Preview

mW
INTERNATIONAL

LABOUR

OFFICE

JOB EVALUATION

GENEVA
1960

STUDIES AND REPORTS
New Series, No. 56
First Printed: May 1960
Second Impression: April 1964
Third Impression : April 1966
Fourth Impression: June 1967

PRINTED BY "COURRIER DE GENÈVE", GENEVA, SWITZERLAND

CONTENTS
Page
INTRODUCTION

CHAPTER I: Some General Considerations

Job Evaluation Defined
Reasons for the Increased Interest in Job Evaluation
The Wage Structure in the Plant
External Factors
Internal Factors
Job Evaluation and the Wage Structure
Job Evaluation in Planned Economies
Job Evaluation and Collective Bargaining
CHAPTER II : Methods of Job Evaluation
Non-Analytical Methods
The Ranking Method
Description
Advantages and Limitations
The Grade or Classification Method
Description
Advantages and Limitations
Analytical Methods
The Factor Comparison Method
Description
Advantages and Limitations
The Point Rating Method
Description
Advantages and Limitations
Examples of Some of the Elements Involved in Establishing a Point
Rating System
Job Description and Analysis
Procedures
CHAPTER III : Some Problems of Analytical Job Evaluation
Technical and Economic Aspects of Wage Determination
Factors, Degrees and Weights
Differences in Factors and Weighting
Differences between Groups of Jobs
Application of Uniform Schemes to Broad Economic Sectors . . . .

8
9
11
11
13
15
16
17
20
21
21
21
22
22
22
23
26
26
26
30
32
32
33
34
39
41
46
46
50
50
56
58

JOB EVALUATION
Page

Problems of Rating
The Determination of Relative Wage Rates
Points and Wage Rates
Classification of Jobs into Grades
Introducing the New Wage Structure
Introduction of Job Evaluation in British Coal Mines
Job Grades and Rate Ranges
Job Evaluation and Incentive Wages
The Problem of " Normal Performance "
Fluctuations in Earnings

62
64
65
67
69
73
75
77
78
82

CHAPTER IV: Some Experiences with Job Evaluation Schemes

Survey of Experience in 73 United States Companies
Surveys of Experience in 96 Canadian Companies and 75 United States
Companies
Survey of Experience of 23 United States Aircraft Manufacturing Companies
Survey of Experience in the United States Pacific Coast Aircraft Manufacturing Industry
Study of the Distribution of Employees in Six Southern California Airframe
Manufacturing Plants
Study of Job Evaluation in the West German Steel Industry

CHAPTER V : Advantages and Limitations of Job Evaluation
Advantages
Limitations
Attitudes towards and Prospects for Job Evaluation
Employers
Trade Unions
Negative Views
Favourable Views
Some Problems Involved in Trade Union Acceptance of Job Evaluation
Some Conditions for Trade Union Participation in Job Evaluation . .
Conclusion

85
88
89
92
92
92

99
99
100
102
102
102
103
106
108
110
112

APPENDIX A: I.L.O. Metal Trades Committee (Sixth Session, Geneva, May
1957): Extracts from the Report of the Subcommittee on Job Evaluation
Methods in the Metal Trades
117
APPENDIX B: International Committee for Scientific Management: Extracts
from the Report of the International Conference on Job Evaluation (Geneva,
May 1950)
127
APPENDIX C: Some Major Features of Job Evaluation Schemes Applied in
Various Countries

133

CONTENTS

LIST OF TABLES
Page

Text
1. Ranking Jobs by Factors under the Factor Comparison Method . . . .
2. Allocation of Money Values to the Different Factors and Ranking of Jobs
under the Factor Comparison Method
3. Comparison of Ranking by Factors and by Money Values under the Factor
Comparison Method
4. Point Values Allotted under N.E.M.A. Point Rating System
5. Rating of Visual Requirements Factor under Belgian " Alphanumerical "
Method
6. Comparative Ranking of Jobs under Steel Industry Plan and Light Industry
Plan in the United States
7. Job Grades and Point Values under the N.E.M.A. and National Metal
Trades Association Schemes in the United States
8. Distribution of Employees by Labour Grade in Six Airframe Plants in
Southern California
9. Effects of the Introduction of Job Evaluation on the Structure of Wages and
Earnings in Six West German Steel Plants

28
29
29
36
38
52
71
94
95

Appendix C
A. Belgium: Factors Used in the Job Classification Scheme Drawn up by the
General Technical Commission
B. France: Factors and Points under the Job Evaluation Scheme for Manual
Workers at the Télémécanique électrique Plant, Nanterre
C. France: Points for Responsibility for Safety of Others under the Job
Evaluation Scheme for All Employees at the Télémécanique électrique Plant, Nanterre
D. German Democratic Republic: Suggested Point Values for Different
Degrees of Knowledge and Experience
E. German Democratic Republic: Suggested Scheme of Factors, Degrees and
Point Values
F. Federal Republic of Germany : Criteria for the Allotment of Points to the
Different Factors under the Volkswagen Scheme
G. Federal Republic of Germany: Factors and Range of Points under the
Volkswagen Scheme
H. Federal Republic of Germany: Relation between Job Evaluation Points
and Wages at the Volkswagen Plant
I. Federal Republic of Germany: Factors Used and Maximum Point Score
Allotted to Each under the Euler-Stevens Method
J. India: Factors and Points under a Job Evaluation Scheme Used in Several
Plants
K. India: Points for Various Jobs under a Job Evaluation Scheme Used in
Several Plants
L. Netherlands: Factors, Point Values and Weights under the Standardised
Method
M. Sweden: Factors and Points under the Metal Trades Scheme
N. Sweden : Classification of Jobs by Degree of Physical Effort under the Metal
Trades Scheme

134
135
136
137
137
138
139
139
139
140
140
141
142
142

JOB EVALUATION
Page

O. United Kingdom: Factors and Points under a Job Evaluation Scheme Used
by a Large Electrical Engineering Company
143
P. United Kingdom: Factors and Weights under the Imperial Chemical Industries Scheme
144
Q. United States : Factors and Maximum Points or Weights under the Southern
California Aircraft Industry, General Electric, Westinghouse and U.S.
Steel Schemes
146
LIST OF CHARTS
I. Typical Stepped Curve Relating Point Values Assessed by Job Evaluation to
the Wage Scale
II. Wage Curve Showing a Proposed Wage Scale and Existing Wage Rates
in a Plant
III. Distribution of the Labour Force in a United States Aircraft Plant
Employing 20,000 Workers
IV. Distribution of the Labour Force in a United States Aircraft Plant
Employing 13,000 Workers

68
69
77
78

INTRODUCTION

The problems of defining and fixing suitable wage relationships
between groups of jobs and workers are of considerable importance in
both the theory and the practice of wage determination. In the process of
collective bargaining, employers and trade unions often base their
position on comparisons with wages or wage changes in other firms and
industries, and it is not uncommon for arbitration awards and decisions
of minimum-wage-fixing authorities to refer to wage adjustments in other
parts of the economy as an important factor in determining a proper or
fair wage adjustment for the workers concerned. Sometimes such
comparisons are made with a view to ensuring equal pay for roughly
equal or, by some standard, equivalent work, sometimes in order to
maintain or create wage differentials that are regarded as justified either
on merely traditional grounds or for more substantial reasons.
The technique ofjob evaluation has been applied to deal with problems
of comparative equity in wage determination mainly, though not
exclusively, at the level of the individual firm or plant. Its aim is to
provide a means of establishing a wage structure that is acceptable to
labour and management and that, as far as possible, is fair or equitable
in the sense of ensuring equal pay for jobs demanding broadly equivalent
sacrifices from their incumbents and of rewarding properly the greater
efforts and hardships involved in some jobs as compared with others.
In this way the method is believed to be helpful in reducing dissatisfaction so often arising from existing wage differentials (or, as the
case may be, from their absence), and thus to contribute to more harmonious human relations in industry.
Job evaluation is a relatively new technique; it has been applied
rather widely in some industries and in some countries but very little
in others; although its basic philosophy is fairly simple, practical application often involves rather complex technical questions; finally, the method
has been criticised on a variety of grounds. For all these reasons it has
been thought useful to bring together in a relatively small volume a brief
description of its aims and methods, some concrete examples and information based on actual experience with j ob evaluation schemes, and a statement
and brief discussion of some of the problems and criticisms that have arisen
in connection with the use ofjob evaluation as an aidin wage determination.

JOB EVALUATION

An effort has been made to view the subject against the broader
background of modern thinking and practice with regard to the wagenegotiating process as a whole. This, it is hoped, will help to illustrate
the significance both of the positive contribution that job evaluation can
make to wage determination and of the limits within which it has to work.
One disadvantage of attempting to discuss the matter in this somewhat
broader framework is that some of the technical details involved in
establishing and operating a job evaluation system can only be discussed very briefly and others only mentioned. On the other hand,
this study is not a technical handbook for the training of job evaluation
technicians; it is at best a contribution to a wider understanding, on the
part especially of employers and trade unionists, of the meaning and
general principles of the method.
The scope of the study is restricted in another direction : although
the principles of the method are applicable to non-manual as well as
to manual jobs (in fact several of the earliest applications ofjob evaluation
related especially to clerical functions), the examples and general discussion in this report are mainly in terms of the latter. The practical
problems of analysing and evaluating salaried jobs, especially senior
and executive positions, are somewhat different from, and perhaps more
complex than, those arising in connection with manual jobs. Full
justice could not have been done to these problems without substantially
increasing the size of the study.
In the first two chapters the method and the problems it seeks to
solve are defined and the general nature of the technical steps involved
are described. In Chapter III some of the theoretical and practical
problems encountered in the application of job evaluation systems are
mentioned and briefly discussed. This chapter—particularly the latter
part—thus constitutes a further elaboration of the technical discussion
in Chapter II. In Chapter IV several studies undertaken in this field are
summarised, and some of the conclusions that have been drawn from these
studies are given. Chapter V reviews the advantages and limitations of
job evaluation, as well as a number of opinions and attitudes toward job
evaluation, especially those voiced by various trade union representatives.
As an illustration of the present state of the debate on the use of job
evaluation in industry, and also as a further source of information on
experiences gained in thisfield,the study is followed by extracts from the
report of the subcommittee on job evaluation methods in the metal
trades established on the occasion of the Sixth Session of the I.L.O. Metal
Trades Committee at Geneva in May 1957.1 These are embodied in
1
Omitted from the subcommittee's report are those parts that deal only with
procedural questions.

INTRODUCTION

Appendix A. Appendix B reproduces the report of the International
Conference on Job Evaluation held under the auspices of the International
Committee for Scientific Management in 1950; several of the questions
raised and discussed at that meeting are still of considerable interest.
Finally, Appendix C describes some of the major features of various
systems that have been applied in industry. Several of the examples
given are taken from the metal trades (in which the method has up till
now been applied most frequently) and are described in the report
submitted by the International Labour Office to the above-mentioned
session of the Metal Trades Committee, which has also been drawn
upon in various parts of the main text of the present study.1

I.L.O. : Job Evaluation Methods in the Metal Trades, Report III, Metal Trades
Committee, Sixth Session, Geneva, 1957 (mimeographed).

CHAPTER I
SOME GENERAL CONSIDERATIONS
Wages are probably the most important single element in conditions
of employment. Accordingly, wage rates and conditions governing
workers' earnings have always raised major issues in negotiations
between employers and workers. The questions examined in the following pages relate to wage differences, or rather to a particular variety of
wage differences.
Such differences occur at several levels. They exist as between
countries and, within each country, as between industries or groups of
industries. Moreover, within any industry there are usually wage differences as between individual regions, firms or plants and within the latter
as between different departments. Many of these differences accompany
and reflect differences in occupations or jobs—quite apart from the fact
that workers engaged on the same job may be paid different wages according to their sex, age or length of service. Finally, individual workers
in any occupation or job may earn different wages, for example according
to differences in personal performance or in working conditions. This
will happen when systems of payment by results are applied or when
special allowances are paid, e.g. for heavy, dangerous, night, shift or
overtime work.
The wage differences with which this study is more particularly
concerned are those between individual—and especially manual—jobs
within a given unit in which common rules govern the fixing of wages
for individual jobs or groups of jobs. The unit may be an individual
firm or plant, but in other cases may cover a whole industry for which
a national collective agreement exists. In exceptional cases, for example
in the post-war economy of the Netherlands, common rules are applied
to even larger groups of workers. The problems to be examined are
mainly concerned with this question : when wages are fixed for a plant,
firm or industry, how are the wage differences established as between
individual jobs and why ?
This in itself is an interesting problem of scientific research. However,
it also represents an eminently practical aspect of wage negotiation and
administration. Generally, wages for individual jobs are determined by

JOB EVALUATION

the employer (unilaterally or after some measure of negotiation with
individual workers) or by collective bargaining. In the former case
wage fixing is a management decision to be adapted to the general
requirements of work organisation and working methods as management
sees them. In the case of collective negotiation the employer will aim
at achieving a result as nearly approaching his views on appropriate
rates as the bargaining permits, while the union will also seek to obtain
that structure of rates which it regards as desirable. An understanding of
the meaning and functions of wage differentials as between jobs should
help in clarifying either side's interest in, and views of, the matter, and
may make mutual agreement easier to achieve.
These aspects have become important with the development of
large manufacturing, transport and service (for example, banking)
enterprises. Such undertakings are now generally recognised to require
an established wage system, if only as an administrative device. Employing large numbers of workers, they have to have some rules for paying
them. Otherwise, estimating and controlling costs would be a very
difficult task. Nor could thefirmbe assured that it could recruit and retain
the manpower it needs. Again, the administrative work of calculating
people's wages would be very complex and difficult.
The fact that many rates are fixed for jobs (and not separately for
each individual worker) in itself corresponds to an important rule. Its
administrative advantages are obvious, but the rule also reflects a more
important feature of modern industrial work. Because of division of
labour and specialisation, any large enterprise may have hundreds of
different jobs, each of which may be performed by a substantial number
of workers. Many people work together on the same, similar or technically interdependent jobs. Their work must be carefully planned and
closely co-ordinated. One result of this is that individual jobs tend to be
precisely defined. Secondly, workers whose jobs are interdependent
must work according to uniform schedules of working hours and observe
other common rules. Persons working under these conditions and
performing identical tasks could hardly be required to work for different
rates of pay. Collective bargaining has accordingly resulted in widespread application of the principle of " the rate for the job " as distinct
from ratesfixedby the employer on an individual basis. But the principle
is now widely recognised to be a rule of good management as well.
A further important aspect of modern large-scale undertakings is
that many jobs are semi-skilled, their performance requiring relatively
little training. In the days when production took place in small units
with master craftsmen, journeymen and apprentices, the question of
relative wages within a firm could be settled by distinguishing wages

SOME GENERAL CONSIDERATIONS

for " skilled ", " semi-skilled " and " unskilled " work. This is still the
basic principle applied in many modern industries. For very large
numbers of industrial workers, however, jobs are no longer differentiated
mainly by the degree of skill required, and in any present-day undertaking a good many may in fact fall within a narrow range of comparatively limited skills. Meanwhile, differences in other job characteristics, e.g. responsibility, effort required and nature of the work, have
assumed greater importance.1 For such jobs, there is clearly slight
justification for paying different wages according to skill requirements
only, and it is understandable that other characteristics should generally
be regarded as a more important basis for differences in wages. As a
result, while skilled craftsmen may be said to be paid in accordance
with what they are able to do, semi-skilled and unskilled workers tend
to be paid according to what they are actually doing.2
The fact that people work together in similar or at least comparable
jobs, and the fact that formally and precisely described jobs may change
when products or the methods of making them are altered, tend to
emphasise differences in job contents and the belief that these should be
reflected in differences in wages. As a result, both employers and trade
union leaders often have to spend a great deal of their time settling claims
and disputes arising from the absence of such differences when workers
believe them to be necessary or, conversely, from the existence of wage
differences which the workers consider unjustified. A wage structure
based exclusively on differences in skill (as most basic wage structures
are) cannot easily cope with this situation. In practice, excessively simple
wage structures have therefore been corrected to some extent, for
example by increasing the number of classes of skill and thus ensuring a
finer correlation between jobs and wages. Thus, in post-war French
industry it has been customary to distinguishfivegroups of jobs: common
labour, specialised labour, and semi-skilled, skilled and highly skilled
work.
In addition, compensation for various types of job deterrents has in
many cases been provided for, e.g. in the form of allowances on account
of dirty, dangerous or heavy work. Such devices have, however, often
been found inadequate, especially when applied on an ad hoc, if not
totally haphazard, basis without much systematic analysis or co-ordination. The wage structure in the United States steel industry before the
1
The fact that a price has to be paid for the danger or unpleasantness of a job
has, incidentally, the advantage of providing an incentive for management to eliminateor reduce such undesirable job features.
8

G. W. TAYLOR, in G. W. TAYLOR and F. C. PIERSON (Edited by): New Concepts

in Wage Determination (New York, London and Toronto, McGraw-Hill, 1957), p. 93.

JOB EVALUATION

days of collective bargaining has been described as " chaotic ", wages
being fixed largely on the basis of the opinions of individual foremen
without relation to wages prevailing in other departments of the same
plant. 1 In other cases an accumulation of general and individual allowances and wage adjustments had resulted in a wage structure so confused
that further adjustments could no longer be made in any rational way.
The wage structure in British coal mining before April 1955 has been
described as so complicated that nobody could tell how many different
occupations there were in the industry nor how wage rates were distributed in each of them. As a result, wage adjustments usually took the
form of uniform flat-rate increases ; in this way the cost of wage increases
could be determined simply by reference to the total number of workers.2
The technique known as job evaluation aims at providing a more
systematic and objective basis for the comparison of job contents as an
aid in the establishment of a more rational wage structure. The question
as to what may be regarded as a " rational wage structure " will be
discussed further on.
JOB EVALUATION DEFINED

Job evaluation may be defined simply as an attempt to determine and
compare the demands which the normal performance of particular jobs
makes on normal workers without taking account of the individual
abilities or performance of the workers concerned. Individual abilities
or efforts may, of course, also be taken into account and reflected in the
workers' earnings, for example under a system of payment by results or
of merit rating, but this is something entirely different from the rating
of the job. Job evaluation rates the job, not the man. As will be seen
further on, the existence of systems of payment by results creates some
difficult problems in the implementation of job evaluation and in certain
cases may defeat its whole purpose.
The following are other typical definitions or descriptions of job
evaluation :
Job evaluation is the process of analysis and assessment of jobs to ascertain
reliably their relative worth, using the assessments as a basis for a balanced
wage structure.3
1
L. G. REYNOLDS and Cynthia H. TAFT: The Evolution of Wage Structure
(New Haven, Conn., Yale University Press, 1956), p. 45. See also J. STIEBER: The
Steel Industry Wage Structure (Cambridge, Mass., Harvard University Press, 1959),
pp. 3-4.
2
W. H. SALES and J. L. D A VIES: " Introducing a New Wage Structure into Coal
Mining", in Bulletin of the Oxford University Institute of Statistics, Vol. 19, No. 1,
Aug. 1957, pp. 201 ff.
3
British Institute of Management : Job Evaluation. A Practical Guide, Personnel
Management Series 4 (London, 1951), p. 77.

SOME GENERAL CONSIDERATIONS

Job evaluation is a method which helps to establish a justified rank order
of jobs as a whole, being a foundation for the setting of wages. Job evaluation,
therefore, is only one of the starting points for establishing the relative differentiation of base wage rates.1
Job evaluation is the evaluation or rating of jobs to determine their position
in a job hierarchy. The evaluation may be achieved through the assignment of
points or the use of some other systematic rating method for essential job
requirements, such as skill, experience and responsibility. Job evaluation is
widely used in the establishment of wage rate structures and in the elimination
of wage inequalities. It is always applied to jobs rather than the qualities of
individuals in the jobs.2
The aim of the majority of systems of job evaluation is to establish, on an
agreed logical basis, the relative values of different jobs in a given plant or
industry. Job evaluation does not, of course, take the place of the established
procedures for bargaining between the representatives of the employers and
workers; but the use of a system of job evaluation may facilitate the development and maintenance of an equitable relationship among the rates for different
jobs. The principle upon which all job evaluation schemes are based is that of
describing and assessing the value of all jobs in the firms in terms of a number
of factors, the relative importance of which varies from job to job.3
It may be noted that some of these definitions refer to the determination of the relative " value " or " worth " of jobs, while others merely
speak of their " ranking ". The latter type of definition seems preferable.
As will be seen from the description of various systems of job evaluation
given in Chapter II, the real object of investigation and comparison is
the content of the job, and not the rather imprecise notion of its " value "
to the firm. It should also be borne in mind that, although job evaluation
will normally be used as an aid in the fixing of relative wages, its results
are no more than one factor determining the wage structure, and the
actual fixing of wages, through collective bargaining or in other ways, is,
in the view of many authors, entirely beyond the realm of job evaluation.
It may also be noted that the Dutch term for job evaluation (werkclassificatie) and the most frequently used French expression (qualification du travail) do not involve any notion of evaluation.
REASONS FOR THE INCREASED INTEREST IN JOB EVALUATION

The principal cause of the increased interest in job evaluation during
recent years has been mentioned above: under modern conditions of
production it has been found desirable to fix relative wages on the basis
1
Netherlands Committee of Experts for Job Evaluation : Standardised Method of
Job Evaluation (The Hague, Jan. 1951 (mimeographed)), p. 1.
2
United States Department of Labor, Bureau of Labor Statistics: Glossary of
Currently Used Wage Terms, Bulletin 983 (Washington, U.S. Government Printing
Office, 1950), pp. 11-12.
3
I.L.O. : Payment by Results, Studies and Reports, New Series, No. 27 (Geneva,
1951), p. 42.

