level, 161165
Conrad, A., 40«
D
Denmark, 97
Dessau, J., 97«
Dicks-Mireaux, K. A., 82«
Directors' fees and emoluments, 155
Dividend freeze, 152-153
Dividend taxation, 154-155
Dow, J. C. R., 80«, 82«, 88«
Dunlop, J. T., 49«, 64«, 96n
Dunn, Peter, 139«
E
Eckstein, O, 18«
Economicfluctuationsin post-war period,
8-10
Economic stability, definition, 5-6
and wages, 7-27
Economist, The, 22«, 155«, 169«
Eicher, H., 38«
F
Flanders, Allan, 134«
France, 38, 98, 152, 169
G
Galbraith, J. K., 54«, 66«
Garbarino, J. W., 114«
General Agreement on Tariffs and Trade,
61«

Germany (Federal Republic), 38, 94
Metal Workers' Union, 165
Gern, J. P., 119«
Godley, W. A. H., 83«
Government exhortation, 94
Guillebaud, C. W., 123
H
Hansen, Bent, 118«
Holzmann, Franklin D., 40«, 57«
Houthakker, H. S., 37«
I
Income distribution, 130-131
Inflation, brief history, 10-13
causes of, 56-76
effects of, 28-55
mechanics of, 77-91
I.L.O., 164, 165«
Investment, factors stimulating, 23
J
Johnston, T. L., 101«
K
Killingsworth, Charles C , 134«
Klein, L. R., 85«
Kuhn, Alfred, 73«
L
Labour, excess demand for, 84
Labour market, effect of shortages in, 24
Labour productivity, concepts of, 115
Labour statistics, need for improvement
of, 166-168
Lanzilotti, Robert K., 63«
Lerner, A. P., 55«, 124«
Levinson, Harold M., 85«
Loans, index-linked, 47
Lytton, H. D., 31«
M
Mack, Ruth P., 65«
Marchai, Jean, 64«
Marketing, strategy of retailers, 66
Markmann, H., 160«, 165«
Massé, Pierre, 49«, 98«, 131«
N
Netherlands, 38, 97, 125, 130, 148, 151
New Zealand, Court of Arbitration, 109

172

PRICES, WAGES, AND INCOMES POLICIES

Non-wage incomes, policy towards, 149156
O
Organisation for Economic Co-operation
and Development 92n, 149«
Organisation for European Economic
Co-operation, 31«
Output growth and price changes, 49-51
Pensions, 37
Pesek, B. P., 40«
Phelps-Brown, H., 32«, 136«
Population changes, 22
Price control and income restraint, 156
Price fixing in retail trade, 68
Price freezing and income restraint, 152
Price indices, movements in, 29
Prices, consumer, 9-10, 12, 16-17
mark-up, 66
rising, see Inflation
Productivity, need for growth of, 135
and economic organisation, 24-25
Profit, maximisation of, 63
regulation and income restraint, 150,
154
Profit-sharing by employees, 160
Public influence over wage fixing, 166-170
Quandt, R. E., 30«
R
Recessions, post-war, 13-19
Reddaway, W. B., 74«
Rent control, effects of, 151
Reuther, Walter, 162
Ruggles, Richard and Nancy D., 31«
Savings, trends in, 41-47
Schultz, C. L., 57«, 65«, 73«
Sellier, F., 95«
Shepherd, J. H., 83«

Smithies, A., 54«
Social security, 36-38
Steindl, J., 116«
Sweden, 100

Tavlor, G. W., 69«
Thorp, Willard L., 30«
Trade cycle, 26-27
Trade unions, 71-72, 130, 143-144, 157159
Turner, H. A., 136«
Tyndall, Sir Arthur, 109»
U
Ulman, Lloyd, 167«
Unemployment statistics, 81
United Kingdom, 37, 92-93, 98-99, 121
United Nations, 24«, 74«
Economic Commission for Europe,
64«
United States, 37«, 38, 47«, 96, 113, 125,
167
V
Vermeulen, A. 160«
W

Wage drift, 20, 136-142
Wage freeze, effects of, 135
Wage increases, distribution of, 120
Wages, and the demand for labour, 75-76
and economic stability, 5-27
maintenance of real, 119
policy, criteria for adjustments, 112-128
methods and institutions, 92-111
problems of, 129-148
processes of fixing, 69-70
reform of structure, problems of, 140
sliding scales of, 119-120
trends in, 16-22
Weidenbaum, M. L., 60«
Workers' participation in profits and
investments, 159-161