JOB EVALUATION

of some comparison of job contents. A main reason for this has been
that wage earners, like most other people, have strong feelings about the
" fairness " or " equity " of relative wages, this being judged normally
on the basis of comparative job contents. A widespread opinion on this
matter is, for example, that people doing the same job with equal efficiency should be paid the same wages; as a corollary of this claim for
equal pay for equal work, most people feel that differences in job contents,
in the way of either job requirements or job deterrents, should be reflected
in unequal pay for the jobs concerned. Some authors have combined
these two rules into the proposition that income differentials would be
" just " if they were merely such that persons holding different positions
would riot wish to exchange these positions. Furthermore, there is a
good deal of agreement as to the sort of job characteristics that matter
from this point of view—the things that make a job decidedly unpleasant
as well as the particular job qualifications that " ought to " be specially
rewarded. This broad measure of agreement provides a sound basis for
attempts to fix wage differences acceptable to the workers through
systematic analysis of job contents.
In several cases the application of job evaluation for this purpose has
been stimulated by the war and post-war abnormalities of the employment market, accompanied in some countries by a freezing of wages or
by repeated wage adjustments to changes in the cost of living (which
distorted the wage structure) and frequently by labour shortages,
especially in semi-skilled and skilled jobs. Thus, wage control by the
National War Labor Board was a strong stimulant to the increased
application of job evaluation in the United States because manifest
anomalies and inequities in the wage structure were often admitted as a
ground for approving wage increases, and job evaluation could be used
to demonstrate the existence of such anomalies. Similarly, under the
post-war system of wage control in the Netherlands, job evaluation was
accepted and encouraged by the authorities as a means of making appropriate wage adjustments for particular jobs or industries without upsetting the general objective of maintaining substantial stability in the
national wage level as a whole. In Germany, too, job evaluation received
wider application in the course of the Second World War and during the
first few post-war years, when wage control continued to be applied.
A structure of wage differentials which broadly satisfies workers has
obvious advantages to workers and employers over a structure that does
not do so; it reduces grievances and disputes and generally makes
people feel happier. To the extent that job evaluation can help in
establishing such a wage structure, the method should therefore be
welcomed. The reason why in many cases it is not enthusiastically

SOME GENERAL CONSIDERATIONS

greeted or is even emphatically rejected may be ignorance concerning
its true nature—although in some instances it has been found that
the disadvantages which job evaluation might involve in any particular
case outweighed its advantages.
Where management has taken the initiative in introducing job
evaluation, the technique has naturally become a matter of great interest
to the trade unions. In some cases the latter have been highly distrustful
for a variety of reasons that will be considered later. Many employers
are, however, also sceptical or critical of job evaluation, sometimes
because they prefer to set wages by simple rule-of-thumb methods or
because they find job evaluation too costly and time-consuming and
believe that it does not result in a wage structure differing substantially
from one that could be obtained through less sophisticated methods. The
fact that both employers and trade union leaders are divided among
themselves as to the value of the method makes the subject of job evaluation one of considerable topical interest.

THE WAGE STRUCTURE IN THE PLANT

In this study the term " wage structure " is used primarily to refer
to rates of remuneration for various jobs, primarily within an individual
firm or plant, in relation to each other and to rates paid for certain jobs
in other firms. In what sense can this wage structure be regarded as more
or less rational? Two aspects of this question may be mentioned here.
First, a wage structure will not be considered rational if it conflicts
sharply with accepted opinions as to the " equity " of relative rates. This
aspect is particularly, though not exclusively, relevant to relative rates
within the firm. Secondly, a wage structure will not be considered
rational unless it permits the firm to hire and retain at reasonable cost
suitable workers to fill the various jobs its productive aims call for; this
is largely a question of the economic suitability of the rates paid by the
firm in relation to a complex of external employment market factors.
In order to be considered rational, a wage structure has to satisfy certain
conditions in both respects.

External Factors
As regards the external aspects of a firm's wage structure,
that at least certain rates paid for particular jobs cannot,
speaking, be much lower than those paid for comparable jobs
other firms. The firm would find it impossible to recruit

it is clear
generally
in certain
qualified

JOB EVALUATION

workers; and the low level of wages might create an atmosphere of
discontent resulting in excessive labour turnover, high costs of training
and other unfavourable effects. It is also clear that the firm's wage rates
should not be too high compared with those paid by other employers.
This would probably involve unnecessary expenditure and would not be
regarded as sound management policy.
Although there are, therefore, significant relationships between rates
paid by an individual firm and its external environment, these relationships are to be regarded as setting upper and lower limits beyond which
the firm cannot normally vary its own rates as compared to those paid
by other employers, rather than as dictating the precise rates which the
firm should pay for each of its jobs. In addition, it has been pointed out
that this range of discretion within which a firm canfixits rates relative
to those paid elsewhere is wider for some jobs than for others.1 In so far
as it is determined by the firm's need to recruit and retain suitable
personnel, it depends largely on the actual mobility of wage earners in
response to wage differentials. This is partly a question of mentality.
Thus it is often assumed that North American workers shift more easily
from one occupation to another than European wage earners. But nowhere in the world will workers move to a new job forfinancialreasons
alone unless the difference in pay is large enough to.overcome natural
averseness to change as well as more rational reasons for staying with
one's job, such as seniority rights in respect of promotion and job
security, pension plans, or a conveniently located workplace.
Furthermore, the actual possibility of changing one's employment is
often limited, especially in jobs other than the lowest-paid—not only
because a worker's accumulated knowledge and experience may be of
lesser value to a new employer, but also because of the practice followed
by manyfirmsoffillinghigher-paid jobs by internal promotion rather than
by external recruitment; in other words, the number of " hiring jobs "
tends to be quite hmited, such jobs, moreover, being often concentrated
at the lower end of the pay scale.2 At the same time new entrants to the
employment market, who will normally qualify for hiring jobs, tend to be
attracted by higher rates more easily than workers in established
positions.
These various aspects of labour mobility may help to illustrate how
the relationship between external conditions and the rate that afirmhas
to pay may vary as between different groups of jobs. The relationship
will be closer in the case of hiring jobs than of other jobs. There are,
1

See G. TAYLOR and F. PIERSON, op. cit., particularly the contributions by

J. T. DUNLOP, E. R. LrvERNASH and A. M. Ross, pp. 127 ff., 140 ff. and 174 ff.
2

E. R. LrvERNASH, in TAYLOR and PIERSON, op. cit., p p . 144-145.

SOME GENERAL CONSIDERATIONS

however, often other factors affecting the firm's discretion in fixing
rates relative to those paid on the outside. In times of labour shortage
the range of discretion will tend to be narrower than when there is
substantial unemployment. Again, when there is a collective agreement
fixing minimum wage rates for a whole industry, the employer cannot
pay less than these rates. There may also be more or less effective
undertakings or agreements between the employers themselves as
regards maximum rates and earnings. Moreover, rates fixed under
a national collective agreement may in turn reflect relationships with
wage rates paid in other industries, and so on.
Whatever the precise situation may be as regards the number and
nature of the firm's jobs for which, and of the external jobs with which,
significant relationships exist, there will normally be some rates strongly
affected by external wages, and the firm has to fix them within a rather
narrow range of discretion. These rates, or at least their lower limits,
are more or less independent of the firm's judgment as to what might
constitute a fair remuneration for the jobs concerned. They necessarily
constitute more or less fixed minima in the firm's internal wage structure,
whether this is determined on the basis of some system of job evaluation
or otherwise. In respect of other jobs the firm has a larger degree of
discretion in choosing rates that it considers appropriate. Finally, since
external rates have the effect of placing lower limits on the rates paid by
the firm, if the latter pays wages which are high compared with those
offered by neighbouring firms, its freedom of choice in respect of its own
wage structure is greater than that of " low-wage firms ".
Internal Factors
The second aspect of a firm's wage structure is the pattern of relative
job rates within the organisation. It is this aspect which, in most cases,
job evaluation seeks to improve more particularly. Here again, much
depends on the practical possibility of filling certain jobs and obtaining
a satisfactory performance from their incumbents; thus, if a very dangerous, heavy or responsible job were not paid more than comfortable,
light tasks not requiring greater knowledge or skill, workers would
shun such a job. Yet the need for filling jobs is apparently not a sufficient
criterion for, or explanation of, the way in which relative rates are fixed.
Thus, it has been pointed out that " the better-paying jobs appear at
any given time to be rationed " * within most organisations. At least in
highly industrialised countries, in normal times there usually are far more
1

LrVERNASH, op. cit., p. 142.

JOB EVALUATION

workers fit for and desirous of promotion than there are highly paid
jobs. This suggests that, from the point of view of equalising supply
and demand, such jobs are overpaid although from the employer's point
of view, as well as from a broader economic angle, their rates might still
be considered justified, for example as an incentive to seek promotion
through good performance and eagerness to participate in training
programmes.1
In some cases wage earners themselves have been known to acquiesce
in wage differentials rather larger than could be accounted for purely in
terms of supply and demand or even on the basis of comparative job
contents. This seems to be the case particularly when the lower-paid
workers are in the position of assistants or apprentices to more highly
skilled personnel and are themselves likely to reach the higher positions
and wages at some later date. Examples that have been cited are those of
the assistant spinner compared with the spinner, the underground
haulage collier compared with the face worker and the labourer in steel
smelting compared with the smelter.2
In general, however, wage earners seem to attach great importance to
what they regard as an " equitable " wage structure, judged largely by
comparative job contents in terms of qualifications required and adverse
working conditions to be endured. They are strongly inclined to compare
their present with their former jobs and their own tasks and wages with
those of others, in terms of appropriate reward and prestige. They will
resent wage differentials to the advantage of others if they cannot find
them justified by differences in job contents or other " good reasons ".
An important problem in this connection is that which arises when
highly skilled workers paid on time rates earn less than less skilled
workers paid according to an incentive system. For example, in a plant
manufacturing asbestos products a machinist supervisor was found to
earn about 20 per cent, less than a pipe-coveringfinisher,although under
a system of job evaluation his work might be rated more than twice as
highly as that of his less skilled colleague paid according to an incentive
scheme.3 The problem is made more difficult because earnings under
incentive systems show a tendency to rise continually as a result of
gradual improvements in productivity. Small changes in raw materials,
tools, methods of work and tacit or formal rules about earnings to be
achieved when new jobs are put on piece rates all exert an upward
1

LrvERNASH, op. cit., pp. 142-143.

* REYNOLDS and

TAFT, op.

cit.,

pp. 58 and

274;

and

H. A. TURNER: "Trade

Unions, Differentials and the Levelling of Wages ", in The Manchester School of
Economic and Social Studies, Vol. XX, No. 3, Sep. 1952, pp. 227 ff.
* W. B. WOLF: Wage Incentives as a Managerial Tool (New York, Columbia
University Press, 1957), pp. 30 ff.

SOME GENERAL CONSIDERATIONS

pressure on piece workers' earnings that may lead to distortions in a
wage structure originally considered as " equitable " when introduced
under a job evaluation scheme. More will be said about this difficult
problem in Chapter HI.
A further parallel may be drawn between the internal and external
aspects of a firm's wage structure. The comparisons that are or can be
made between various jobs are more direct or evident between some jobs
than between others. These differences in the degree of comparability or
precision of possible comparisons are due partly to the magnitude of the
differences in job contents and partly to the extent to which they can be
observed. For example, comparisons are less easy between broad groups
of managerial, clerical and manual jobs than between jobs within each
of these groups; between different departments than within a given
department; and between work crews than within a single crew. A
group of jobs which are closely related for purposes of comparing job
rates and contents may be called a " job cluster ", 1 In so far as the
internal wage structure aims at avoiding or removing wage inequities, it
should, therefore, give particular attention to the fixing of suitable wage
differentials within job clusters.
JOB EVALUATION AND THE WAGE STRUCTURE

It is claimed that job evaluation enables various requirements of
a rational internal wage structure to be satisfied. Especially in continental
Europe, the method is generally regarded as a way of determining
" equitable " wage rates (rather than " the value of the job "). Indeed,
a basic complement of most job evaluation systems consists in provision
for grievance procedures enabling any remaining inequities to be
corrected. Furthermore, the method is usually applied in the first
instance to manual jobs, which tend to raise more important and acute
problems of equity and lend themselves perhaps somewhat more easily
to its application than other types of work. Managerial posts are often
excluded, and clerical and commercial jobs may be either excluded or
evaluated under an entirely different system. When the jobs are analysed
direct supervisors are usually consulted, and in larger organisations
where the committee system is applied the committees are usually set up
on a departmental basis and with departmental representation. These
features, bringing in technical knowledge concerning the jobs to be
1
" A job cluster is defined as a stable group of job classifications or work assignments within a firm (wage-determining unit) which arc so ¡inked together by (!) technology, (2) the administrative organisation of the production process, including
policies of transfer and promotion, or (3) social custom that they have common

wage-making characteristics." (J. T. DUNLOP, in TAYLOR and PIERSON, op. cit., p. 129.)

JOB EVALUATION

evaluated, help in establishing acceptable wage relations, particularly
within each department and its " job clusters ".
At the same time, all job evaluation systems recognise the need for a
suitable relationship between at least some of the firm's wage rates and
external rates. To this extent, the method is concerned not only with the
internal but also with the external aspects of the firm's wage structure.
JOB EVALUATION IN PLANNED ECONOMIES

The procedures and, to a lesser extent, the objectives of wage fixing
in the planned economies of eastern Europe differ rather substantially
from those in unplanned economies. Yet there is a striking similarity
between many of the technical problems encountered in planned and
unplanned economies, and this is true in particular as regards some of the
questions with which this study is concerned. While it has not been
possible to analyse detailed case material from eastern European countries (with the exception of the German Democratic Republic) a few
brief remarks on wage problems in these countries may be made here to
illustrate both certain common features of planned and unplanned
economies and the importance of one of the problems of wage differentials to which reference has been made above.
In principle, basic wage scales in the U.S.S.R. and in other eastern
European countries have for a long time borne some relation to the
characteristics of the jobs concerned: skill, responsibility, the complexity
and difficulty of the work, and also working conditions—e.g. exposure
to high temperatures and (at least in the Soviet Union) the location of
certain plants, due account being taken of the harsh climate, isolation,
and inadequate housing facilities in some parts of the country.
On this structure of basic wage scales a large number of additional
allowances and other features have been superimposed, including particularly systems of incentive wages covering an unusually high proportion
of production workers. In the course of the years the widespread application of incentive systems has led to an increasingly wide gap between
earnings and basic rates. In the U.S.S.R. few changes were made
between 1932 and 1956 in the basic rates and the job classifications to
which they relate, while output " norms " for incentive workers were not
systematically revised either. As a result, basic wages of incentive
workers came to constitute a progressively falling proportion of their
total earnings, while premium and bonus payments became in some cases
increasingly independent of personal application and effort.1 At the
1
As will be seen below, rather similar problems were encountered, for example,
in connection with the introduction of job evaluation in the steel industry of the
Federal Republic of Germany.

SOME GENERAL CONSIDERATIONS

same time the earnings of less-skilled workers paid by results sometimes
surpassed those of highly skilled workers, who are not paid under
incentive systems, diminishing worker's interest in improving their skills or
seeking advancement. These and other distortions in the wage structure
gave rise to decisions, after the 20th Congress of the Communist Party
of the Soviet Union in 1956, to revise the country's wage system. This
decision was followed by similar developments in several other eastern
European countries.1
It has been suggested that the discussion concerning a basic improvement of the wage system in these countries, if it is not to be conducted
on too narrow a basis, should not concentrate only on questions of
" work norms " and payment by results, but should be concerned also
with questions of job content and the systematic application of job
evaluation.2
JOB EVALUATION AND COLLECTIVE BARGAINING

As will be seen from the description of job evaluation procedures contained in Chapters II and III, it is perfectly possible to combine job
evaluation and collective bargaining. Some trade unionists attack job
evaluation on the ground that it " limits collective bargaining " or even
discuss the whole subject as if job evaluation were an alternative to
collective bargaining; they argue, for example, that " collective bargaining is better than job evaluation ".
If " job evaluation " is understood in the widely accepted sense given
to it in this study, this is a false distinction. Wages may or may not be
fixed in collective negotiations, and in such negotiations both parties
may or may not agree to use job evaluation to provide some of the
criteria to be used in their discussions. It is perfectly possible that both
parties may wish to do so. It is also possible that the workers' representatives may not regard job evaluation as a useful instrument for
collective bargaining. Some of the reasons why they may take this line
are mentioned in Chapter V; perhaps the most important is that they
may feel that acceptance on their part of job evaluation as an important
criterion in the fixing of relative wages would unduly restrict their
freedom to negotiate and would therefore not be in the best interests of
1
For a more detailed discussion of these problems, see " Some Aspects of Wage
Policy in the Planned Economies of Eastern Europe", in International Labour
Review (Geneva, I.L.O.), Vol. LXXIX, No. 1, Jan. 1959, p. 1; and United
Nations: Economic Survey of Europe in 1956 (Geneva. 1957). Chapter I, pp. 36 ff.
* E. HAMPICKE: Grundfragen der qualitativen Arbeitsbewertung, published under
the auspices of the Karl Marx University (Berlin, Verlag Die Wirtschaft, 1957),
pp. 29 ff.

JOB EVALUATION

their members or of their organisations. Again, it is quite possible that
employers will not wish to use job evaluation, either because they feel
that the results would not be worth the time and trouble involved, or
because they feel that job evaluation would create obstacles to the
negotiation of changes in relative wage rates called for by changes in
market conditions.
In other cases, however, even where trade unions do not accept job
evaluation as a basis for further negotiations, the employer may regard
it as a useful management tool for his own guidance. When discussing
with the workers' representatives problems of relative remuneration he
will normally have his own ideas as to what constitutes the most desirable
internal wage structure, and he may base these ideas on more or less
logical and objective standards of job evaluation, hoping that in his
negotiations he can reach a large degree of agreement with the unions on
what he regards as a good wage structure.
Again, the employer may or may not seek formal acceptance of his
job evaluation system by the workers' representatives as a useful basis of
discussion. Similarly, he may or may not seek the active participation of
workers' representatives in the introduction and application of the
system. As will be seen in Chapters II and III, there are many job
evaluation schemes in which there is full joint participation by labour and
management. There are, however, also cases in which workers' representatives know that management applies job evaluation and are
prepared to discuss the ranking of jobs resulting from the method, but
do not wish or are not allowed to participate in the process of evaluation
itself, and do not necessarily accept the results of the method in the
process of wage bargaining.1 Finally, there are cases in which the
employer decides to apply job evaluation as a basis for clarifying his own
ideas as to what constitutes the most desirable wage structure, without
even informing the workers.
Whatever may be the formal status of job evaluation in the process of
wage bargaining, and whatever the reasons may have been for choosing
one course rather than another, it will be clear from the foregoing that
the mere application of job evaluation as such has nothing to do with
the question whether wages arefixedby collective bargaining or not. An
agreed system of job evaluation cannot be applied without the agreement
of the workers; but, if employers wish to use it as a management tool for
1
" Unions have simply stated that all they are interested in are the wages paid to
their members and that management may use whatever means it desires to arrive at
what it thinks wages should be. The union simply reserves the right to bargain on
the results of any management wage determination, regardless of how the determination is made." (Collective Bargaining Report (Washington, A.F.L.-C.I.O.), Vol. 2,
No. 6, June 1957, p. 33.)

SOME GENERAL CONSIDERATIONS

their own guidance, this is essentially a management decision over which
workers have little or no control. The real questions which trade unions
have to settle are whether they wish to encourage application of the
method and, if they are asked to do so, whether they are willing to
participate in the operation of the method and to accept its results as a
guiding criterion for negotiations concerned with questions of relative
remuneration.
Finally, it should be recalled that job evaluation is concerned with
problems of relative remuneration only. It has nothing to do with the
fixing of absolute wage levels or, for that matter, absolute wage differentials between evaluated jobs: even if both parties have agreed to apply
job evaluation and to use its results as a basis for wage determination,
wage levels and the absolute amount of wage differentials remain to be
negotiated.

CHAPTER II
METHODS OF JOB EVALUATION
There are four main types of job evaluation schemes based on four
different methods, namely the ranking method; the grade or classification method; the factor comparison method; and the point rating
method. The first two are usually referred to as " non-analytical "
(and sometimes as " non-quantitative " or " summary ") methods, while
the latter are known as " analytical " (or " quantitative ") methods.
The terms " quantitative " and " non-quantitative " are often used in
the United States ; " analytical " and " non-analytical " correspond to
the usage in other languages, including German. In Eastern Germany
the term " qualitative " is used to designate job evaluation in general,
the word " quantitative " being reserved for incentive wage systems.
These questions of terminology are of small importance; in the following
pages the two main groups of methods are described as " analytical "
and " non-analytical " respectively, but the other terms would be equally
suitable.
Although point rating systems are the ones most widely applied,
very brief descriptions of all four methods are given in the following
pages. As will be seen from these descriptions job evaluation can be
applied in different ways with varying degrees of detail. Underlying
particular systems there are, however, certain broad common principles.
First of all it is necessary to be quite clear as to the criteria to be
applied. Under all methods except the simple ranking method it is
necessary to define either grades in which jobs will be placed or " factors "
in terms of which they will be evaluated. These criteria must be defined
in clear, unambiguous terms. The next step consists of a more or less
thorough examination and description of the jobs to be assessed in
terms permitting them to be evaluated according to the grade or factor
descriptions established for the purpose. When an analytical method
is applied this examination has to be more detailed than when a nonanalytical system is used. In the former case, job analyses have to be
made, breaking down the content of the job into components corresponding to each of the criteria or factors distinguished in the particular
method used.

METHODS OF JOB EVALUATION

Then the jobs have to be compared with each other or with some
other jobs. Again, this comparison may involve no more than an overall judgment of the jobs as a whole or it may consist of a more detailed
comparison in terms of the individual factors involved. Finally, the
jobs have to be arranged in a certain order, or " ranked ".
The definitions of such basic elements as grades and factors have to
be drawn up very carefully. This is normally done after extensive
discussion and consultation between various people with special knowledge of the jobs. Usually, after some provisional definitions have been
prepared, these are tested by applying them to certain important jobs
so as to see whether the definitions can form a basis for acceptable wage
relations between these jobs. In the light of such experiments the definitions may be revised. Once they have been established in final form
the various definitions are usually laid down in manuals that may also
contain illustrations of specific job descriptions and ratings. -A good
manual will not only clarify the practical interpretation to be given to
more or less abstract definitions, but will also help in ensuring that the
system is applied in a consistent manner. Simplified versions of the
manuals may also be circulated to the staff to help them understand how
their jobs and wages have been fitted within the general structure of
their firm.
It has already been observed in the section of the previous chapter
dealing with job evaluation and collective bargaining that each and
every step involved in the introduction and administration of a job
evaluation system can be fitted into the framework of joint consultation
and collective bargaining. In practice this has often been done. In any
case it will usually be inevitable to have one or more committees set up
to prepare the definitions and to rank and evaluate jobs. In some cases
these tasks are entrusted to a single individual, but if more than a few
jobs, especially jobs with rather widely varying characteristics, are involved, it is usually necessary to establish a group of persons whose
combined knowledge and experience will enable them to carry out these
tasks in a responsible manner.
NON-ANALYTICAL METHODS

The Ranking Method
Description.
The ranking method is the simplest of all job evaluation methods.
It merely entails placing placing the jobs in order of the demands they
are considered to make on those who perform them, on the basis of

JOB EVALUATION

their titles only or of simple descriptions of each job as a whole. These
operations are quite simple in the case of a small plant with only a few
jobs, but may prove unsatisfactory in a larger plant with a considerable
number of jobs, some of which may be quite complex. In this event
grouping of jobs on a somewhat arbitrary basis may be necessary,
especially if it is desired not to have too many individual grades and
wage rates.
Advantages and Limitations.
The ranking method is relatively simple, inexpensive and expeditious.
A disadvantage, of course, is that jobs may be ranked on the basis of
incomplete information and without the benefit of well-defined standards.
This means that the ranking may be somewhat superficial and lead to
the ranking of the job holder rather than the job itself. The rank position
of the different jobs is also likely to be influenced by prevailing wage
rates.
The Grade or Classification Method
Description.
The grade (or classification) method is widely used for salaried jobs,
including those in government and service occupations. It is also used
for manual jobs in some industries in which the same worker is called
upon to perform a variety of more or less standardised jobs for which
standard rates of remuneration can be fixed, as in engineering workshops and in the electrical and shoe industries.1
This method differs from the ranking method in that labour grades
are established before actual job characteristics are examined. The
number of grades is decided upon, and the functions corresponding to
these grades determined; only then are descriptions prepared for each
job and matched with those of the different grades established.
The grade descriptions are framed in such a manner as to cover
discernible differences in degree of skill, responsibility and other job
characteristics. The lowest grade may cover jobs which merely require
the workers concerned to follow simple instructions under close supervision. Each succeeding grade will reflect a higher level of skill and responsibility, with less and less supervision. Since the method is non-analytical,
jobs are not broken down into their component parts but, as in the
ranking method, considered as a whole.
1

J. WIBBE: Entstehung, Verfahren und Probleme der Arbeitsbewertung (Munich, Carl
Hanser Verlag, 1953), p. 23. The author quotes an agreement of 1949 for the shoe
industry in Western Germany, under which about 200 different jobs were classified
into six wage groups.

METHODS OF JOB EVALUATION

Advantages and Limitations.
The classification method is also relatively simple and, on the whole,
inexpensive and easy to apply. In organisations with a relatively small
range of jobs the results achieved may be fairly satisfactory. When the
range increases, however, it becomes more and more difficult to compress
so many different jobs into a limited number of grade descriptions.
A given job may present various characteristics in widely differing
degrees: such a job will be classified in one grade or the other depending
on which characteristic is considered. In such cases it will become
necessary to increase the number of grades, but this tends to destroy
one of the main virtues of the method—that of simplicity.
Sometimes the classification method is preferred to analytical methods
precisely because of its comparative vagueness and hence greater flexibility. For example, when the nature of a job changes owing to new production methods its classification will normally have to be negotiated
anew, and analytical methods, with their pre-determined definitions of
factors and degrees, leave less freedom to the parties in such negotiations.
As a result, trade unionists have sometimes expressed a preference for
the classification method on this ground.
For instance, under the job evaluation system introduced in 1955 for
time workers in the British coal mines, after an original list of over
6,500 " jobs " emerging from an inquiry into all occupations and job
titles in use in various areas had been condensed into a standard list
of about 400 national jobs, the latter were divided into a number of
grades. Three such grades were established for the craftsmen's jobs,
five for underground jobs and five others for surface jobs. Although an
analytical method could have been used, " the method of examining the
job as a whole and assigning it to one of a few grades by common agreement produced results which in this case were substantially confirmed
by an independent job evaluation based on a points system ".1 The
key feature of the scheme, it seems, is the small number of grades. It
was considered that most people would recognise two extreme classes
of jobs and a " middle class " in between ; from this point on, there is
only one further step to admitting two additional classes, one between
each extreme and the middle, so that a division into five classes is
obtained.
An early example of the classification method is provided by the
German " Wage Group Catalogue for the Iron and Metal Industries "
of 19422, which at one time was rather widely applied in the German
1

SALES and DAVIES, op. cit., p. 213.

Lohngruppenkatalog Eisen und Metall. Briefly described in WIBBE, op. cit.,
pp. 21 ff. and 49 ff.

JOB EVALUATION

metal trades. It divided jobs among eight wage groups, and contained
a general rating of each group in terms of basic job requirements. The
job characteristics distinguished were skill, physical and mental effort,
responsibility and working conditions, the main emphasis being placed
on skill and training requirements. The general descriptions relating
to the wage groups were not sufficient for general application of the
method throughout the industry, and for the most important jobs
(e.g. turning, grinding and milling) so-called " wage group characteristics " were therefore published, laying down detailed criteria for the
classification of each of these jobs in one wage group or another according to the degree of difficulty involved in each particular case. In this
way, the system was made to approach an analytical method. Even so,
the many different jobs found in the industry could not be easily classified,
and a further catalogue was issued containing detailed descriptions of a
few thousand jobs which were to serve as examples for the proper classification of jobs in individual firms.
A somewhat similar system exists at present in the German
Democratic Republic. Under this system, industrial manual workers are classified in eight wage groups under a simple scheme based
on three criteria: training required; difficulty; and muscular effort.
For example, wage group 3 includes jobs requiring either (a) six
months' training on the job (but no pre-employment training) if the
work involves simple operations on machines; (b) six weeks' training
on the job if the work is " heavy "; or (c) no training if the work is
" very heavy ".
" Heavy " jobs requiring six months' training are placed in wage
group 4, as are " very heavy " jobs requiring six weeks' training; the
same is true of jobs requiring at least 12 months' training on the job
and involving " difficult " operations on machines.1
On this basis, 88 lists of jobs have been drawn up for individual
industries and one or two other classes of activity showing in what wage
group each job is to be classified. A catalogue with examples for the
classification of commercial, technical and " economic " employees in
salary groups has also been worked out. It is based on three job requirements: difficulty and variety of work; supervisory responsibility; and
training and experience.2
A detailed and comprehensive classification system for civil servants
exists under the United States Classification Act of 1949 which has
1
E. KASCHEL: Handbuch für die Lohngruppeneinstufung (Berlin, Verlag Die Wirtschaft, 1959), pp. 20 ff.
2

K. FLASCHE, A. LORENZ and W. RIECHERS: Die Ausarbeitung, Einführung und

Handhabung des Gehaltsgruppenkatalogs (Berlin, Deutscher Zentralverlag, 1959).

METHODS OF JOB EVALUATION

superseded an earlier Act of 1923. Under this system, an exceptionally
broad range of professional, scientific, clerical and administrative jobs
(called "positions") are placed in 18 "General Schedule" grades. 1 A
few extracts from this important enactment may help to clarify the
nature of the classification method used.
Grade GS-1 includes all classes of positions the duties of which are to
perform, under immediate supervision, with little or no latitude for the exercise
of independent judgment, (1) the simplest routine work in office, business,
orfiscaloperations, or (2) elementary work of a subordinate technical character
in a professional, scientific, or technical field.
This describes the lowest level of jobs. Each higher grade description
makes greater demands on the individual, until in GS-18 one finds
included, inter alia, jobs, the duties of which are—
. . . to plan and direct or to plan and execute frontier or unprecedented professional, scientific, technical, administrative, fiscal or other specialized programs of outstanding difficulty, responsibility and national significance,
requiring extended training and experience which has demonstrated outstanding leadership and attainments in professional, scientific, or technical
research, practice, or administration, or in administrative, fiscal, or other
specialized activities;
For example, grade GS-2 comprises jobs " with limited latitude for
the exercise of independent judgment " and although the work is still
" routine " it is no longer " the simplest routine work ". Instead of
" elementary work ", jobs in this grade may involve " subordinate
technical work of limited scope, requiring some training or experience ".
For jobs in grade GS-3 supervision may be " immediate or general ",
the work " somewhat difficult and responsible ". The job should require
" some training or experience ", " working knowledge of a special
subject matter " or " to some extent the exercise of independent judgment
in accordance with well-established policies, procedures, and techniques ".
Jobs in GS-4 are " moderately difficult and responsible " and require
" a moderate amount of training " and (not " or ") the exercise of
independent judgment, though still within the framework of wellestablished policies. In GS-5 work is " difficult and responsible " ; it
requires " considerable training ", " broad working knowledge " and
" independent judgment in a limited field ". In this grade appears,
however, also a different type of job involving low level scientific work.
The duties which it entails are—
. . . to perform, under immediate supervision, and with little opportunity for
the exercise of independent judgment, simple and elementary work requiring
1

Title VI of the Act.

JOB EVALUATION

professional, scientific or technical training equivalent to that represented by
graduation from a college or university of recognized standing but requiring
little or no experience;
Work in GS-9 is " very " and in GS-10 " highly " difficult and
responsible; in GS-11 it becomes " o f marked difficulty and responsibility " and so on, until in GS-18 one finds, in addition to the scientific
posts mentioned above, heads of bureaux for whom " the scope, complexity, and degree of difficulty of the activities carried on is exceptional
and outstanding among the whole group of positions of heads of
bureaux ".
It may be noted that in both this and the earlier example of the East
German catalogue method, the same grade or wage group may include
jobs of quite different types. In the German case a worker doing somewhat difficult but light work is in the same group as a worker doing simple
but very heavy work. In the United States civil service a " moderately
difficult and responsible " administrative job and a simple scientific
job involving little responsibility are both classified in grade GS-5. Thus,
for purposes of classification, entirely different job characteristics are,
as it were, substitutes for each other and a certain degree of one is
considered equivalent to some degree of the other. This, indeed, is a
basic feature of job evaluation, especially of point rating methods, in
which (as will be seen further on) rather precise measures of these
equivalents are applied.
ANALYTICAL METHODS

Analytical job evaluation methods differ from the two just described
in that comparisons of job contents are made separately in respect of
each of a smaller or larger number of job characteristics called " factors ".
These partial comparisons are then " added " so as to arrive at a reasoned
comparison of the job contents in their totality. The first part of this
process, namely the comparison of individual factors, may again be
carried out by simple ranking or by classification. In the former case
the factor comparison method is used, in the latter the point rating
method.
The Factor Comparison Method
Description.
The factor comparison method, though less refined than point rating
systems, is rather complicated in practice. For this and also for some
other reasons, it is not very widely used.

METHODS OF JOB EVALUATION

As originally developed, this method involves the ranking of different
jobs in respect of certain factors and, as will be seen from the description
below, usually also involves the assigning of money wages to the jobs.
The fact that the method leads directly to the determination of wages
(for which reason it has been called the " weight-in-money " or " directto-money " method) has been regarded as an undesirable feature by
those who consider that the only task of job evaluation is the comparison
of job contents, " an act which logically is completely distinct from the
fixing of wages ".1 As will be mentioned below, the factor comparison
method may, however, also be applied in a different way.
The first task in applying the method is to select and clearly describe
the factors to be used. These may be, for example: (1) skill; (2) mental
and educational requirements; (3) physical requirements; (4) responsibility; and (5) working conditions. Each of these may be subdivided;
for example, working conditions may be broken down under such
headings as " temperature," " dust " and " other inconveniences," and
responsibility into " responsibility for other people," " equipment,"
" progress of the work of a crew," etc.
The ranking of jobs in respect of each of the factors and the subsequentfixingof wages are first carried out for a number of " key jobs " or
" benchmark jobs " which serve as points of reference for the later
evaluation and ranking of the other jobs. For this purpose, the key
jobs should satisfy a number of conditions. First, they should be capable
of clear description and analysis in terms of the factors used. Secondly,
the group of jobs selected should cover a sufficiently wide range as
regards the importance of each factor. Thirdly, when the rates for the
key jobs are to be used as the standard for the fixing of appropriate
wages in the other jobs, these rates should be regarded as appropriate
by all concerned and, in the case of some jobs, should not differ too
much from the rates paid for these or similar jobs in the local employment market. Finally, the key jobs should include jobs covering a
sufficiently wide range of pay grades including, specifically, the lowest
and one or more among the highest-paid jobs in the plants concerned.
The number of key jobs thus required as a basis for a job evaluation
plan of the factor comparison type depends on the number and
variety of jobs in the organisation: it may range from less than 15 to
more than 100.
First, the key jobs are ranked successively by reference to each of the
factors chosen. For example, if there are five jobs and also five factors
1

WIBBE, op. cit., p. 26; and G. W. HAGNER and H. WENG: Arbeitsschwierigkeit und

Lohn, second edition (Cologne, Bund-Verlag, 1952), p. 9 and passim.

JOB EVALUATION

the ranking might be as in table 1. When the ranking is done by a committee the usual practice is for each member to make his own ranking,
the results being then averaged. For example, if the averages for skill for
fivejobsare 1.2 for job No. 1,2.3 for job No. 2, 3.5 for job No. 3, 3.6 for
job No. 4 and 4.2 for job No. 5, the final ranking for skill will be 1, 2, 3,
4 and 5.
TABLE 1. RANKING JOBS BY FACTORS UNDER THE FACTOR
COMPARISON METHOD
Job

Toolmaker
Machinist
(grade 1)
Electrician
(grade 1)
Assembler
(grade 1)
Janitor

Skill

Mental
requirements

Physical
requirements

Responsibility

Working
conditions

4
5

5
1

5
4

2
1

After such agreement bas been reached, if the method is to be used
for immediate fixing of money wages, the jobs are ranked again but
according to a different procedure. The wage rate for each key job is
broken down and distributed among or apportioned to the factors in the
proportions in which these are considered to contribute to the total
" price " paid for each key job in the form of its wage rate. For example,
if toolmaking is a key job and if its wage rate is 20 money units, it may
be decided to allot nine of these to skill, five to mental requirements,
two to physical requirements, three to responsibility and one to working
conditions. Similarly, if the wage rate for another key job, for example
that of a first grade machinist, amounts to 18 money units, eight of these
may be allotted to skill, three to working conditions, and so on. When
the rates for all key jobs have been divided in this way the jobs have
implicitly been ranked again with respect to each of the factors. In the
example given, the toolmaker ranks above the machinist as regards skill
requirements, but below the machinist if the jobs are ranked on the basis
of working conditions.
After the results have been averaged by a committee, in the manner
described above, the allocation of wage rates and the implied ranking by
factors of the jobs covered by table 1 might work out as indicated in
table 2.

METHODS OF JOB EVALUATION

TABLE 2. ALLOCATION OF MONEY VALUES TO THE DIFFERENT
FACTORS AND RANKING OF JOBS UNDER THE FACTOR COMPARISON
METHOD

„„
Wage
rate in
money
units

Job

Toolmaker
Machinist
(grade 1)
Electrician
(grade 1)
Assembler
(grade 1)
Janitor . .

require-

'al? -s"-

11
>"9
o >••£
I l'i
I* l'i
li
s
s
s
CO

>"2 'S °

•c
3o
O

' 3

3 o
•3""
5<*-

lï

¿a&

M
CXI

* 3
>>.£>

2.2»

il
S

1°

li 35*°
1°
SII
' 3
>.XJ

•e""

. .

9.0

5.0

2.0

3.0

1.0

. .

8.0

4.0

1.0

2.0

3.0

. .

6.0

3.0

1.5

2.5

. .
. .

14
12

4.0
2.0

4
5

2.0
1.0

4
5

1.5
4.0

4
1

1.0
0.5

4
5

5.5
4.5

1
2

The two rankings of the key jobs are undertaken independently
of each other and need not coincide. Their respective results, as illustrated by the two examples given above and embodied in tables 1 and 2,
are compared in table 3.
TABLE 3. COMPARISON OF RANKINGS BY FACTORS AND BY MONEY
VALUES UNDER THE FACTOR COMPARISON METHOD

Job

"§

K lì
Toolmaker . . . .
Machinist
(grade 1) . . .
Electrician
(grade 1) . . .
Assembler
(grade 1) . . .
Janitor

Physical
requirements

re^e-

Responsibility

•a a

*1
CO >

Working
conditions

•aá

CO >

4
5

5*
1

4*
1

5*
4*

4*
5*

2*
1*

1*
2*

* = different results obtained from ranking by factor and by money value.

It will be noted that there are ten differences in ranking at the two
stages (shown by asterisks in the table). These differences have to be

JOB EVALUATION

removed by either increasing or decreasing the money values of the
different factors for the jobs concerned or by examining the job contents
again to see if a suitable adjustment can be made in the factor rating.
If it is not possible to reconcile the rankings for a particular job it is
eliminated from the list of key jobs.
The results obtained can now be used to rank the other jobs in the
plant and to fix their wage rates. On the basis of job descriptions the
content of each job is analysed and compared with the key jobs in terms
of each of the factors separately. For example, the job of a "drill press
operator (grade 2) " might be found to call for skill requirements between
those of " electrician (grade 1) " and " assembler (grade 2) ". The wage
component corresponding to skill for the job would then also be fixed
somewhere between the amounts of money allotted for skill to the two
key jobs (six and four money units respectively) ; it might be agreed that
the appropriate reward for skill for the drill press operator's job would
be 4.5 units.
Applying the same procedure to the other factors, one arrives at the
total wage rate for the job.
Advantages and Limitations.
The main advantage of the factor comparison method over the two
non-analytical methods described earlier is that it permits of a more
systematic comparison of jobs than could be obtained under either of
these methods. On the other hand, the method as described above is
difficult to explain to workers. Furthermore, one of its essential features
is that the wage rates for the key jobs are presumed to be correct and
that all other rates are determined by reference to the latter. As has been
mentioned above, many people believe that the operations of evaluating
jobs and fixing their wages should be kept entirely apart. Furthermore,
the apportionment of wages among different factors has been denounced
as necessarily arbitrary.
Acceptance of prevailing rates for the key jobs as the basis for the
whole wage structure does not, however, constitute an essential feature
of the factor comparison method. What is essential is the ranking of
jobs in respect of each of the factors, and this may be done without
reference to existing rates. For example, the job which ranks highest in
respect of a particular factor might be given a certain point score for
that factor; the job ranking lowest for this factor might be given zero
points, and appropriate intermediate values might be assigned to other
jobs. Thus modified, the method would come closer to the point rating
method (which is discussed below) although it would still differ from the

METHODS OF JOB EVALUATION

latter in that there would be no precisely defined degrees and weights for
skill requirements, responsibility and other factors.
An example of a factor comparison scheme not involving direct
translation of ratings into wages is found in the method proposed by
Hagner and Weng in Western Germany.1 This is based on seven major
factors: job knowledge, including experience; dexterity; responsibility;
muscular effort; sensorial and nervous effort; mental effort; and environmental influences (job conditions). The responsibility factor is subdivided
into responsibility for equipment and output; for the safety of other
workers ; and for the work process (utilisation of plant and equipment
and co-ordination of the work of others). Job conditions are broken
down under eight headings: temperature; water, dampness, acids and
fumes; dirt, grease, oil and dust; gases; noise and vibration; lighting;
risk of catching colds; and risk of accident. Thus the total number of
separate job characteristics is 16.
The jobs to be evaluated are ranked in terms of each factor and
fitted into scales ranging from zero to 100 points. The top of each scale
(100 points) expresses the highest degree in which the factor concerned
is involved in any job within the plant or (if the method is applied at an
inter-plant or inter-industry level) group of plants concerned. In other
words, the degree in which a factor is found in a particular job is expressed
as a percentage of the maximum. The authors regard their method as
applicable to most existing jobs and give examples of jobs ranking
highest in terms of each individual factor. Thus, the top-ranking job in
terms of knowledge and experience is that of a moulder in steelcasting,
handling pieces weighing up to 140 tons and requiring up to 400 hours
for moulding; that cited for " responsibility for the security of others "
is the job of a person supervising from 20 to 30 workers in the construction of underground tunnels, river dams and similar works; and the
example given for muscular effort is the carrying of 100-kilo sacks on
ladders, for example in ports.
The final ranking of jobs results from addition of the points assigned
to each job for each factor. In order to demonstrate the general applicability of the method, the authors give complete examples, including the
1
HAGNER and WENG, op. cit. ; in a publication issued by the West German metalworkers' union under the title Arbeitsstudien (Frankfurt, Industriegewerkschaft Metall,
1958, pp. 6 ff.), this method is put forward as reflecting the union's views on job evaluation. The same publication contains a critical discussion of other methods used in the
Federal Republic. Other variations of the factor comparison method calculated to
avoid the difficulties of the " weight-in-money " method have been proposed in the
United States, for example by W. D. TURNER (see Personnel (New York, American
Management Association), Vol. 24, No. 6, May 1948, pp. 476-492; and Vol. 25,
No. 2, Sep. 1948, pp. 154-160) and also by E. N. HAY and D. PURVES (ibid., Vol. 28,
No. 2, Sep. 1951, pp. 162-170; and Vol. 31, No. 1, July 1954, pp. 72-80).

JOB EVALUATION

description, analysis and ranking of ten rather varied jobs. These, Usted
in the order of the ranks finally assigned, are the following: taxi-driving
in a medium-sized city; housewifery; a particular type of toolmaking;
watchmaking; sand-shovelling; delivery of mail; the duties of a female
clerk in a workshop with 50 workers; turret-lathe operating; the fixing
of cylinder heads to engines on a conveyor belt; and dish-drying.
The Point Rating Method
Description.
The point rating method combines comparative analysis of job
contents in respect of factors with the principle of grading or classification.
As with the factor comparison method, a number of factors are distinguished, but instead of jobs being simply ranked in terms of the various
factors, a number of degrees are first distinguished for each factor (e.g.
very small, small, average, large or very large). These degrees must, of
course, be defined in a fairly detailed manner, the jobs being analysed
by reference to such definitions rather than with respect to each other,
—although the latter operation is also performed as a check on rankings
obtained on the basis of degree definitions. In some systems account
is further taken of the relative frequency or duration of the various
factors as they occur in the normal performance of the job.
The various factor degrees are assigned certain point values expressing
the importance attached to the various elements composing a job. The
simple or " weighted " addition of these points leads to a single measure
of the content of the job which can then be used for thefixingof suitable
relative wages.
Thus, the first steps involved in applying the point rating method are
(a) to select and define the factors that will be used in evaluating jobs ;
(b) to determine the number of degrees to be distinguished under each
factor, and to describe them; and (c) to assign point values to each
degree of each factor.
Job descriptions are next prepared, on the basis of which the jobs
are evaluated. This can be done either by taking each individual
job and comparing the job description with the degree definitions in
respect of all factors or by evaluating all the jobs in terms of the first
factor, then of the second, and so on. In practice both methods are
used, the latter being employed as a check on the results of the former.
As under the factor comparison method, it is customary to apply all these
steps first to a number of suitably selected key jobs. When a number of
such jobs have been evaluated it will be possible to judge whether the
definitions and point values selected are adequate as a basis for compari-

METHODS OF JOB EVALUATION

son of jobs within the plant and for the fixing of an acceptable wage
structure. Should this not be the case, the definitions and point values
will have to be altered.
Advantages and Limitations.
Compared with the original version of the factor comparison method,
point rating systems have the advantage of using points and not money
values. This enables job evaluation and wage fixing to be kept separate.
Since the jobs are evaluated on the basis of predetermined definitions of
factor degrees it may also be claimed that the method is less subjective
and fairer than any version of the factor comparison method. It should,
however, not be thought that this makes the method entirely objective,
let alone scientific. Basically, the selection of factors, the definition of
degrees and the point values attached thereto involve many arbitrary
and subjective elements, even though the evaluation of individual jobs
on the basis of these elements may be more objective than in the case
of other methods.
The method has also been said to suffer from lack of flexibility. As
in the case of the classification method, a large variety of job characteristics has to be covered by a limited number of factors and degrees.
Since no predetermined degrees are applied under the factor comparison
method, the latter may be somewhat more convenient in some cases.
In fact, the two methods may supplement each other. As will be seen
further on, the choice of factors and weights must be such that an
acceptable wage structure can be obtained. Few new wage structures
would be regarded as satisfactory that would completely upset existing
relationships. In other words, the new structure will normally have to
reflect to some considerable extent the old wage structure. The various
steps involved in the factor comparison method may be very helpful
in arriving at factors and weights that satisfy this condition. For example,
the job evaluation plan in the United States steel industry (based on a
point rating system) is largely the outcome of an earlier plan developed
by the companies and known as the Co-operative Wage Study; this was
derived from existing rates by a method rather similar to the factor
comparison method. 1 Finally, the point rating method has been regarded
as inflexible in the sense that it does not enable economic changes to be
reflected in a changing wage structure. This, and some other problems
arising in connection with the method, will be discussed in Chapter III.
1
STŒBER, op. cit., pp. 27 ff. The approach is criticised, on the general grounds
mentioned earlier, in C. W. LYTLE: Job Evaluation Methods, second edition (New
York, Ronald Press Company, 1954), p. 117.

34
Examples of Some
Rating System.

JOB EVALUATION

of the Elements

involved

in Establishing

a Point

It may be useful to give a few examples of the various concepts that
have been mentioned in the above description of the point rating method.
First, an example of a factor definition.
Definition of factors. U n d e r the Netherlands standardised method of
j o b evaluation the " knowledge " factor is described as follows 1 :
Knowledge of facts and data required for the job and obtained through
training, experience or otherwise. Knowledge includes theoretical knowledge
as well as knowledge of the daily practice of work, and knowledge resulting
from repeated sensorial observation, e.g. of degrees of acidity, temperatures,
properties and structures of material.
Also included is general knowledge, indispensable for normal performance
of the job, even if the nature of this knowledge cannot be defined precisely.
This may include, for example, general knowledge about social, economic and
political conditions when this is necessary for a correct insight into and understanding of such problems in the performance of the job.
The application of knowledge, including the analysis of problems and
choosing between different possibilities, is not regarded as knowledge but is
covered by other factors, particularly independence.2
The manual then gives further details, defines factor degrees and
quotes examples of the evaluation of jobs in terms of this factor.
Definition of degrees. A n example of definitions of degrees of a
factor, taken from The United States Steel Corporation and United Steel
Workers of America Manual for Job Classification of Production and
Maintenance Jobs 3 is given below. The same system is applied by many
companies engaged in iron mining, in the mining, smelting a n d refining of
non-ferrous metals, in steel fabricating, aluminium making, can manufacturing and in railroad car making. 4 The factor to which the degree
1
Hoofdcommissie voor de Normalisatie in Nederland : Genormaliseerde Methode,
Document NEN. 3000 (The Hague, 1958), p. 33. The standardised method was
devised for application to manual jobs in all industries as well as for technical, administrative and supervisory staff. Accordingly, its factor descriptions are couched in
rather broad terms. Job evaluation is fairly widespread in the Netherlands. Early
in 1958 systems of one sort or another had been introduced in practically the whole
of the ceramics, limestone, paint, timber, ready-made clothing, paper and cardboard, flour, dairy, oil, brickmaking, stone cutting, button making, rubber working,
electrical, sugar and confectionery industries, and in plumbing. The system was also
applied in many firms in the metal trades, the iron and steel industry and the chemical
and textile industries, and preparations were being made for the introduction of job
evaluation in coal mining, construction, the police and even agriculture.
2
Defined as the nature and extent of supervision under which the job is carried out.
3
Quoted in R. TILOVF : " The Wage Rationalization Program in United States
Steel ", in Monthly Labor Review (Washington, U.S. Department of Labor, Bureau of
Labor Statistics), Vol. 64, No. 6, June 1947, pp. 967 ff. After the point rating system
from which these definitions have been borrowed had been agreed in 1947 between
the United States Steel Corporation and the United Steel Workers, it was adopted by
several other companies so that it now affects the wages of a large majority of the
country's basic steel workers.

* STIEBER, op. cit., p. xix.

METHODS OF JOB EVALUATION

definitions apply is called " responsibility for operations " which is of
outstanding importance in steelmaking where crews of men are operating
large and expensive units of equipment and their performance affects not
only the utilisation of their own equipment and the level of their earnings
(incentive rate systems are usual in these operations) but also those of
workers engaged in previous and subsequent processes. Accordingly,
the " responsibility for operations " factor is weighted rather heavily
under this system. The different degrees are indicated by the letters in
the first column of the table.
RESPONSIBILITY FOR OPERATIONS
Consider the obligation imposed on the workman for utilizing capacity of equipment
or process by maintenance of pace and machine speeds. This includes planning,
instructing, and directing the work of others.
Consider the size of crew and teamwork required, the importance and size of equipment and the degree of control exercised by the workman on the job.
Excess capacity and storage facilities between process operations are definite indicators
for the lowering of the classification in this factor.
Code

Job requirements

Little or no responsibility beyond use
of own time.
Work as member of a gang on simple
work closely directed.
Work on simple highly standardized
jobs with little equipment or no other
operations closely dependent.

Mill janitor.
Laborer.
Wire bundler.

Operate an important part of a major.
producing unit.
Operate a medium-sized producing unit
when closely associated with other
operations.
Responsible for continuity of operation
for a number of medium-sized units.

Craneman—
soaking pit.
Pusher operatorcoke plant.

Has high responsibility for complex
work planning to meet production
schedules.
High responsibility for continuity of
operations of a large producing unit.

Benchmark jobs

Numerical
classification

Base
Chipper—
conditioning.

3.0.

Bloom shearman.
Speed operator—
hot strip.

First helper—
open hearth.

4.0.

Sets pace and assumes joint responsibility with supervision for production
of a large unit.

Welderbutt weld.

5.0.

Has responsibility for maximum production from a major producing unit.

Rollerblooms.

6.5.

JOB EVALUATION

Factors and points. Table 4 shows how point values are assigned for
the various factors and sub-factors under a point rating system widely
applied in the United States—that adopted by the National Electrical
Manufacturers' Association (N.E.M.A.), which covers several thousands
of metalworking plants.
Two comments may be made on this system. First, the number of
points that can be allotted to individual jobs is rather higher than under
many other schemes, including that applied in the United States steel
industry, which has already been referred to. In practice, the importance
of the number of points is not always very great, since many systems
with low point values achieve equal precision by using half-points or
(in most cases) first decimals. In the application of systems under which
hundreds of points can be scored, jobs coming within certain ranges of
points are usually grouped into job classes with uniform wages.1 For
systems with smaller numbers of points this is not always necessary.
Secondly, it will be noted that the maximum number of points
assignable to each factor and sub-factor is not the same. The total
TABLE 4. POINT VALUES ALLOTTED UNDER N.E.M.A. POINT RATING
SYSTEM
Points for degrees
Factors and sub-factors

Skill :
Education
Initiative and ingenuity
Effort :
Physical demand . . . . . . . .
Mental or visual demand . . . .
Responsibility :
Equipment or process
Material or product
Safety of others
Work of others
Job conditions :
Working conditions
Unavoidable hazards

Total
points
possible

14
22
14

28
44
28

42
66
42

56
88
56

70
110
70

250

10
5

20
10

30
15

40
20

50
25

5
5
5
5

10
10
10
10

15
15
15
15

20
20
20
20

25
25
25
25

100

10
5

20
10

30
15

40
20

50
25

Source: National Electrical Manufacturers* Association: Job Rating Manual. Definitions of the Factors
Used in Evaluating Hourly Rated Jobs (New York, 1946), p. 2.
1

See below, Chapter III.

METHODS OF JOB EVALUATION

number of points that may be scored on account of skill is as large as
that for effort, responsibility and job conditions taken together. This
means that the system is " weighted ". The problem of weighting is
both difficult and controversial. It will be discussed at greater length
in Chapter III. At this stage it may, however, be noted that there are
three methods of weighting the factors, namely—
(a) assigning equal maximum point values to all factors or subfactors, which are thus given equal weight; although it is customary to
call such systems " unweighted " it will be clear that they may also involve
a form of weighting, depending on the selection of factors and subfactors. For example, under the Hagner-Weng system referred to above,
the same weight is given to all of the 16 sub-factors distinguished; but
since no less than eight of these have to do with working conditions,
this aspect of job content is in effect weighted very heavily by comparison
with the others.
(b) assigning equal maximum point values to all factors, but
multiplying them by different coefficients; this is done, for example,
under the Netherlands standardised method; here most factors and
sub-factors are allotted eight points at the most, but in the case of
knowledge the score is multiplied by five, dexterity by two, responsibility by four and so on; and
(c) assigning different maximum point values to various factors and
sub-factors, as in the N.E.M.A. and United States steel schemes; this
is the most frequent case.
Factor degrees and frequencies. Reference was made above to the
fact that some systems distinguish not only factors and degrees but also
their relative frequency or duration in the case of each job. This is done,
for example, under the Euler-Stevens method1, which is applied in a
number of West German iron and steel plants (employing hundreds of
thousands of workers), in parts of the iron ore mining industry (affecting
about a quarter of the total labour force in that industry) and in lignite
mining in the Rhineland. There are 16 factors and the maximum total
score is 53 points, including eight for muscular effort (the largest maximum allotted for any one factor2). Suppose, now, that out of an eighthour day a job calls for heavy work during four hours, less heavy work
during two hours and no heavy work during the remaining two hours,
1
H. EULER and H. STEVENS: Die analytische Arbeitsbewertung als Hilfsmittel zur
Bestimmung der Arbeitsschwierigkeit, Sozialwirtschaftliche Schriftenreihe, No. 3,
third edition (Düsseldorf, Verlag Stagleisen, 1954).
a
In the United States steel industry plan this factor is given little weight. As
explained earlier, this plan places a comparatively heavy premium on responsibility
for which, under the Euler-Stevens method, only three points can be scored.

JOB EVALUATION

TABLE 5.

RATING O F VISUAL REQUIREMENTS FACTOR
BELGIAN " ALPHANUMERICAL " METHOD

UNDER

Frequency with which required
Factor degree

Good
Very good
Especially good

. . .

Infrequently
(less than 10
per cent, of time)

Frequently
(10 to 60 percent.
of time)

Very frequently
(more than
60 per cent, of time)

al
bl
cl

a2
t>2
c2

a3
b3
c3

including rest periods. For the fact that the duties include heavy work
at all the job may be given, for example, one point. On top of this
1.5 points may be allotted for the four hours during which heavy work
is normally performed and 0.5 points for the two hours during which
less heavy work is done.
Similar principles are applied under the Netherlands standardised
method and under that developed by the Belgian General Technical
Commission, a tripartite body set up by the Government in 1945.1
Under the latter method a system of " alphanumerical " classification
is used indicating, on the one hand, the degree of the factor concerned
(marked by a letter) and, on the other hand, the frequency with which
this degree occurs in the job (marked by a number). For example, for
the " visual requirements " factor the classification is as shown in table 5.
This system enables jobs to be grouped quickly for other purposes
besides job evaluation, such as the drawing up of training programmes
and the assessment of labour redeployment and transfer possibilities.
For example, " a2-bl " is a convenient shorthand formula for
identifying jobs that often call for a low degree of a particular quality
and sometimes for a medium degree of that quality. When applied to job
evaluation the system enables finer distinctions to be made than would
otherwise be possible. Suppose that a fitter's job has to be evaluated
for visual requirements. If most of the time the job calls for " very
good eyesight " and if quite frequently for " especially good eyesight "
(b3-c2) the job can be (though it need not be) rated more highly than a
job which calls for " especially good eyesight " only infrequently (b3-cl).
Some trade union spokesmen have questioned this principle. For
example, Hagner and Weng believe that when a job is evaluated in terms
of a factor it should be assigned the number of points corresponding
1
Ministère du Travail et de la Prévoyance sociale: Méthode de qualification du
travail de la Commission technique générale (Brussels, 1956). The principles underlying this method were applied, for example, in the wage rationalisation programme
for the steel industry in Liège in the years 1946-48.

METHODS OF JOB EVALUATION

to the highest degree in which that factor occurs, irrespective of whether
this degree is required all or only part of the time: " It is not the worker's
fault if his abilities are not fully used by the firm. Does the manager
think of accepting a lower remuneration for the time during which he
is travelling by car to a meeting than that which he receives for the
managerial work he performs during that meeting? " 1 A similar comment is made as regards " knowledge, skill or ability " requirements in
the A.F.L.-C.I.O. bulletin quoted above.2 The case for this point of
view as regards the rating of jobs in terms of qualities required is further
discussed below.3 As far as job conditions or physical effort are concerned
it will, however, seem to many people that differentiation according
to frequency or duration is quite justified—that, for example, a job
requiring heavy lifting all day should be rated more highly in respect of
this factor than a job calling for the occasional lifting of some heavy
object; and that continuing exposure to heat in the case of blast furnace
operators should count for more than that endured occasionally by
maintenance men, many of whose tasks do not involve such exposure.
On the other hand, and more generally, it might be argued that
whereas skill differentials remunerate the worker for what, as a result of
extensive training, he is able to do, differentials based on other factors
are merely intended to reward the worker for what he is actually doing
or enduring.
It may be noted that the problem of workers performing duties that
involve different levels of job requirements may also arise under the
classification method.4
JOB DESCRIPTION AND ANALYSIS

The purpose of a job description is to provide information about
the actual content of a job. Usually, job descriptions include indications
of the purposes of the various tasks involved, the frequency with which
these tasks have to be performed, the environment and conditions in
which the work is done, the tools and equipment used, the nature and
degree of supervision received or given, the degree of skill, knowledge,accuracy, judgment and attention required, and the responsibilities
involved. Such information is useful for many purposes other than job
evaluation, such as the selection and recruitment of personnel, training,
transfers and promotions, and the establishment of safety programmes.
1

HAGNER and WENG, op. cit., p. 35.
Collective Bargaining Report, op. cit., p. 39.
» See Chapter III.
4
KASCHEL, op. cit., p. 13.
1

JOB EVALUATION

Sample Job Grade Substantiating Data from N.E.M.A. Job Rating Manual

Job Title : MACHINIST

GRADE 2

Job description : Plan and perform all operations necessary to the construction, repair
and maintenance of important and expensive standard and special machinery and equipment.
Must be able to diagnose and correct difficult mechanical trouble. Involves accurate fitting
and aligning. Work from complicated drawings, sketches and samples, or where design information is incomplete.
Job factors

Evaluation

Substantiation of factor rating

Degrees Points
Education

Requires trades training such as machinist. Interprets complex
machine drawings and makes calculations, using advanced shop
mathematics.

Experience

In addition to trades training, satisfactory performance can be
attained after a period of 3 to 4 years consisting of experience on
lower-graded work as a machinist and training on this job.

Initiative
and
ingenuity

Uses high degree of judgment in diagnosing machine difficulties
on complicated equipment and takes corrective action on own
initiative. Handles unusual or very difficult jobs on a wide variety
of shop machinery requiring considerable ingenuity.

Physical
demand

Occasionally handles heavy machine parts and motors ; however,
considerable time is spent planning operations and directing
helpers and co-workers in performing the physical work.

Mental and/or
visual
demand

A high degree of concentrated mental and visual attention is
necessary to determine cause of machine trouble and plan
course of action in making repairs.

Responsibility
for equipment
or process

Uses tools such as portable drills, machines such as grinders and
instruments such as micrometers. Damage to equipment will
normally range between $5.00 and $20.00 for any one loss
because of carelessness.

Responsibility
for material
or production

Makes repairs and instructs others in making repairs to expensive
machines. Carelessness resulting in damage to equipment and
tie-up of production may cause losses from $250.00 to $350.00
but seldom over.

Responsibility
for safety of
others

Care required in properly fastening, tightening and balancing
revolving machine parts and installing guards securely to insure
safety of others and prevent lost-time accidents.

Responsibility
for work of
others

Responsible for instructing and directing helpers and co-workers
who may be assigned to work with machinist.

Working
conditions

Maintenance and repair work on machine shop equipment and
machines involves dirty, oily or noisy conditions that are somewhat disagreeable.

Hazards

Exposed to lost-time accidents, such as eye injury, severe abrasions of hand, loss of fingers and hernia from awkward lifting,
while making repairs to, trying out or operating machines.

Total points

340

JOB GRADE 2

SCORE RANGE 338-359

DATE:

Source: National Electrical Manufacturers' Association: Guide for Use of N.E.M.A. Job Rating Manual (New York,
1946), p. 26.

METHODS OF JOB EVALUATION

In particular, it has often been found to reveal numerous possibilities
of improving the organisation and methods of work in the plant, thus
raising productivity and reducing costs. Therefore, although the preparation of job descriptions for the purpose of job evaluation in a large
plant involves a great deal of work, the cost of the operation may be
more than fully repaid by the many other uses to which the information
collected can be applied. This is one of the main reasons why job
evaluation in practice is often not nearly as expensive as it might at
first sight appear to be.
Depending on the purposes for which the material is to be used, the
information contained in the job descriptions is broken down and
recorded in a convenient way. When the purpose is job evaluation the
jobs are analysed in terms of the grade, factor and degree definitions of
the system applied. Comparison of the components of the jobs with
these definitions enables the jobs to be rated for each of the components.
An example of a description and rating of a particular job under the
N.E.M.A. job evaluation method is reproduced on the opposite page.1
PROCEDURES

As was mentioned in Chapter I, a decision to introduce job evaluation
(as distinct from the fixing of wage rates) may be taken by management
as a management task and perhaps as a starting point for negotiations
with a trade union with or without the approval of the union.
Similarly, the implementation of this decision may or may not involve
trade union participation, depending, inter alia, on the wishes of both
parties. Whether or not there is trade union participation, outside
consultants may be called in to advise on the technical problems
involved.
A second decision which must be taken concerns the choice of the
system to be applied. Sometimes an entirely new scheme has to be
developed for the plant concerned ; in other cases an existing scheme can
be used, suitably adjusted to the special needs and conditions of the
plant. Adoption of a well-tried existing scheme has obvious advantagesin that the difficult work of selecting and defining factors and degrees
has already been done and that existing manuals provide a useful guide
for such operations as the preparation of job descriptions and job
analyses, and the rating of jobs.
In the United States steel industry a new system was built from the
ground up by joint consultation and collective bargaining, initially
1
It may be recalled that under this scheme five degrees are distinguished for each
sub-factor. The system of weighting is illustrated by table 4 above.

JOB EVALUATION

between the United States Steel Corporation and the United Steel
Workers of America. The process by which this plan came into being
has been described as " an application of industrial engineering and
collective bargaining without parallel in American industry n.1 A joint
union-management committee first drew up job descriptions of more
than a thousand key jobs, starting from a proposal for improvement of
the industry's wage structure that had been submitted by 12 steel companies in 1943. The committee then worked out a system of job evaluation which, in accordance with the wishes of the union, differed from the
original employers' proposal in that it assigned much less weight to
skill and much more to responsibility. In the course of the difficult
negotiations which this entailed, a number of tentative understandings
were reached; these made it possible to proceed with further exploratory
work without waiting in each case for a formal agreement to be signed.
For instance, a tentatively agreed manual was applied to the classification
of jobs in a large steel plant; thus both parties had an opportunity to
test it before it was formally adopted. Special agreements were concluded
in the case of maintenance and repair jobs (which raised particularly
thorny problems) and for the establishment, in all of the company's
plants, of wage funds for the gradual reduction of inequalities in the
wage structure revealed by the job evaluation plan. In other words,
agreement was reached on the over-all cost of wage increases. In the
agreement of 8 May 1946 this was fixed at 3 5 / 8 cents per man-hour in
each of the company's plants (the increase being distributed among
individual jobs in such a manner as to reduce or eliminate wage inequalities).
Once it has been established what system is to be used, the most
important next steps are the description, analysis and rating of jobs.
This work is often entrusted to a committee, particularly in the larger
firms. Whereas job description can often be left to technicians, in consultation with the workers and their foremen or department heads (in
some cases the job holders themselves are asked first to prepare their
own job descriptions), final decisions as regards the contents, rating and
classification of jobs are in many cases taken by the committee. When
the union participates in the plan the committee will often be a joint
body, the members making independent ratings of the same jobs, which
are then compared and further discussed or simply averaged.
In many cases the committees include not only labour and management representatives, but independent experts as well. While the implementation of a wage programme directly affects the interests of both
1

TILOVE, op. cit.,

967.

METHODS OF JOB EVALUATION

management and the union, there are at the same time many purely
technical aspects to the matter. Questions concerning the exact description of a job can often best be judged by the foreman or department
head; general questions of rating may be judged most appropriately
either by outside consultants or by a staff specialist in the plant; and in
questions of wage-fixing, the personnel department or the economic department of the plant, if there is one, will normally have useful advice to give.
Procedures combining representation of interests with expert participation in the establishment of job evaluation systems are recommended
in the directives for the introduction of job evaluation in the iron and
steel industry of the Federal Republic of Germany.
An example taken from one West German firm may illustrate the
principle.1 According to a works agreement concluded in 1949 in this
undertaking, the labour manager is in charge of the introduction of
the system2, its implementation being entrusted to a committee. The
latter is composed of (a) the labour manager or his special representative, as chairman; (b) the chief of the economic department; (c) the
chief of the department concerned ; (d) a permanent representative of
the works council; and (e) a member of the works council competent
in respect of the department under consideration. Also participating
in the work of the committee, but without voting rights, are (a) one or
more workers' spokesmen 3; (b) one or more foremen 3 ; (c) one or
more members of the economic department3; and (d) in cases of doubt,
Dr. Euler, Dr. Stevens or a representative designated by them.4 The
committee is convened by its chairman. Minutes are taken of each
meeting and signed by the five voting members; the minutes are kept
as the only conclusive documentation regarding the introduction and
application of the system. The method is to be applied in accordance
with a number of principles discussed with Dr. Euler.
In the German Democratic Republic it has been recommended that
job evaluation schemes be prepared by a working party comprising the
officer in charge of the firm's manpower planning, the technician responsible for fixing " work norms " as the basis for payment by results, a
technologist and a representative from the efficiency department. When
1
Quoted in Haute Autorité, Communauté européenne du charbon et de l'acier:
La qualification du travail et ses applications dans la sidérurgie de la Communauté :
Allemagne (provisional mimeographed edition, Luxembourg, Sep. 1957), pp.130-131.
2
Under the 1951 Act concerning co-management in the coal and steel industries,
the labour manager is a member of the firm's managerial board, nominated by labour
representatives.
* Competent in respect of the jobs to be evaluated in the department, under consideration.
4
The system used in this case (as in most West German iron and steel undertakings)
is the Euler-Stevens method, already discussed.

JOB EVALUATION

analysing jobs in individual departments this group should call for advice
on the foreman and " a few workers " from the department concerned.
The possibility of using job analysis also as a means of improving efficiency and of determining how many workers are really needed in any
job is specially emphasised.1
In India the collective agreement of 8 January 1956 between the Tata
Iron and Steel Company and the Tata Workers' Union provides that—
. . . a Joint Committee consisting of an equal number of representatives of the
Company and of the Union shall be appointed with an independent expert as
Chairman for the work of job evaluation. The parties agree that the Joint
Committee shall consist of suitably qualified representatives of the Company
and the Union. The parties agree that the classification of jobs embodied in the
report of this Joint Committee shall be accepted by both parties as forming
the basis of the future structure of wages and other emoluments. Thereafter
the parties shall negotiate the rates of the future structure of wages and emoluments.
A job evaluation system, once established, has to be kept up to date.
There should be provision for the evaluation of new jobs that did not
exist at the time when the system was introduced and for the review
of jobs whose contents change as a result of reorganisation, new methods
of production and so on. In addition, some workers or the management
may feel that certain jobs were, after all, not properly classified and
should be reconsidered. In some cases all this work may be entrusted
to the committee that established the system in the first instance, thus
making it a standing committee. In other cases simpler procedures have
been adopted.
As an illustration of procedural arrangements following introduction
of a job evaluation system, some provisions may be quoted from an
agreement between the International Harvester Company (Canada)
and the United Steel Workers of America. Under this agreement a joint
union-management job evaluation committee was set up which—
. . . will continue to write job descriptions and evaluate all the remaining
occupations in the plant in the following manner:
(a) The Company will submit a draft of the job description for each
occupation.
(b) These job descriptions will be checked by the Union members on the
Committee.
(c) The Joint Committee will agree on the job descriptions.
(d) Each occupation will be evaluated by the Joint Committee on the basis
of the twenty-eight (28) benchmark occupations.
(e) In the event that agreement is not reached on either the description or
evaluation of an occupation, such occupation will be referred to the
Works Manager for a decision.
1
A. ERHARD and R. KÜHNEMUND: Zur Verbesserung der Planung des Arbeitslohnes
in den Betrieben der sozialistischen Industrie (Berlin, Verlag Tribüne, 1958), pp. 38 ff.

METHODS OF JOB EVALUATION

2.
3.
4.
5.

(f) In the event that agreement is not reached on an item at this point,
such matter will be set aside until all occupations have been considered,
at which time all such unresolved items will be reviewed and those not
then agreed upon will be submitted to arbitration.
All new or revised occupations will be described and evaluated by the Joint
Committee in accordance with the above procedure.
The description or evaluation of any occupation developed by the Joint
Committee by the above procedure will remain fixed unless affected by a
change that justifies revision.
The implementation of the new evaluations resulting from the program
will be withheld until agreement has been reached on the evaluation of all
occupations.
It is agreed that this program disposes of all outstanding grievances that
have arisen from the Job Evaluation Plan and that any adjustment in job
group numbers and wage rates as a result of this program will be made
effective from April 11, 1955, and employees involved will be reimbursed
to that extent.
The parties likewise agree that the objective date for the completion of the
Evaluation Program will be October 1, 1956, and that all new rates will be
made effective on and retroactive pay calculated as of November 1, 1956,
unless the entire re-evaluation is completed prior to that date. If, however,
this program is not entirely completed by that date, the program will be
continued.

The following relatively simpler provisions are found in an agreement
between the Republic Aviation Corporation (United States) and the
International Association of Machinists :
. . . The Company will furnish the Union with a copy of the hourly job rating
plan now in effect, including factor and degree definitions and point scores.
Section 32.3. In the event that the Company after the date of this Agreement establishes a new job or makes a change in any existing job under the
basic hourly job rating plan now in effect, the Company shall evaluate any
such new job or change in any existing job and shall establish and put into
effect its labor grade for any such job and shall immediately notify the Union
of its action, furnishing the Union with a copy of the job description sheet
for the job and the point scores for such jobs as established by the Company.
The Union may within ten (10) working days after receipt of such notice,
through a member of the Union's Job Evaluation Committee, request a joint
review of the evaluation of any such job with the Company. If a Union request
for such review is not received by the Company within such ten (10) working
days' period, the evaluation placed upon any such job by the Company shall
remain in effect during the term of this Agreement. If after a review no satisfactory adjustment of the evaluation of any such job is reached by a representative of the Union's Job Evaluation Committee and the Company within ten
(10) days after the filing of the request for review by the Union, the grievance
shall be referred to Step Three of the Grievance Procedure.

CHAPTER III
SOME PROBLEMS OF ANALYTICAL JOB EVALUATION
In the previous chapter the main methods of job evaluation have
been briefly described and some of the steps involved illustrated through
practical examples. One question that has not been discussed so far
is that of the application of the results of job evaluation in the fixing
of wages by converting points into wage rates. This question will be
discussed below. It would, however, seem desirable first to dispose of a
few other questions that have been raised in connection with the very
principle of using job evaluation in the determination of wage rates,
since it has even been contended that, whatever its merits for other
purposes, the method is not a suitable basis for wage-fixing. Questions
have also been raised in connection with the selection and weighting
of factors. A few aspects of both problems will first be discussed, especially with reference to the point rating method; some of these matters
are also relevant to other analytical methods.
TECHNICAL AND ECONOMIC ASPECTS OF
WAGE DETERMINATION

In Chapter I job evaluation has been described as a means of establishing a pattern of wage differentials that will be accepted as " equitable " and " fair " within certain limits, such as those set by a firm's
external environment. It has been noted that most people seem to
judge the fairness of wage differentials with reference to comparative
job contents in terms of qualifications required and adverse job conditions to be endured. An attempt has been made in Chapter II to show
how job evaluation aims at ensuring a fair wage structure according to
these criteria by comparing, and to some extent measuring, the demands
made upon the holders of different jobs, and by establishing on this
basis a hierarchy of jobs for wage-fixing purposes.
It has, however, been objected that the idea offixingwage differentials
in terms of " equity " is altogether ill-conceived. It has been argued that
relative wages, like relative prices for commodities, ought to be determined by the relative supply of and demand for various types of labour,

SOME PROBLEMS OF ANALYTICAL JOB EVALUATION

i.e. by economic factors. Any attempt to substitute ethical considerations and purely technical factors for these economic aspects is, according to this view, bound to cause serious difficulties.
For example, according to a study of the job evaluation scheme
applied in the aircraft manufacturing industry in southern California,
" the price of the job in the market is one well-established measure of
its worth. The S.C.A.I. plan established a different measure of its worth.
The conflict between these different standards made for continuous
obstacles to the consistent administration of the plan." x
Another critic, while admitting that a proper system of job evaluation
will give the firm, at the moment of introduction of the plan, " an up-todate knowledge of the duties and content of all the evaluated jobs and a
rationalisation of their relative values ", raises the following pertinent
questions regarding the lasting value of job evaluation:
How long can this situation last; how long can the enterprise accept with
confidence the value-relationship of jobs originally established by the plan;
are there any indications that this initial value-relationship will be disturbed
and become outdated? The answers to these questions depend mainly upon
the degree of economic change operating in the economy and the industry
of which the enterprise is a p a r t . . . .
Job evaluation, as we know it today, makes no provision to meet these
dynamic processes so prevalent in our economy. All the existing plans, whether
simple or complicated, whether technically sound or otherwise, have up to
now lacked a method by which external or internal economic changes and
structural adjustments can be automatically incorporated. Some authorities
fail to realise the need for this adjustive capacity, while others, realising it, are
prepared to accept the shortcoming. As a result of this technical limitation, all
existing plans retain the initial job-value structure throughout their operation.
Trying to solve this problem of change in a piecemeal fashion, certain " emergency " measures are introduced. Sometimes a complicated " maintenance "
procedure is devised by which individual jobs are re-examined, job descriptions
are revised. Where a discrepancy seems serious or where the economic change
affects the labour market so that the supply of a particular skill is threatened,
special bonuses are paid to retain or attract employees. This step is always
taken in the expectation that the disturbance is of short duration and things
will return to " normal " again. If in the meantime more jobs show similar
signs of manpower scarcity, the same piecemeal solution is adopted . . . .
The main weakness of job evaluation is that it cannot adjust adequately
and successfully to the dynamic elements of our economy as they affect the
process of wage determination. In the long run, in spite of " emergency "
adjustments to change, under the impact of the dynamics of the labour
market,
the whole job evaluation structure will disintegrate and collapse.2
1
C. KERR and L. H. FISHER: " Effect of Environment and Administration in Job
Evaluation ", in Harvard Business Review (Cambridge, Mass., Harvard University
Graduate School of Business Administration), Vol. XXVIII, No. 3, May 1950, p. 81.
s
L. G. NICOLGFOULOS: Formal Job Evaluation and Some of Its Economic Implications, Research Report No. 1 (Montreal, McGill University, Industrial Relations
Centre, 1954), pp. 36-41. The report was based on discussions in a seminar on job
evaluation attended by representatives of 29 leading Canadian companies.

JOB EVALUATION

These are serious objections and they should no doubt be borne in
mind when a firm introduces job evaluation as well as in the further
development of the system. At the same time, the defects which job
evaluation may be said to have should not be exaggerated. The following
remarks may help to put the criticisms quoted above into their proper
perspective.
For one thing, the statement that, as a general rule, " the price of
the job in the market is a well-established measure of its worth " hardly
constitutes a valid objection—if only because for many jobs there is
no such a thing as a readily assessable " market price ". The process of
wage determination cannot be compared in any simple way to the processes through which the prices of commodities are set. As pointed out
earlier, there are certain wage rates which cannot be fixed much below
those paid by other employers.1 This is especially so for " hiring jobs ";
but even for such jobs the employer will often be free to pay a higher
rate than that offered by others if he wishes to do so. In other words,
for this group of jobs the market may set a lower limit to the wage to be
paid rather than a precise, well-established price. Again, this lower limit
will often not be easy to define; workers may accept a job at a somewhat
lower wage than they could obtain elsewhere, for example if prospects
of promotion are better or conditions of work more pleasant. For jobs
other than hiring jobs the " market price " is even harder to assess ; often
a " market " can hardly be said to exist for jobs which are filled by
internal promotion. Briefly, while there are some jobs for which the
wage has to be fixed within a more or less narrow range determined by
what other employers pay, this range of discretion is much larger for
other jobs. For those jobs for which there are no clearly established
market rates the problem of a possible conflict between job evaluation
and the forces of supply and demand scarcely arises.
Of course, rates for the various jobs should be fixed in an acceptable
and more or less logical relation to each other and to those that are
determined externally. Viewed as a strictly mathematical problem, this
might be an impossible task. From this angle there might appear to be
more conditions, some of them perhaps even contradictory, than any
job evaluation formula could satisfy simultaneously. Practical wagefixing is, however, not a matter of mathematics or precision engineering.
As will be seen from the discussion below concerning the conversion of
1
In certain cases the influence of external rates may be very strong. Job evaluation
was introduced in the southern California airframe industry during the war, when
employment in the industry increased tenfold while other regional industries (particularly shipbuilding) experienced a similar rate of growth. Very keen competition for a
highly mobile labour force created a situation in which " market rates " came to play
an exceptionally important part.

SOME PROBLEMS OF ANALYTICAL JOB EVALUATION

points into actual wage rates, this process can be carried out in a very
flexible way. It will often be possible to accommodate to a large extent
both the internal and the external requirements of a firm's wage structure.
It remains conceivable that the wages for certain jobs might be so strongly
determined by those paid by other employers, that these wages might be
so completely at variance with broad negotiated principles reflecting
people's view of equity as laid down in the firm's job evaluation system,
or indeed that the factors to be rated in such jobs might be of such an
unusual or elusive nature, that the jobs could not possibly be fitted into
the job evaluation system in question. Thus, under few existing systems
would it be possible to rate satisfactorily the poise and beauty of fashion
models employed in a department store or even the special combination
of qualifications required from a hall porter in a first-class hotel. Assuming that job evaluation could be used as a basis for fixing appropriate
wage differentials for most of the other jobs in the department store or
in the hotel, the only possibility might be to exclude the few very special
jobs that could not be fitted into this system. The fact that such exceptional cases exist does not, however, mean that job evaluation as such
is useless. Of course, if a particular system does not afford a suitable
basis for the fixing of relative wages for any substantial part of the labour
force in a particular firm, that system ought to be abandoned or replaced
by another.
A further question is that of changes in job contents and in the
supply of (or demand for) particular skills in the employment market.
It is true that such changes occur and that many of them cannot be
" automatically incorporated " in systems of job evaluation. In fact,
they could not be automatically incorporated in any system of wage
determination. It should, however, be emphasised that job evaluation
as an aid in the fixing of relative wages is not intended to do away with
the normal practice of periodical review of a wage structure. In most
cases the level and structure of wages are reviewed at yearly or twoyearly intervals, as for instance when collective agreements have to be
renewed or when arbitration awards expire. When job evaluation has
been made part of the wage-negotiating procedures it is only normal
that such reviews should also cover the working of the job evaluation
system. Individual jobs may have to be regraded, the money value of
points may have to be changed and other adjustments may have to be
made. There is no reason why the application of job evaluation should
necessarily make the wage structure more rigid than any wage structure
negotiated without resorting to the method. The advantage that may
be claimed for job evaluation is that it enables such adjustments to be
made on a systematic basis and in a manner consistent with the whole

JOB EVALUATION

internal wage structure of the firm. In this respect, it provides a less
" piecemeal " solution than other procedures would do.
Finally, a new production method may call for skills not previously
required in the firm. The usual practice is to give a number of suitable
workers the necessary training at no cost to themselves. Often this
takes no more than a few days or weeks. In such cases the scarcity of a
particular new skill is not a very serious problem. In some cases a
specialised skill may, however, have to be recruited from outside at
wages that cannot possibly be reconciled with the factors, rates and
points of the firm's job evaluation system. It is by no means certain that
such cases are particularly frequent or important, but when they do
occur the jobs may have to be paid according to other criteria. Again,
this does not mean that the system has become completely useless;
normally, there will remain a large variety of jobs to which job evaluation
can still be applied as a basis for the fixing of wage differentials. It is not
claimed that job evaluation could solve all existing problems of wage
differentials, but it is claimed that in most cases the number of jobs to
which it can be usefully applied is large enough to justify its introduction
and maintenance.
These considerations suggest that, when job evaluation is applied
within an individual firm, any difficulties that might arise out of possible
conflicts between technical and economic criteria for wage determination
can usually be kept within manageable bounds. Conflicts of this kind
may, however, be more serious if efforts are made to apply job evaluation
(particularly a uniform job evaluation scheme) on a more general level.
In such cases greater importance would seem to attach to the questions
whether and how far the factors and degrees used in job evaluation
schemes reflect conditions of supply and demand in the employment
market, and whether they could be made to do so to a greater extent
than they actually do. These questions will be discussed in detail in a
subsequent section.
FACTORS, DEGREES AND WEIGHTS

Differences in Factors and Weighting
The point rating systems applied in various industries and countries
differ rather widely as to the number and nature of factors and subfactors used for evaluating jobs, the number of degrees distinguished,
and the weights attached to them. While the method developed by the
Belgian General Technical Commission used 32 criteria, many other
systems use little more than ten or 15 factors. As was mentioned in

SOME PROBLEMS OF ANALYTICAL JOB EVALUATION

Chapter II, the system most widely applied in the West German steel
industry places a premium on muscular effort, whereas in the United
States steel plan this same factor is given relatively little weight. Again,
whereas the maximum score for working conditions under the EulerStevens plan is 19 points out of a possible total score of 53, the HagnerWeng method (which has been recommended by the West German
metalworkers' union as an alternative to the former), though formally
unweighted, places much greater emphasis on working conditions.
These differences have been interpreted by some observers as evidence
of inconsistencies and contradictions in the selection of factors and
weights. One writer has stated the problem in the following terms :
. . . who can reconcile the differences in weighting between, for example, the
National Electrical Manufacturers' plan and the Carnegie-Illinois Steel Company plan ? The former has eleven factors; the latter twelve. In each case these
factors are logically placed in the four groups that are nearly universal in all
job evaluation plans, namely skill, effort, responsibility and job conditions.
But in the N.E.M.A. plan the skill group receives 50 per cent, of the total
points, while the Steel plan assigns only 31 per cent. But the N.E.M.A. plan
allots but 20 per cent, of the points to the responsibility group while the Steel
plan gives it 45 per cent. The other two groups of factors are weighted about
alike in the two plans. Now, it must be obvious that there will be some discrepancies between the final grading of a group of jobs as rated by these two
plans. If so, which plan is correct? And if there is no difference in grading of
the same jobs between the two plans then why do 1the experts bother to make
such a fuss over the factors and their weightings ?
This matter has also been discussed by another authority who, in
the course of a panel discussion on job evaluation, pointed out that if a
job evaluation scheme were developed by a committee there would be—
. . . as many different opinions as to what those weights should be as there were
people on the committee. But invariably some sort of compromise was achieved
by a pooling of their relative opinions. The very fact that they do not all agree2
to me is evidence of the fact that there is no weighting that is specifically correct.
The speaker then went on to raise the question whether in view
of this it was desirable to use any weighting. Why not, he asked, use
exactly the same scale for all factors? He then described an experiment
which he had carried out. He had taken a scheme covering about 80
jobs, with 11 or 12 weighted factors. He and his collaborators, acting
as a job evaluation committee, had assigned weights to each of the factors,
put them on chips, dropped the chips into a hat and stirred them up.
1
" Job Evaluation—A Discussion ", in Personnel Journal (Swarthmore, Pa.),
Vol. 28, No. 8, Jan. i950, p. 309.
2
Prof. H. B. ROGERS, quoted in Proceedings of the 18th Annual National Time and
Motion Study and Management Clinic (Chicago, Industrial Management Society,
1955), p. 80.

JOB EVALUATION

The weights were then re-allocated to the various factors in the order in
which the chips were taken out of the hat. A " scrambled " weighting of
the various factors was obtained as a result. It was found that there was
a better than 95 per cent, correlation between the results obtained by
applying the new scrambled weighting scale and the original weights.
The conclusion drawn from this experiment was that the idea of assigning weightings to the various factors might be an unnecessary complexity.1
Another authority has, however, contested the validity of this claim.
At the same conference he stated that he had tried out a similar procedure
on 2,000 jobs and the best correlation he could get was from 70 to 73 per
cent. These jobs covered a very wide range, including all kinds of machine
work and all types of fabricating, processing and inspection work in
various industries.2
Clearly, if sufficiently large differences are made in weights assigned
to the various factors, there will also be significant differences in the
ranking of jobs. For example, one author, comparing the United States
steel industry plan, with its heavy weighting for responsibility, with a plan
for light industries with the more usual emphasis on skill requirements,
found that certain important steel jobs would be ranked under the two
plans as shown in table 6.3
TABLE 6. COMPARATIVE RANKING OF JOBS UNDER STEEL INDUSTRY
PLAN AND LIGHT INDUSTRY PLAN IN THE UNITED STATES
Job

Roller, blooming mill
Machinist
First helper, open hearth
Common labourer
Assembler, light bench work

Light industry plan

Steel plan

1
2
3
4
5
6

3
1
4
2
6
5

. . . .

Source: STIEBER, op. cit., p. 39.

If in practice it turns out that many systems, though apparently
different in their choice of factors and weights, lead to practically the
same results, this would seem to indicate that fundamentally people are
in broad agreement as to the order in which various jobs should be
1

Proceedings of the 18th Annual National Time and Motion Study and Management
Clinic, loc. cit.
2
Statement by Mr. L. V. FISHER, ibid., loc. cit.
3
P. EDWARDS: " Statistical Methods in Job Evaluation **, in Advanced Management,
Dec. 1948, pp. 158 ff., quoted in STIEBER, op. cit., p. 38.

SOME PROBLEMS OF ANALYTICAL JOB EVALUATION

ranked, although they may give or seem to give different reasons for
their opinions.
Another matter concerned with weighting which has been considered
by some writers is the assumption implicit in analytica ¡schemes that
valid total scores can be obtained from the sums of the scores for the
individual factors, weighted or unweighted. This proposition has been
called " highly questionable " on the ground that it is based on the
" unverified assumption commonly made . . . that the effects of the
pertinent factors is additive . . . . This assumption implies in turn that
those factors are independent." 1
The International Conference on Job Evaluation held at Geneva in
1950 decided to approach these problems experimentally and called for
an inquiry aimed at determining the degree of comparability between
job evaluation schemes used in different countries applying the point
rating system. It was also decided that the results of the experiment
should be submitted to the Ninth International Management Congress
to be held at Brussels in 1951.
For the purposes of the inquiry, 20 representative industrial
jobs were selected and agreed definitions of the duties and responsibilities
of each were circulated to 12 job evaluation groups using different
systems in Belgium, Finland, France, the Federal Republic of Germany,
the Netherlands, the United Kingdom and the United States. The
definitions and descriptions of 17 of the jobs were taken directly from
the N.E.M.A. job rating manual.
The major objectives of the experiment were to determine (1) the
extent to which the various evaluations made by the different groups
differed from one group to another and from one job to another; (2) the
ranges of variation inherent in job evaluation; and (3) the usefulness or
otherwise of international comparisons and research of this nature.
After the necessary adjustments had been made for differences in the
schemes used by the 12 groups it was found that the evaluations made by
two French groups (there were five French groups altogether) were not
compatible with those made by the other groups. There were also
significant variations in the evaluations of three out of 17 jobs which.
suggested that major differences might exist between countries in the
duties and responsibilities of these jobs. The report submitted to the
Ninth International Management Congress concluded that the results
of the experiment had given—
1
S. B. LiriÁüfcK and A. ABKÜ¿ZJ: " Expci¡mental Criteria for Evaluating Workers
and Operations ", in Industrial Relations and Labor Review (Ithaca, N.Y., New York
State School of Industrial and Labor Relations, Cornell University), Vol. 2, N o . 4,
July 1949, p. 512.

JOB EVALUATION

. . . a clear indication of the need for further research and comparison of
job evaluation schemes on a practical basis. The results also indicate the wide
differences which occur between existing schemes and between the interpretations placed upon job descriptions by different groups of experienced assessors.
There is certainly a need for the development and adoption of a uniform
method of job evaluation if the present variances between groups and jobs are
to be reduced.1
There is indeed scope for further research along these lines; at the
same time, it should not be forgotten that many people regard the main
objective of job evaluation as the establishment of a more equitable
internal wage structure. People have strong ideas as to who should earn
more or less than (or the same amount as) other people and these opinions
have often given rise to grievances and disputes. Job evaluation seeks to
settle this problem by providing each firm with a logical and systematic
framework for correlating relative wages and relative job contents and
thus enabling it to construct and maintain a satisfactory wage structure
of its own. Viewed from this angle, the question whether the factors and
weights applied in one industry or firm differ from those used in another
would not seem very important. It is true that some systems (such as the
N.E.M.A. scheme in the United States, the Euler-Stevens method in
Western Germany and the standardised method in the Netherlands)
have been found useful in a variety of firms and industries. This suggests
that the criteria by which these systems evaluate different jobs in relation
to each other and the wage structure arrived at on this basis reflect rather
adequately the judgments that people in general apply when considering
the equity of wage differentials. It does not mean, however, that any of
these systems is " correct " in any purely scientific or philosophical sense.
For example, in the Netherlands the iron and steel industry, the metal
trades and the railways found other systems better suited to their particular needs than the standardised method. Is there really any need for
reconciling differences between the various schemes if each sector succeeds in settling its wage problems to its own satisfaction ?
It may therefore be asked what importance should be attached to the
finding that there are significant differences in the ranking of the same
jobs according to different systems and by different evaluators—differences which, it may be added, would probably have been even more
striking if it had been possible to extend the survey to countries with an
entirely different industrial structure, such as India, Burma, Mexico or
Chile? The answer, in the light of the foregoing considerations, would
seem to be that such differences might constitute a serious shortcoming
"Comité national belge de l'organisation scientifique: IXe Congrès international
de l'organisation scientifique, Vol. II: Comptes rendus (Brussels, 1951), VI-D, pp. 8-9.

SOME PROBLEMS OF ANALYTICAL JOB EVALUATION

of job evaluation (or at least to indicate that it is still a long way from
achieving its objectives) if the method were to be regarded as a purely
scientific undertaking concerned with the finding of objective truth.
Some of its earlier advocates may have given the impression that this is
indeed its purpose and this may, in turn, explain some of the trade
unions' mistrust of the method. Trade unions consider the fixing of
wages to be a matter for negotiation, not for scientific assessment and
certainly not a part of " scientific management ". The descriptions and
examples given above may, however, have made it clear that at least at
its present stage of development and application job evaluation is not a
scientific discipline, even though it does provide a basis for systematic
discussion of wage problems. In the words of the report of the International Job Evaluation Conference held at Geneva in 1950: " J o b
evaluation can be regarded as being ' scientific ' only to the extent that
it treats its problems in ' the spirit of science ', that is in an ordered and
systematic fashion." 1
This modest claim of job evaluation to be no more than a basis for
systematic discussion confirms the statement made earlier that the method
is not a substitute for, but rather a possible subject of, collective bargaining. When wage negotiators agree to apply the method, the particulars
of the system to be adopted, and especially the choice of factors and
weights, are matters of negotiation—not of scientific truth. This also
explains and justifies the cautious procedure under which a system is
first tried out on a number of key jobs. When the factors and weights
chosen in the first instance do not result in a ranking of key jobs satisfactory to both parties, it is the factors and weights that require adjustment, not the views of either party on what constitutes suitable ranking
of the jobs.
Finally, when job evaluation is regarded merely as an aid to systematic discussion of relative wages, the question whether factor ratings for
a given job can be added or not is likely to seem much less serious than
might appear at first sight. From a purely scientific point of view, the
problem is no doubt a challenging one, involving many complex and
even philosophical questions having to do with such matters as equity,
effort, and combination of physical and psychological job characteristics.
At the present stage, however, job evaluation is no more concerned with
such matters than wage negotiators are in their day-to-day practical
discussions, and for this more restricted purpose job evaluation seems
to have been quite a useful device in spite of its limitations.
1
Comité national belge de l'organisation scientifique: IX* Congrès international
de l'organisation scientifique, Vol. I: Rapports présentés aux sections (Brussels, 1951),
section VI, p. 2.

JOB EVALUATION

Differences between Groups of Jobs
As an aid in the establishment of suitable wage differences, job
evaluation is concerned with jobs for which questions of comparison
arise. As was noted in Chapter I, the degree of comparability between
jobs varies from one group of jobs to another. The term "job cluster "
was used to describe a group of jobs so closely related as regards the
nature of the work and other factors that close comparisons of work and
wages can be made.
When the nature of the work, the social background and status of the
workers, and the organisational framework in which the work is carried
out are all substantially different, there is generally much less need for
applying the same factors and weights to all the jobs. In practice it is
often found desirable to apply different systems of job evaluation to each
group. For example, the work of manual, clerical and managerial staff
is often considered to be so different that a uniform system would have
to include large numbers of factors that are essential for one group but
irrelevant for any of the others. For this reason separate systems of job
evaluation are sometimes applied in the same firm for manual and nonmanual workers. Senior executive jobs are often regarded as altogether
unsuitable for application of any method of job evaluation.
At present it is customary in most countries to fix individual job rates
at the level of the plant (possibly on the basis of minimum rates laid
down in national collective agreements), to treat wage and salary earnings
more or less separately, and to consider salaries of senior staff as being
beyond the realm of collective negotiation altogether. As long as this
is the case, it is to be expected that any particular job evaluation scheme
will be fairly limited in scope as regards the firms and groups of workers
covered. It is, however, not inconceivable that at some future date
people might grow into the habit of comparing their jobs and wages not
only with those of colleagues in their own department, but also with jobs
and incomes on the outside. In this case the problem of differences
between factors and weights under the different schemes might become
more important than it seems to be at the moment.
For example, Barbara Wootton has raised a number of questions that
could conceivably come to be asked more generally:
One may search in vain for the rational compound of skill, responsibility,
effort and working conditions of a system which would explain why the ward
sister in a general hospital should be paid (at the top of her scale) about onesixth of the salary of the Dean of Westminster; why the male probation officer
should start at about one-quarter of the top transatlantic airpilot's rate, and
at his maximum just pass the starting rate of the university lecturer in Greek;
or why the sub-officer in the fire brigade should end a little above where the

SOME PROBLEMS OF ANALYTICAL JOB EVALUATION

graduate school teacher began, and the police constable after six years'
experience get for his full-time employment aboutfive-sixthsof the salary of a
part-time governor of the B.B.C.1
The author was by no means convinced that job evaluation would be
capable of resolving such apparently irrational differences :
. . . even in the simplest cases the objectivity of job analyses is far from
absolute. At two points the contingent quality of the measures employed
betrays itself. The first is in the assignment of weights to the various factors
involved. This is obviously bound to be arbitrary: effort, skill, responsibility
and environment are mutually inconvertible. The second is inherent in the
translation of job-points into money-wages —quantities which are even more
patently inconvertible, for to attempt directly to translate job-points into
money-values would be like comparing Hyde Park Corner with members of
the House of Lords.2
However that may be, some efforts have been made to broaden the
scope of job evaluation schemes. In the Netherlands, for instance, the
method is already applied over a much broader field than has so far been
customary elsewhere, and suggestions have even been made for further
extension of the scheme. In that country, wages are controlled by the
Government which has to decide on claims for wage adjustments. An
important principle of the system has been that inequalities in pay for
equal work in different industries constitute grounds for wage increases.
This has led to the use of job evaluation in dealing with claims submitted
on these grounds and in comparing, for example, jobs and wages in
agriculture with those in manufacturing. 3
Whatever reservations may be made concerning the validity of such
comparisons, they at least have the merit of providing a systematic basis
for dealing with wage claims. The standardised method applied in the
Netherlands, at any rate, is believed to constitute a workable scheme
applying broadly the same factors and weights to job evaluation and the
study of income differences in a wide range of industries. Indeed, the
extension of job evaluation to executive and even entrepreneurial
occupations with a view to arriving at a more equitable distribution of
'Barbara WOOTTON: The Social Foundations of Wage Policy (London, George
Allen and Unwin, Ltd., 1955), p. 146.
2
Ibid., loe. cit.
3
While this is largely a peculiar feature of the wage policy pursued in the Netherlands, mention may be made of a not entirely dissimilar development in the United
Kingdom where, under the wage settlement of May 1958, the British Railways Staff
National Tribunal recommended that an inquiry should be made comparing wage rates
for the many grades of railway employees with those paid for comparable work in
outside occupations, as a basis for making adjustments in the railway rates. In an
article published in The Times of 18 August 1958 attention was drawn to similar
inquiries already undertaken by the Civil Service Pay Research Unit with a view to
comparing representative civil service jobs with comparable jobs in industry.

JOB EVALUATION

income has been recommended by the Netherlands Federation of Trade
Unions.1
Uniformity of factors and weights is apparently also regarded as
necessary in planned economies. According to one East German author,
jobs involving " labour of equal complexity " should be assigned the
same value irrespective of the firm or industry in which they are found
and of the persons evaluating them.2 By the same token, this author
opposes the practice of most other countries whereby weights and factors
are negotiated at the plant or industry level.
Application of Uniform Schemes to Broad Economic Sectors
When a uniform system of job evaluation is applied to a large proportion of a country's labour force and has a significant impact on the
national wage structure as a whole, the economic and technical problems
involved in the choice of factors and weights are somewhat different from
those which arise in the far more frequent cases where the method is
applied by individual firms using systems adapted to their own particular
needs. In the latter case the individual firm, operating within a far
broader economic context, has to adjust its job evaluation scheme to
certain external rates without normally being itself in a position to
influence these conditions to any appreciable extent. Suppose, however,
that job evaluation is adopted on a wider scale than it now is in most
countries and that uniform systems are increasingly applied to the fixing
of wage differences in several or all of a given country's industries—a
practice largely confined at present to the Netherlands and possibly some
of the planned economies of Eastern Europe, but which might conceivably be adopted in the future in other countries as well.3 In such a case,
instead of having to be adjusted to a given external wage structure, job
evaluation would itself to a large extent determine that structure. Would
there not be a risk, in such conditions, of job evaluation either proving
unworkable because of the failure of the factors commonly used under
the method to reflect supply and demand or, conversely, distorting the
wage structure, with labour shortages and surpluses, as well as maldistri1
Nederlands Verbond van Vakverenigingen : Wenkend Perspeclief (Amsterdam,
1957), pp. 75 and 83.
2

HAMPICKE, op. cit., pp. 40 ff. and Appendix III.

In Australia the national wage structure is determined largely by the awards of the
Commonwealth Commission (formerly Court) of Conciliation and Arbitration. While
no formal job evaluation scheme has been used for the fixing of wage differences
for skill and other job characteristics, the system raises questions somewhat similar to
those discussed here. A short note on the subject appears in Appendix C.

SOME PROBLEMS OF ANALYTICAL JOB EVALUATION

bution of the labour force among existing industries and occupations, as
the consequence? Although the immediate practical importance of these
questions may not be very great in most countries, a brief discussion
may help to bring into clearer focus some of the problems already
discussed in the section dealing with the economic and technical factors
affecting wage determination.
It may be noted, first of all, that modern employment markets do not
function like stock or commodity exchanges where small changes in
supply and demand have immediate effects on the price. Wage differences
among firms, industries and groups of occupations are not altogether
independent of differences in the supply of and demand for various types
of labour; but neither are they determined exclusively by such purely
economic factors. For example, it is widely believed that the reduction
in wage differences based on relative skill that has occurred in many
countries in recent years was due in part to such factors as the increasing
strength of semi-skilled and unskilled workers' unions and to widespread
cost-of-living adjustments under which the earnings of the lower-paid
workers increased proportionately more than those of the higher-paid
groups. Similarly, although shifts in demand and supply may have played
a part in the narrowing of wage differentials according to sex, to some
extent this tendency has been due to deliberate policy. In other words,
even if job evaluation could be shown to be incapable of adjustment to
short-term changes in supply and demand, this would not necessarily
mean that in this respect it would be worse than other processes of
wage determination.
On the other hand, it may be noted that job evaluation tends to rate
the more difficult and unpleasant jobs more highly than others, owing
to the widespread belief that particularly demanding or unpleasant jobs
ought to be better paid as a matter of fairness. At the same time the
number of people willing and able to perform difficult and repugnant
jobs is smaller than that of workers prepared to perform simple and more
pleasant duties. To this extent, job evaluation implicity takes account
of differences in the supply of certain qualifications such as a high degree
of skill, ability to assume a heavy responsibility or willingness to endure
particularly harsh working conditions. This is the main reason why job
evaluation systems in practice seldom clash violently with the requirements of the employment market.
This point may be further elaborated upon. Whereas certain qualifications are scarce, particular combinations of such qualifications may be
even scarcer. For example, there is a limited number of people prepared
to work on jobs exposing them to excessive heat. There are also relatively
few people able to perform jobs requiring outstandingly accurate judg-

JOB EVALUATION

ment. The number of people combining these two qualities may, however,
be no more than a minute fraction of the number of people satisfying
only one of these requirements. Hotel porters, in particular, have been
mentioned as an example of an occupation requiring an exceptional
combination of qualifications, such as a knowledge of languages, abihty
to deal promptly with unexpected and sometimes extraordinary emergencies, a considerable degree of tact and an excellent memory.
To some extent, of course, job evaluation takes account of such cases
by assigning points separately for each of the qualifications required.
Yet a combination of several of these qualifications in a high degree may
be so rare that it is difficult or impossible to rate the job concerned highly
enough as compared with jobs involving a high degree of only one such
qualification. It has therefore been suggested that in such cases higher
point values might be assigned to certain qualifications when found in
combination with certain others.1
Thus it is quite possible for a job evaluation scheme to reflect a
scarcity of desired qualifications in the employment market. The problem
is to determine how to do this in precise numerical terms. To this end a
wage-fixing agency operating a national job evaluation scheme would
have to decide on the specific factors and weights to be applied and on a
procedure for converting the evaluated job structure into actual money
wages, taking account of the broad requirements of the market as a whole.
If, for example, at the time of introduction of such a system the general
level of wage differentials according to skill in the economy is broadly
adequate, in the sense that there is no obvious shortage of skilled labour,
factors and weights will presumably be selected in such a way as to yield,
for most jobs, approximately the prevailing skill differentials. As in the
case of the individual firm, however, weights and degrees will have to be
reviewed from time to time to take account of changes in the supply of
and demand for particular qualifications. Generally speaking, such
changes are slow and there will normally be no need for very frequent
changes in the system. Factors that may bring them about include
gradual changes in a country's educational system, in the technical
structure of its industries and in people's attitudes towards particular
types of work. For example, rising living standards appear to be accompanied by an increasingly general distaste for occupations which involve
unpleasant working conditions, such as coal mining and (sometimes)
agriculture. These changes are generally slow and gradual, as are in most
countries the corresponding changes in the wage structure. This is
1
J. TINBERGEN: De criteria van de classificatie en hun weging, Studiedagen over
werkclassificatie in de ijzer- en staalindustrie der Gemeenschap, Rapport 6 (Luxembourg, Europese Gemeenschap voor kolen en staal, 1958), pp. 11 ff.

SOME PROBLEMS OF ANALYTICAL JOB EVALUATION

another reason why job evaluation systems applied to individual firms can
often be used for relatively long periods without major changes.
While a job evaluation scheme designed for widespread application
and based on a broad view of the general labour requirements of the
economy as a whole would give an indication as to what wage differentials
might be considered appropriate for remunerating certain general
qualifications such as skill, dexterity and willingness to endure unpleasant
working conditions, it could not reflect differences in supply of and
demand for individual jobs. For example, two jobs requiring exactly the
same level of skill, one in the textile industry and the other in the metal
trades, might be assigned the same point score and wage rate, even though
the demand might be much stronger in the metal trades than in the
textile industry.
It is commonly assumed that if industry A is willing to offer a higher
wage than industry B for jobs calling for the same amount of skill,
experience, endurance and other general qualifications, this can only
mean that the marginal productivity of labour in industry A is higher
than in industry B. This being the case, the higher wage paid in industry
A may be considered as a useful device for attracting additional workers
to high-productivity jobs; or, on the other hand, it may be considered
preferable to fix identical wages in the two industries, which will have the
effect of reducing the number of workers who can be profitably employed
in industry B and increasing that number in industry A.
Without discussing the relative merits of these two points of view, it
may be noted that proponents of the former have suggested that job
evaluation systems should include, in addition to the usual factors, a
further one expressing the state of demand in a particular industry or
occupation ; such a factor might, for example, be based on some measure
of the number of unfilled vacancies.1 Although this suggestion was made
with particular reference to a nation-wide system such as that applied in
the Netherlands, it might well be of even greater interest in connection
with systems applied at the level of the individual undertaking. As
mentioned earlier, it may not be necessary or even desirable that wage
differentials as between all individuals in a firm should reflect the state
of demand in the economy as a whole. The individual firm, however,
may often find it useful or necessary to try to attract new workers by
offering a somewhat higher wage, and the inclusion in job evaluation
schemes of a factor based on the actual demand for workers in particular
jobs might add a further element of flexibility to an already highly
flexible method.
1

TlNBERGEN, o p . Cit., p . 9.

JOB EVALUATION
PROBLEMS OF RATING

The rating of jobs is the operation of ranking them after having
compared the various job descriptions and analyses (a) with each other
and (b) with the factor and degree definitions. As mentioned earlier,
this is generally done by a committee; each member makes his own
ratings, any discrepancies being then removed by discussion, reconsideration of the facts as brought out by the job descriptions, and other means.
It has also been mentioned that there are two possible methods for
making the ratings. First, every job may be rated in respect of all factors
before proceeding to the next job; secondly, all jobs may be rated in
respect of one factor before proceeding to the next factor. The latter
method has the advantage of enabling the rater to make more accurate
comparisons between the jobs in terms of each factor, but the former is
also important as a check on the fairness of the ranking of each job as a
whole. In practice, both methods may be applied.
It is important to ensure that the rating is done with a good deal of
accuracy. If job evaluation is to result in a wage structure which is
accepted as equitable by the workers, the job ratings should at least be
capable of providing a sound basis for wage discussions, even though
in the course of these discussions it may be found necessary to make
changes in the original rating and grading of certain jobs. The major
difficulty of rating lies in the fact that there are few or no yardsticks for
measuring the degree in which a job calls for various requirements.
It is true that the amount of physical exertion involved in a particular job
can sometimes be measured by the weight of the objects handled, oxygen
consumption, expenditure of calories or other physiological criteria.1
On the whole, however, the practical possibilities of using such objective
measurements are limited, and in some cases—especially for factors such
as " mental effort " or even " skill "—virtually non-existent. Therefore,
rating is essentially a matter of estimation. This, of course, does not
mean that it is a matter of guesswork. Experience gained through the
widespread use of time study has in fact shown that it is possible to learn
the art of appraising various aspects of labour effort quite satisfactorily.
Accurate ratings can be achieved through proper training of those in
charge of this operation, use of carefully written manuals with definitions
and examples, and effective consultation procedures. One author lists
the following mistakes as commonly made in the process of rating 2 :
1
See WIBBE, op. cit., pp. 28 ft", and the literature quoted therein. This author
believes that scientific labour physiology could make important further contributions
to the problem of rating for purposes of job evaluation.
a
Elizabeth LANHAM: Job Evaluation (New York, McGraw-Hill, 1955), p. 220.

SOME PROBLEMS OF ANALYTICAL JOB EVALUATION

(1) " halo effect ", i.e. a tendency to assign a high or low rating to
several factors of a job simply because the first or the most important
factor was so rated ;
(2) " central tendency ", i.e. reluctance to assign high or low values
to most factors, and tendency to rate them at mid-points on the scale;
(3) leniency, i.e. tendency to assign high rather than low ratings in
case of doubt, or to avoid altogether assigning low values either to
individual factors or to the jobs themselves ;
(4) harshness (the opposite of leniency) ; and
(5) bias or prejudice.
In particular, strong evidence of leniency and " halo effect " has been
found in an analysis of 101 manual jobs in a variety of industries in the
Netherlands evaluated according to the standardised method mentioned
above. A suspicion of leniency is justified by the finding that very few
zeros were assigned, e.g. for knowledge, dexterity and burdensome
attention. As the authors put it—
t

... apparently the " normal adult male " is unable to carry out even simple,
unskilled tasks without having acquired some knowledge and dexterity, while
there are only few industrial jobs requiring so little attention that the normal
Dutchman would be prepared to give it to his job without some extra reward.
Probably there is also a certain disinclination on the part of the raters to assign
zeros.1
The authors also constructed a correlation table for the points allotted
to the various jobs for each factor. This table reveals an astonishingly
close correlation between the numbers of points assigned for knowledge,
independence and responsibility (each of which is heavily weighted under
the Netherlands method). There were few jobs which, if given high (or
low) point values in respect of one of these factors, were not also allotted
high (or low) values for the two others. This suggests that either the
rating of the jobs examined was influenced by a " halo effect " or that
some of the factors used could have been dropped, without affecting the
ranking of the jobs.
A particular case of leniency is noted by the West German metalworkers' union in a comment on the application of job evaluation in
steel plants.2 In this case, it was found that the remuneration for certain
jobs might have to be reduced as a result of technical change if the job
1
J. SANDEE and R. RUITER: " Beroepseisen van de industrie-arbeider ", in Economisch-Statistische Berichten (Rotterdam, Nederlands Economisch Instituut), Vol. 42,
No. 2106, 6 Nov. 1957, pp. 944 ff.
* Haute Autorité, op. cit., p. 263.

JOB EVALUATION

evaluation manual were followed strictly; accordingly, there was a
tendency to assign a higher point value to factors not or only indirectly
affected by the technical change.
THE DETERMINATION OF RELATIVE WAGE RATES

The final step, that of translating the evaluated job structure into a
structure of wage rates, is a decisive one. It must result in an acceptable
system of wage differentials. The subject of what constitutes a rational
wage structure has already been discussed in Chapter I. The requirements
which such a structure should meet are briefly recapitulated below.
(1) The rates should stand in a suitable relationship to comparable
rates paid by other employers. In particular— (a) hiring rates should not
be substantially lower than relevant external rates ; (b) other rates should
not be so much lower than those external rates with which workers are
apt to make comparisons as to cause serious dissatisfaction; and (c)
where the firm's wage rates exceed relevant external rates, the difference
should be kept within reasonable bounds.
(2) The firm's rates should stand in an acceptable relationship to each
other. For example—(a) when time and incentive payment co-exist, the
relative rates should be fixed so as to ensure an acceptable relationship
between the earnings of different jobs ; (b) particularly within narrow
job clusters, relative rates should be in accordance with people's views
concerning the equity of wage differentials compared with differences in
job contents; and (c) the new wage structure as a whole—and the differences between the new and the old systems—should be acceptable
to a substantial majority of the workers.
The meaning and importance of these rather abstract rules depend on
the circumstances of each case. For example, hiring rates may exist for a
few bottom jobs only, but they may also apply to a rather large number of
skilled maintenance jobs. " Relevant external rates " may be those paid
in other factories in the same region—but they may also be those fixed
by nation-wide collective agreements for the entire industry or for one or
more of the crafts employed in the firm. The relevant external rates for
some jobs may be found in an area quite different from that in which
relevant rates for other jobs occur. For example, for production workers
in an oil company regional rates for other production workers will normally be relevant ; but if the company also operates a tanker fleet, rates for
the crews will have to be fixed with regard mainly to other maritime
wages. In the United States local wage surveys are often recommended
as a means of collecting information on relevant external rates when job

SOME PROBLEMS OF ANALYTICAL JOB EVALUATION

evaluation is introduced. Such surveys are mutually advantageous to the
parties, and appear to be held quite regularly in many industrial centres.1
In a number of United States firms where no such surveys are made, the
introduction of job evaluation is linked to the firm's existing wage
structure. For example, the lowest existing rate may be assigned to a job
which ranks lowest in the evaluated structure and the highest existing
wage to a job which ranks highest, the rates for the other jobs being
suitably interpolated in accordance with their ranking.
In Europe less importance seems to be attached to extensive wage
surveys. The more widespread practice of negotiating wage agreements
on an industry-wide basis for a large area, often comprising the national
territory as a whole, is probably responsible for this difference. In
countries such as the United Kingdom and Western Germany, for
instance, collective agreements fix minimum rates, and these are of course
binding on firms introducing new wage schemes based on job evaluation
as well as on others. 2
As regards the internal aspects of the new wage structure, the choice
of factors and weights applied should normally ensure acceptable wage
relationships within the firm. If the system fails substantially in this
respect, it will probably be necessary to change it altogether. Nevertheless,
almost any new wage structure is liable to cause dissatisfaction in some
quarters if only because it changes traditional wage relationships. In
addition, the external requirements of the wage structure may clash with
some of its internal conditions. For example, the switchboard operator
of a factory may have to be paid at the same rate as employees of the
local telephone company even though that rate may be high by comparison with the firm's other office jobs; similar problems may arise on a
larger scale, e.g. in the case of maintenance and office jobs, as compared
with production jobs. 3 How such difficulties can be settled will be
discussed at a later stage.
Points and Wage Rates
In principle, the conversion of the evaluated job structure into a
wage structure is achieved by adding to a certain minimum wage rate an
amount which is determined on the basis of the point value allotted to
each job.
1
LYTLE, op. cit., pp. 236 ff.; and LANHAM, op. cit., pp. 269 ff. Both authors give
detailed indications as to the best way of organising such surveys.
2
Even in the United States, it may be recalled that the steel industry operates a
nation-wide job evaluation scheme. The widespread application of other systems,
such as the N.E.M.A. plan mentioned above, has also contributed to greater uniformity
in the national wage structure.
3
LlVERNASH, Op. Cit., p. 156.

JOB EVALUATION

Theoretically, there is an infinite variety of possible functional
relationships between points and wage rates. The most widely used
method is that of assigning for each point a fixed and uniform amount
of money, i.e. so many cents, pence or pfennigs per point. In a graph
measuring points scored along the horizontal axis and corresponding
wage rates along the vertical axis, the " wage line " connecting the points
representing individual jobs would be an upward-sloping straight Une.
The larger the money value for each point, the steeper the line would be.
One could, however, also choose any type of curved wage line. That
encountered most frequently is a curve becoming progressively steeper,
meaning that the money value per point increases as the number of points
allotted becomes higher. Such a procedure may be justified, for example,
on the ground that ability or willingness to perform jobs of increasing
difficulty becomes progressively scarcer, or that a progressive wage scale
provides an extra incentive for workers to seek advancement by acquiring
higher qualifications, or, more pragmatically, that such a wage structure
more closely resembles the existing one and is therefore less likely to
arouse opposition or to clash with the external wage structure. Some
authors, for example Hagner and Weng, object to a curved wage line
on the ground that it means paying the higher-ranking jobs better " at the
expense " of the lower and middle groups.1 They feel that the wage line
should be straight. Lytle, on the other hand, sees no justification at
all for a straight line and believes that—
... unions, if smart, know that a straight line overpays all intermediate jobs when
the end jobs are staked correctly
Managements suspect that some geometric
curve would be more correct but few managements
have the slightest idea as to
what geometric curve should be used.2
It may be doubted whether there is such a thing as a wage Une which
is " correct " in aU circumstances. As long as the application of job
evaluation tends to be limited to particular groups ofjobs (production or
clerical) with a relatively small range of variation in job conditions and
normal wage rates, a straight Une, it would seem, is hardly open to
serious objection. But if the same system has to accommodate a far
wider range of jobs, including labourers as well as managing directors,
difficulties may arise. In practice these are often avoided either by
excluding executive, supervisory or clerical jobs from the job evaluation
scheme altogether or by applying a different system in their case. It may
also be noted that the real issue is less one of constant versus varying
money values per point than of relative rates of remuneration. Practically
any desired wage structure could be constructed with a constant money
1

Op. cit., pp. 35 and 216; see also WIBBE'S comment, op. cit., p. 107.
* Loc. cit.

SOME PROBLEMS OF ANALYTICAL JOB EVALUATION

value per point, through a suitable choice of factors, weights and point
values assigned to successive degrees. While in many systems each succeeding degree is assigned a fixed number of points, in addition to those
allotted to the next lower degree, there is no real need for doing so:
for example, the points for degrees 1, 2, 3, 4, etc., might be 0, 2, 4, 8
and so on. Reference has already been made to a suggestion that a
certain point score might be allotted for a high degree of one factor
only where the same job also required a high degree of another factor,
in the event that combinations of a high degree of both qualifications
were particularly scarce. Conversely, whatever may be the factors
and weights used in a particular system, almost any wage structure can
be obtained by selecting an appropriately curved wage line. Thus, in
theory at least, two firms may evaluate identical jobs under an identical
system, and even have the jobs rated by identical persons, and still have
completely different wage structures.
Classification of Jobs into Grades
Under the point rating and factor comparison methods, it would be
logical to fix the wage rates for all individual jobs in accordance with
their rank or the number of points allotted to them, and this is sometimes
done; for example, it is done in four of the six steel plants whose job
evaluation schemes were analysed in the E.C.S.C. study mentioned earlier.1
In other cases, however, the jobs are classified into a smaller or larger
number of grades within each of which the same rate (or range of rates),
is fixed for all jobs. It should be noted that this classification of jobs
after—and not before—they have been analysed and evaluated is entirely
different from that under the grading or classification method of job
evaluation. In the latter case, a number of grades is established on the
basis of some general description of job characteristics and each job is
then placed in the grade where it seems to fit best.
The grouping of jobs into grades may be justified on several grounds.
For one thing, thefixingof a separate rate for each point score assumes
that job evaluation has achieved a degree of precision of which in fact,
at its present stage of development, it is quite incapable. Secondly, in
cases where a worker regularly has to perform different jobs slightly
differing in point value, he would have to be paid different wages every
time he changed from one job to another; this is avoided when the same
rate is paid for a group of jobs with similar point values. Finally, a
reduction in the number of different wage rates means a corresponding
reduction in the cost of wage administration.
1

Haute Autorité, op. cit., p . 127.

JOB EVALUATION

On the other hand, especially if the inter-grade wage differences are
large, difficulties may arise because workers become interested in
demonstrating that they should be classified in a higher grade. Some
workers will also fail to understand why, although their job has been
recognised to be more " difficult " than others, they should nevertheless
be paid the same wage ; since the primary purpose of job evaluation is
precisely to establish appropriate wage differences, such criticisms may
often be justified.
Chart I illustrates the classification of jobs into grades; all jobs which
score between 100 and 139 points are placed in grade 1, those from 140
to 179 in grade 2 and so on.
CHART I. TYPICAL STEPPED CURVE RELATING POINT VALUES
ASSESSED BY JOB EVALUATION TO THE WAGE SCALE
Wage scale
in money units
per hour

Wage scale
in money units
per hour

100|

"1100

4
Job grade numbers

100

150

—1—
200

250

Job assessments in points

300

350

SOME PROBLEMS OF ANALYTICAL JOB EVALUATION

INTRODUCING THE NEW WAGE STRUCTURE

When all jobs have been rated and their ranking has, through some
appropriate bargaining procedure, been converted into a structure of
wage rates, the next problem is that of changeover from the old wage
structure to the new one. In chart II jobs have been marked from left
to right in the order in which job evaluation would rank them according
to point values measured along the horizontal axis. The existing wage
rates are measured along the vertical axis. The dotted line indicates the
wage rates corresponding to point scores assigned under the job evaluation scheme. It has been assumed that the jobs are not grouped in
grades but that each job receives a wage rate corresponding exactly to
its point score.
It will be noted that some of the dots representing existing wage
rates fall above the dotted wage line and others below it. The usual
practice is for the wage rates below the line to be brought up to it as
soon as possible, this depending mainly on the total cost involved and
CHART II. WAGE CURVE SHOWING A PROPOSED WAGE SCALE
AND EXISTING WAGE RATES IN A PLANT
Money units
per hour

Money units
per hour

16 •

100

-I.
150

-I200

-u
250
Job assessments in points

Notes.
1. The dots represent existing individual wage rates.
2. The line
is the trend line for these rates.
3. The line
represents the proposed new scale.

300

350

JOB EVALUATION

on the over-all wage increase that the firm is able and willing to make
at the time when the new wage structure is introduced. Mention was
made earlier of the agreements in the United States steel industry under
which certain amounts of money were set aside for this purpose.
The rates above the wage line (often called " red-circled rates " in
the United States, " encircled rates " in Germany and " red plusses " in
the Netherlands) are more difficult to deal with. Generally speaking, it
is neither desirable nor possible to reduce such rates to the level corresponding to their point scores. This is the main reason why introduction
of job evaluation usually involves an over-all wage increase: some rates
are raised but none are reduced. Some further observations on the
way in which red-circled rates may be dealt with are made below, but
first a few other, more general questions will be discussed briefly.
There are many ways of approaching the problem of adjusting the
wage structure while neither exceeding the financial limits within which
the firm has to operate nor reducing existing rates. One of these is to
spread the process over a longer period. Each new agreement on wage
increases may then be used to raise " below the Une " rates while " above
the line " rates are kept constant or are raised by smaller amounts. For
example, a large engineeringfirmin the United Kingdom, having adopted
a job evaluation scheme designed to provide what it considered an ideal
wage structure, took more than ten years to change over to the new
system.
Especially if the difference between the existing wage structure and
that resulting from the job evaluation scheme is very large (a high proportion of the existing rates lying far above or below the new wage line)
quick adjustment will not only be very costly but the drastic changes in
the wage structure may also cause discontent. Though aware of certain
anomalies in the existing system, people may not be prepared to accept
wholesale changes. If such a situation arises it should be asked whether
the system is after all a good one, and whether a different selection of
factors and weights would not have resulted in a more acceptable and
less costly ranking of jobs as compared with the existing structure.
When the ranking of the jobs is, however, accepted as being generally
fair, the cost of the system may be varied depending on the shape of the
wage line. Reference has already been made to the possibility of using
either a straight, curved or stepped line. If a straight Une is chosen, its
slope, starting point and the breakdown of job grades can still be varied.
The location of a wage Une of a given shape is determined by the level
at which the lowest wage rate is fixed. The slope of a straight Une depends
on the money value per point, which may be fixed at a larger or smaUer
amount. In some cases the wage Une does not begin to rise before a

SOME PROBLEMS OF ANALYTICAL JOB EVALUATION

relatively large number of points is reached. For example in table
7, which illustrates the grading scheme used by the N.E.M.A. and the
National Metal Trades Association in the United States, all jobs carrying
up to 139 points are grouped together in the lowest grade and each
higher grade covers a group of jobs within a range of 22 points.
TABLE 7. JOB GRADES AND POINT VALUES UNDER THE N. E. M. A
AND NATONAL METAL TRADES ASSOCIATION SCHEMES
IN THE UNITED STATES
Points

Up to 139
140-161
162-183
184-205
206-227
228-249

Grade

.
.
.
.
.
.

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

.
.
.
.
.

12
11
10
9
8
7

Points

250-271
272-293
294-315
316-337
338-359
360-381

Grade

6
5
4
3
2
1

Depending on how the various grades are fixed on the basis of point
scores, the number of jobs placed in each grade will vary and the total
wage bill will also be different. For example, if the rates for grades 2,
3 and 4 are $1.60, $1.80 and $2 respectively, total wages will be lower
if the point ranges for the grades are 80-110, 110-125 and 125-140 respectively than if they are 80-100, 100-120 and 120-140. In the latter case
each grade will include some jobs that in the former case would have
been placed in a lower grade.1
When these questions have been settled there will normally remain
a core of jobs that do not fit into the wage pattern. Some jobs, in particular, may have been rated too low either with respect to external
rates or as compared with established internal wage rates. In such
cases further adjustments may sometimes be made by re-evaluating the
job and, where necessary, raising its point score. Or it may be possible
to change the contents of a job by adding one or two elements designed
to raise its point score, if that can be done without impairing operating
efficiency. Another possibility is to base the wage rate partly on factors
not assessed by job evaluation and to introduce or maintain the payment
of some sort of extra allowance or bonus.2 If it proves impossible to
fit a job satisfactorily into the wage structure through any of the above
means, the job may be left entirely out of the scope of the scheme.
1

For an illustration of this point see charts III and IV below.
This may be necessary in any case. Thus, it has been contended that job conditions in coal mining are so variable that it is impossible to apply a predetermined
grading scheme for this factor to underground jobs. Job conditions can then be
taken out of the job evaluation system and made the subject of separate, more or ¡ess
ad hoc allowances. In most of the steel plants surveyed in the E.C.S.C. report referred
to above, special temporary efforts which cannot be foreseen in the process of evaluating jobs are also remunerated separately (Haute Autorité, op. cit., p. 125).
2

JOB EVALUATION

Some people may regard certain of these devices as somewhat contrived and not quite compatible with the basic philosophy of job evaluation. It is quite true, of course, that if one regards job evaluation as
forming in itself a sufficient basis for the establishment of " equitable "
wage rates or for measuring the " true " value of a job, any admixture
of trial and error or compromise—not to mention outright exclusion
of certain jobs from the process—can only be regarded as unjustified
tampering with the basic principles of the method. The more common
view of job evaluation is, however, that it is no more than an aid to,
or one of the factors in, the process of wage determination. The factors
determining a rational wage structure are so numerous and complex,
and some of the forces limiting the range within which rates can be
fixed so strong, that it would be unreasonable to expect any system of
job evaluation to make due allowance for all of these factors or, for that
matter, to expect any firm to base its wage structure exclusively on a
particular job evaluation system if that were to cause undue internal or
external difficulties.
Reference has already been made to the problem created by jobs
classified and paid more highly than the scheme would justify. Reduction of rates is, of course, a painful operation and apt to cause considerable discontent. In most cases measures are therefore taken to avoid
such reductions and to announce this intention at an early stage in
order to forestall fears and opposition, as well as excessive efforts on the
part of workers to overstate the requirements of their job.
There are various ways of making reductions in wages unnecessary,
such as those resorted to in the German steel plants surveyed in the
European Coal and Steel Community study mentioned above.1 In two
of these plants the workers concerned were given other jobs with a
higher point score; in other cases conditions of production were improved
so that the workers could maintain their earnings thanks to higher
productivity. In three plants the same wages continued to be paid, but
the differences between these and the rates based on job evaluation
were partly offset by excluding the workers concerned from subsequent
wage increases. In some cases wages were maintained by introducing
special premiums in addition to the new wage based on job evaluation.
Finally, in one plant the difference between the old and new rates was
gradually reduced over a six-month period.
Unless the old wage structure was very inefficient, thefinancialburden
of maintaining the earnings of overpaid jobs is relatively small. Therefore,
even when there is no possibility of dealing with the problem by pro1

Haute Autorité, op. cit., p. 127.

SOME PROBLEMS OF ANALYTICAL JOB EVALUATION

motion, transfer, retraining or in some other way, there is usually no
compelling financial reason for reducing wages, and the firm may maintain the old rates until a new person takes the job, simply regarding the
amount of overpayment as " the penalty of earlier mismanagement ".1
Some trade unionists in the United States have claimed, however,
that this has resulted in downgrading and has been a means employed
by some managements to reduce the total payroll. It has been contended
that if, for example, there are three jobs in a department paid at $2, $1.75
and $1.50 per hour respectively (a total of $5.25) and the worker on
the $2 job has higher qualifications than are necessary the job may be
re-rated at $1.75 an hour. The $1.50 man is then put on to this job
(and is therefore upgraded) and a new man is assigned to the $1.50 job.
The $2. man is transferred to a $2 job in another department. The total
wage bill for the first department now amounts to $5 instead of $5.25.
Later on it may be found necessary to abolish the job the $2 man is
holding. Because of the operation of seniority rules he will be given the
chance of going back to his old job but at the new rate of $1.75. He will
now " bump " his replacement who will go back to his former job at
$1.50, thus bringing about the dismissal of the new man who was engaged
to fill this job.
Introduction of Job Evaluation in British Coal Mines
Sometimes the introduction of job evaluation results in a complete
upheaval of the old wage structure. In the British coal mines job evaluation had to be undertaken for a wage and salary structure described as a
" cobwebbed and rickety tangle, almost without rhyme or reason ".2
Some jobs, which by common consent should have had the lowest
remuneration, were paid at rates exceeding those of other jobs which
evaluation had ranked much higher. On the other hand, even at individual collieries there were considerable differences in the wage rates
paid to men performing identical jobs.3 The problem was further complicated by the large size and varied nature of the undertakings; the 850
collieries operated by the National Coal Board were scattered over
1

LYTLE, op. cit., p. 295.

* SALES and DAVIES (quoting The Economist), op. cit., p. 214.
3
It is interesting to note this in view of the argument sometimes advanced that job
evaluation ignores market forces of supply and demand and that collective bargaining
conducted without job evaluation results in a " natural ", " economic " or " intuitive "
system of wage differentials. Sales and Davies feel thai " it is difficult to credit that
such situations arise mainly as responses to economic stimuli; it seems more plausible
to attribute them to the subjective assessments and preferences of management,
unrestrained—at least within wide limits—by market considerations " (op. cit., p. 212).

JOB EVALUATION

many different parts of the country with widely differing wage and job
structures.
The problem, after all jobs had been analysed and classified into a
number of grades, was (a) to find appropriate rates for these grades,
(b) to avoid widespread dissatisfaction and (c) to keep the cost of the
revision of the wage structure within reasonable bounds. As regards the
first point, the existing wage structure gave no guidance except by way
of national minimum weekly wages. As to the second point, it was
considered that, in order to be acceptable, " changes in wage rates must
be one way—upwards ". It was also believed desirable to make as
many people as possible benefit from the change because in this way
" the impact of some men getting more than others would be lessened ".1
At the same time, the high cost involved in a complete reshaping of the
wage structure had to be borne in mind; in other words, the extent to
which the desirable wage structure could immediately be brought into
being was limited by budgetary conditions.
The problem of distributing a given aggregate wage increase in such
a way as to award the maximum number of workers some increase while
establishing meaningful differentials between the new grades was approached with mathematical techniques, graphical methods and special
tables enabling quick and precise answers to be given to such questions
as what would be the distribution of wage increases by grades, by
districts and by numbers of men affected, and the cost which various
possible patterns of wage adjustment would involve. Over a hundred
such possible patterns were examined and compared, careful attention
being given to their respective advantages and disadvantages. This
formidable undertaking apparently proved quite worth while, as evidenced
by the ease and remarkable lack of friction attending the introduction
of the wage structure eventually selected.
This final wage structure included the following features. Instead of
the old pattern of minimum rates supplemented by an involved system
of more or less " personal " rates, " standard " rates were established
for each grade. The standard rate was the only one recognised for the
job and grade concerned, and rates below it were immediately adjusted.
Standard rate differentials were fixed temporarily (in view of the limited
resources available) at one shilling per hour between successive grades.
Secondly, a general flat wage increase was awarded to all workers,
provided that the wage rate thus arrived at might not exceed the limit
for their grade. Those workers whose rates were already equal to or
above the limit " marked time ", i.e. they got no increase. The differences
1

SALES and DAVIES, op. cit. p.

216.

SOME PROBLEMS OF ANALYTICAL JOB EVALUATION

between actual and standard rates were expected eventually to disappear
as a result of two factors : first, a worker was entitled to his personal rate
only so long as he remained in the same grade at the same colliery;
secondly, personal rates would tend to be overtaken by a succession of
later general wage increases.
Job Grades and Rate Ranges
When jobs are classified into grades, the question arises whether the
remuneration for all jobs in each grade should be fixed as a single rate
or as a rate range with minimum, maximum, and a smaller or larger
number of intermediate rates.1 The advantage of a single rate is its
administrative simplicity. The disadvantage is that a single rate leaves
no scope for remunerating differences in performance and in seniority.
This may cause dissatisfaction among older workers who believe that
their longer experience and their loyalty to the firm entitle them to higher
earnings. In addition, a rate range may provide some incentive for
workers to seek an increase in wages by putting forth their best efforts.
Sometimes rate ranges can be used to make a system more flexible;
for example, new workers may be recruited at a fairly high rate within
the range instead of at the minimum rate.
It will be clear that this question is somewhat beyond the scope of
job evaluation problems because it relates to remuneration of persons
rather than of jobs. When an improvement of the wage structure is
sought through the introduction of job evaluation it is, however, natural
to ask whether the introduction of rate ranges might not further improve
the wage structure. Alternatively, when afirmintroducing job evaluation
already has rate ranges, it is necessary to consider how these can be
fitted into the new wage structure without defeating the objectives of
job evaluation. Here again there are many possible arrangements to
choose from. The two most important aspects of rate ranges are their
spread (which may be expressed either as the difference between the
maximum and minimum money rates or as a percentage of the minimum)
and their relative position, i.e. whether, and how far, successive ranges
overlap.
When there are no special reasons for deciding otherwise, it may be
regarded as equitable for workers in jobs at all grades to be given equal
opportunities of advancement by fixing the maximum for each range as
a uniform percentage of the minimum and dividing each range into an
1
It is not necessary to fix uniform rates for all jobs in a given grade: differing rates
(or rate ranges) can also be fixed depending on the job.

JOB EVALUATION

equal number of steps. In practice the maximum is often fixed at about
50 per cent, above the minimum, with four or more steps in between.
Rates in the lower half of the range may then, for example, be paid to
new workers who are still being trained, and rates above the mid-points
may be paid on the ground of seniority or of especially deserving performance, or both. The justification for advancement on the ground of
seniority may be that with length of service there is a normal increase
in individual proficiency; special merit increases may be paid for superior
progress.
Wage ranges may conflict with the objectives of job evaluation, for
example as a result of excessive overlapping. When the ranges overlap
to a large extent, jobs placed in very different grades may be paid the
same rates even though in the process of job evaluation they were rated
very differently. Excessive overlapping may occur, for example, when
the number of seniority steps is larger than the normal scope for individual
improvement would justify.
In charts III and IV two examples are shown of wage structures
based on job evaluation and using job grades and overlapping rate
ranges. The examples were taken from two large aircraft manufacturing
plants in the United States employing 20,000 and 13,000 workers respectively, the operations performed in both being substantially the same.
Both plants used job evaluation schemes based on exactly the same
factors (although in one case slightly more weight was given to effort
and job conditions than in the other—a difference unlikely in itself to
affect the relative point scores substantially). Both systems also used the
same number of job grades and broadly the same total range of wage
rates. In both cases there was a great deal of overlapping, the minimum
rates being in several cases below the maximum for one or more lower
grades (although in neither case was there any overlapping, for example,
between grades 7 and 11, both of which include a large number of
workers).
The two cases, however, differ, considerably as regards the distribution of the labour force among the various grades. For instance, chart IV
shows 30 per cent, of the staff as being in grade 7, as compared with only
about 20 per cent, in chart III. While no indication is given as to the
point ranges covered by each grade, comparison of the two charts may
serve to illustrate the earlier observation that the total wage bill resulting
from a given job-evaluated wage structure may differ considerably
depending on how the total range of point scores is divided into job
grades. Attention may also be drawn to the fact that, in the case illustrated by chart III, special premiums were paid to workers classified in
grades 1, 2, 3, and 4.

SOME PROBLEMS OF ANALYTICAL JOB EVALUATION

CHART III. DISTRIBUTION OF THE LABOUR FORCE IN A UNITED
STATES AIRCRAFT PLANT EMPLOYING 20,000 WORKERS
Hourly wage
rates in dollars
2.50

•r
JftT e

<?J

Pol

D 1

• "

[

— i

"©

• . .

100

60
50
40
30
Cumulative percentage of total labour force

Negotiated premium.
j | ¡ ¡ = Rate range of labour grades.

JOB EVALUATION AND INCENTIVE WAGES

As was emphasised in Chapter I, job evaluation is concerned with the
rating of jobs, not of the work of particular persons. Individual effort
and performance may, but need not, be rewarded separately through a
system of merit rating or payment by results. In principle, the division
seems to be clear. The requirements of " normal performance " on any
job can be defined and rated according to some system of job evaluation
as a basis for rate-fixing. Performance which is either above or below
standard can be rewarded or penalised according to some system of
premium payments which will necessarily affect earnings. In practice,
however, the matter is not so simple. A number of difficulties commonly
encountered are discussed below.

JOB EVALUATION

CHART IV. DISTRIBUTION OF THE LABOUR FORCE IN A UNITED
STATES AIRCRAFT PLANT EMPLOYING 13,000 WORKERS
Hourly wage
rates in dollars
2.40

CTS

235

2.30
2.25

2.20

2.15
2.10

1(2H(^B

E SiiÄ

(bj

2.05

2.00
1.95

Te] wSBÊÊËSÈ^jÊÈffÊÈBÊÊÊlÈia

1.S0
1.85

sÉjICjO

1.80

[io BBS

1.75
1.70

—BEE

1.65
1.60
1.55
1.50

p—©

1.45

100

Cumulative percentage of total labour force

The two main problems arising when job evaluation is to cover also
incentive workers seem to be (a) the practical difficulty of determining
" normal performance ", especially when the fixing of incentive wages
is the subject of collective bargaining, so that agreement on this question
is needed; and (b) the fact that fluctuations in incentive earnings cannot
easily be controlled.
The Problem of " Normal Performance "
Job evaluation looks at comparative job contents. One job requires
greater effort or entails a greater degree of responsibility than another
and the difference may be expressed in general grade descriptions (under
the classification method) or in factor degree definitions (under the
point rating method). When one of the jobs is on a time wage and the

SOME PROBLEMS OF ANALYTICAL JOB EVALUATION

other on a piece rate, the same comparison has to be made, but with the
qualification that there are now two reasons why earnings may differ,
namely—(a) because the jobs are differently rated, and (b) because of
differences in workers' individual performances. An attempt has to be
made to eliminate the effect of the second factor by comparing the jobs
on the basis of a " normal performance ".
Suppose it is found that in the past the piece worker earned 20 per
cent, more than the time worker, although job evaluation rates the two
jobs as equivalent. If this difference can be attributed entirely to the fact
that the piece worker's individual performance was superior to a " normal " performance accurately determined by means of work measurement, it can be concluded that the piece rate was a fair one. In other
cases, job evaluation will have shown that the piece rate was either too
" tight " or too " loose ". Thus, if job evaluation is to cover piece rates,
difficulties are hable to arise unless it is possible through work measurement to determine normal performance in an acceptable manner.
The problem is comparatively simple in the case of certain wellestablished operations not subject to frequent or rapid change. Copytyping is a good example. Even so, when job evaluation is initially introduced, the standard workload fixed under the incentive scheme—
e.g. six pages per hour—must be reviewed and, if necessary, revised so
as to bring the piece rate into line with other piece or time rates. In other
words, the employer and the trade union must reach agreement on how
many pages per hour represent a copyist's " normal performance," and
this may prove rather difficult.
On the other hand, piece rates in many industries (e.g. apparel-making,
engineering and coal mining) have to befixeddaily for jobs that did not
exist before or that differ from existing ones in the quality of the cloth
to be cut, the degree of precision required, the tools supplied by the
employer, or in other respects. In such cases piece rates may be determined by rate-fixers on the basis of rough experience, with no great claims
to precision and with no clear-cut aim beyond that of striking some sort of
balance between " loose " and " tight " rates so that the average worker
can get enough different jobs of both sorts and earn some suitable overall bonus. Even when work study is applied (and rapidly changing job
conditions do not always permit this), the rates are not necessarily
accurate and mutually consistent. In such cases the objective of job
evaluation can easily be upset by incorrect piece rates.
When pieces rates have to be fixed frequently and quickly, it is very
difficult to ensure that relative earnings always correspond to evaluated
job contents. Job evaluation may rate two jobs equally high, but suppose
that each job comprises a variety of tasks differing, for example, according

JOB EVALUATION

to the texture of the cloth and the precise pattern to which it is to be
cut; or according to the time during which the specific task is to be
continued, so that there may be more or less opportunity for gaining
experience and improving individual output. These and similar factors,
and the accuracy with which they are foreseen when the piece rate for a
task is fixed, may make rates " tight " or " loose ". When the rate for
one job is tight, the same effort results in lower earnings than on a job
for which a loose rate has been fixed.
Sometimes a relatively stable operation (for example steel rolling)
has been performed for a long time by crews paid according to some
group incentive scheme under which each man receives a certain percentage of the total bonus. These percentages may have been determined
in part by the degree in which each member is deemed to be able to
influence the group's output rather than, for example, the effort or skill
required on the job. Thus some men may earn as a bonus a higher percentage of their basic pay than others, and relative earnings may differ
from relative basic wages. It is possible, moreover, that in the course of
time, with slight changes in equipment or working methods, various
special allowances and bonuses have been added to individual wages, so
that in the end relative earnings and relative basic rates are completely
different. Job evaluation may then be applied to rate the jobs and " correct " basic rates. But it may then be necessary also to revise the whole
" inherited " system of premiums and bonuses. Technically this might
not be a very difficult task but it is apt to raise complicated human
problems. Perhaps a close-knit group may have its traditional earnings
relationships altered drastically, and bonuses and premiums that had
been regarded as justified in some sense may be abolished or reduced.
The problems that may arise are particularly well illustrated by the
history of the protracted and, in the end, unsuccessful attempts by the
United States steel industry to include incentive wages in its otherwise
impressive wage inequities programme. 1
In 1945 and 1947 the United States Steel Corporation and the union
agreed that the basis of incentive wages would be a " fair day's work ",
defined as " that amount of work that can be produced by a qualified
employee when working at a normal pace and effectively utilising his
time where work is not restricted by process hmitations ". Parties
would develop " guideposts " or principles for determining what this
" fair day's work " was to mean in practice. Serious efforts to that
end started in 1947 and continued until mid-1950, "when the situation
was recognised as hopeless on both sides ".2 This disappointing conclu1

This experience is described in detail in STIEBER, op. cit., pp. 175-231.

STIEBER, op. cit. p.

184.

SOME PROBLEMS OF ANALYTICAL JOB EVALUATION

sion is the more significant since, on the one hand, about half the production and maintenance workers in the industry were on incentive
wages and, on the other, the parties had been able to reach agreement
on both the general principles of job evaluation and their application
to time workers.
At the same time it is clear that agreement on any " guideposts " for
the setting of wage rates on the basis of such question-begging concepts
as " qualified employee ", " normal pace ", " proper allowances for rest
and personal needs " and " idle time " is intrinsically most difficult to
achieve. The impact of any precise definition agreed at an industry or
company level on thousands of piece rates and other incentive wages
fixed at the plant and department level can hardly be foreseen by either
side (and perhaps particularly by the trade union) with sufficient clarity
to enable parties to renounce the right of ad hoc negotiation in individual
cases.
Even while negotiations on the matter were in progress, something
had to be done about existing incentives and any new ones that might be
adopted. In 1947 the parties agreed that existing systems were to
remain in effect until they were replaced by mutual agreement or, in the
following cases, by the employer who, in discussion with local unions
and subject to the existing grievance provisions, was—(a) allowed to
establish incentives for new and " submerged " 1 jobs, and (b) obliged
to do so in the case of new conditions resulting, for example, from
improved methods of work or from the introduction of new equipment.
In carrying out this latter task the company used performance
standards laid down in a manual of its own which, it had hoped, would
form a basis for agreement on the general question of a fair day's work.
This phase of the programme, however, also raised serious difficulties.
In particular, the revision of rates by the company, where compulsory
under the 1947 agreement, was a frequent cause of friction. It raised,
of course, a general problem common to all incentive systems—that of
deciding at what point conditions had changed sufficiently to make a
new incentive system necessary. Apart from this, however, the new
incentives often yielded lower earnings than the old ones. True, in
accordance with another provision of the 1947 agreement, the incumbents
of the jobs concerned were paid " out-of-line differentials " so as to
maintain hourly earnings at the average level of the three months preceding the change in the system, and even when, due to higher performance, a worker reached or surpassed his old earnings on the basis
of the new incentive, the differential continued to be paid. The effect of
1
Defined as jobs on which earnings were lower than the applicable standard
hourly wage rates (STIEBER, op. cit., p. 177).

JOB EVALUATION

this practice was, of course, to maintain the incentive element in the
new rates, but it was interpreted by union officials as an attempt to
make the incumbents accept new rates that really required more than a
fair day's work from their successors. It was also feared that the company was trying to reduce crews and in general to alter established
" work rules V
Following recommendations by the War Stabilisation Board in
connection with a dispute, the parties finally agreed in August 1952 to
drop their undertaking concerning the establishment of " guideposts " for
setting standards of performance and defining " a fair day's work ".
Furthermore, they agreed that out-of-line differentials on incentive
jobs were to be paid to all workers on those jobs and not to " present
incumbents " only. New incentives would continue to be fixed by the
company subject to the grievance procedure. As a result, many questions
on which parties could not agree, or about which they had been deliberately vague in their agreements, had to be referred to arbitration.
Fluctuations in Earnings
Related to the difficulty of defining " normal performance " as a
basis for the fixing of equitable wage differentials is the problem arising
from the tendency of incentive earnings to fluctuate. In so far as this
reflects variations in individual performance it is a normal phenomenon,
fully in keeping with the philosophy of incentive wages, namely that an
increase or a fall in effort should be compensated by a change in the
bonus earned.
But there are other reasons, quite unconnected with individual
performance, why earnings can fluctuate even though there has been no
change in actual rates of remuneration. Such fluctuations can affect
different jobs in widely varying degree and result in disparities not reflected
in the rates fixed for the respective jobs. Thus, even though the workers
may have accepted the rates as fair and equitable in terms of effort or
job contents (or both), they will tend to resent what they regard as
unjustified differences in earnings.
For example, in spite of all the precautions normally taken when
fixing rates by work study methods, actual earnings on a job may rise
considerably above those attained during a period of testing and experimentation. As the worker gains experience he may improve his own
working methods in ways that the industrial engineer had not foreseen
when carrying out his time and motion studies. Or he may benefit from
1

The question of work rules was a major issue in the 1959-60 steel strike.

SOME PROBLEMS OF ANALYTICAL JOB EVALUATION

a more regular supply of parts or materials. For example, if the firm
receives more orders, it may attempt to improve the over-all organisation of its operations, enabling piece workers to increase their output
and earnings without greater effort.
Furthermore, especially when incentive systems are initially introduced, piece rates may be deliberately set at a rather high level in order
to overcome suspicions and resistance on the part of the workers. The
need for such " loose " rates may, however, be greater in some departments than in others. As a result, differences in earning opportunities
will appear and may persist for many years after the introduction of the
system.1
In all these cases relative wages tend to be distorted and thus to
defeat the purpose of job evaluation. A wage structure that is acceptable
at the moment of its introduction may show sharp increases in earnings
on some jobs, provoking claims for adjustment of other incentive and
time wages. And once job evaluation has been accepted as the basis
for determining " equitable " wage relations, it is hard to turn down
such claims.
One way of reducing difficulties of this kind is to apply incentive
systems under which earnings rise less than proportionally with output.
For example, it appears that in some West German steel plants premium
curves in new rolling mills have, in recent years, been made less steep
than those for workers using older types of equipment, so that excessive
differences between earnings in different mills are avoided. This, however, reduces the incentive effect of the wage system—a consequence
which will normally not be considered desirable. One might therefore
approach the problem from the other end and, instead, provide compensation for those who fall behind as a result of rising incentive wages.
Various devices have been applied for this purpose. Sometimes
time workers are given a flat wage increase to compensate for the absence
of opportunities of earning incentive premiums. In other cases they are
paid a bonus, depending on the earnings of the piece workers, in still
other cases special incentive schemes have been devised for jobs not
amenable to application of straight piece rates; for example, a group
bonus may be introduced for maintenance workers, depending on the
ratio of maintenance hours to the plant's total output over a certain
period, or a degressive piece rate may be applied, the rate of the bonus
falling more or less steeply when output rises above the level beyond which
the quality of the work may suffer.2 In some cases such procedures
1
J. P. DAVISON and others: Productivity
George Allen and Unwin, Ltd., 1958), p. 54.
2

W O L F , op. cit., pp. 62

ff.

and Economic Incentives

(London,

JOB EVALUATION

have proved satisfactory, in others they have not. Flat or varying
bonuses for time workers may be resented by the piece workers—who
have to make a specific measurable effort in order to earn their bonus,
which time workers do not—and particularly by those piece workers
who, for one reason or another, earn a lower bonus themselves. Thus,
application of incentive schemes to jobs that do not really lend themselves
to such systems may not be effective in achieving the desired level of
relative earnings or may cause a fall in the quality of the work performed.
Special problems arise when premiums earned under incentive
systems do not really reflect personal effort. For example, when group
incentive schemes are applied, some workers of the group may have no
influence on the rate of output except in the negative sense that they
could hold up the work done by those workers who do have an influence
on output. It may then be necessary to supplement the evaluation of
job contents as a basis for relative wages with some evaluation of the
extent to which various workers may be deemed to contribute to the
actual speed of production. In other cases old and involved systems
of premiums, allowances and bonuses are applied, on top of very low
basic rates, in such a way that some bonus is earned under almost any
circumstances without any discernible relation to the human effort
which these bonuses are normally intended to reward. In such cases
introduction of job evaluation may raise the question of the exact components of the wages to which the results of job evaluation should be
applied. When this entails drastic revision of long-established systems
of payment, it may not be an easy task.

CHAPTER IV
SOME EXPERIENCES WITH JOB EVALUATION SCHEMES
In the previous chapter a number of problems arising in the application of job evaluation, and especially of point rating systems, were
described and briefly discussed. Many readers will, however, be interested
in the experience gained in concrete cases. The following pages therefore
contain a summary of a few inquiries into the functioning of actual
schemes. While most of the examples given are taken from the United
States, where job evaluation plays a particularly important part, the
thoroughgoing studies of job evaluation in the West European iron and
steel industry which have been carried out under the auspices of the High
Authority of the European Coal and Steel Community are also of considerable interest; some of the findings relating to the German steel
industry are given below. Readers are also referred to the description
given in Appendix C of some systems of job evaluation applied in various
countries.
SURVEY OF EXPERIENCE IN 73 UNITED STATES COMPANIES

One of the most comprehensive surveys of job evaluation schemes
was undertaken by the Industrial Relations Section of the Department
of Economics and Social Institutions of Princeton University.1 This
covered 73 companies, many in the metal trades, employing from 278 to
140,000 workers. Thefindingsare based on replies to questionnaires sent
out to the companies as well as on interviews with company executives,
and on information obtained through interviews with one or more
officials of 20 national or local unions.
Of the 73 companies surveyed, 41 reported general satisfaction with
their job evaluation schemes, 11 reported definite dissatisfaction and 12
felt that they had had both satisfactory and unsatisfactory experience.
The nine other companies could not express an opinion, as their schemes
had not been operating long enough. Some companies reported on more
1
See Helen BAKER and J. M. TRUE: The Operation of Job Evaluation Schemes
(Princeton, N.J., Princeton University, 1957).

JOB EVALUATION

than one scheme. Of 20 schemes which had proved unsatisfactory
(including discontinued schemes), 14 had been established in 1939 or
earlier. It appeared, therefore, that the longer a scheme had been in
operation the greater the chances were that difficulties would arise which
would require the scheme to be discontinued or completely revised. On
the other hand, there was evidence that a large number of plans which
had been established in 1939 or earlier were still proving satisfactory.
Of 56 point rating and factor comparison schemes, 34 per cent, were
felt to be unsatisfactory, while of 20 ranking and classification schemes
only 20 per cent, were considered unsatisfactory. The failure of a number
of point rating schemes was said to be due to an effort by managements
to adhere rigidly to a system which did not take into account such
important .elements in wage determination as the effects of collective
bargaining and conditions in the employment market. In the case of
the ranking and classification systems a major weakness seemed to be the
lack of adequate job descriptions as contrasted with the more elaborate
point rating and factor comparison schemes.
The following factors led to the failure of various schemes:
(1) weaknesses in installation;
(2) inadequate administrative organisation (generally a failure to assign
staff to maintain the scheme properly and resolve difficulties as they
arose);
(3) changes in key executive personnel resulting in the withdrawal of
top management's support for the scheme;
(4) trade union opposition to the scheme after it had been introduced;
and
(5) inflexibility in the administrative arrangements which led to inability
to deal with changed conditions such as new production methods.
On the other hand, the survey revealed that the job evaluation
schemes had several advantages.
(1) The schemes had, in many cases, brought about a more logical and
equitable wage structure, which had considerably improved the
workers' morale. This comment was made most often by administrators of recently established schemes, whose experience in dealing
with wage problems without the help of job evaluation was fresh
in their minds.
(2) It was also found that the schemes had facilitated the processes of
collective bargaining, since the wage rates which had been determined
by collective bargaining could be defended as having been arrived at

SOME EXPERIENCES WITH JOB EVALUATION SCHEMES

quite impersonally. Certain emotional elements were removed from
the bargaining and both management and trade union representatives
were able to approach wage problems more dispassionately. Industrial relations had improved as a consequence.
(3) The number of wage grievances was reduced in many cases and
workers' confidence in the fairness of management in rate setting
was increased.
The experience of the 73 companies surveyed indicated that the
following conditions are necessary for the successful operation of a job
evaluation scheme:
(1) It must be carefully established by ensuring (a) that management's
aims are clear to all concerned and that not only the manual workers
but also all levels of supervision and management fully understand
its implications, and (b) that all relevant internal and external factors
have been taken into account in arriving at the final form of the
scheme.
(2) It must have the full approval and continued backing of top management.
(3) The trade union's acceptance of the scheme must be obtained.
(4) Adequate administrative controls must be set up to ensure (a) centralised co-ordination of the scheme, (b) the evaluation of new and
changed jobs, (c) the proper control of individual rate changes and
(d) the conduct of wage surveys to provide the necessary information
about intra-plant wage rates.
(5) The importance of factors other than job content in wage rate
determination, for example employment market conditions, sex wage
differentials, geographic wage differentials and the relative bargaining
power of management and the trade union concerned, must be
recognised.
The companies' experience further revealed that a fundamental difficulty with job evaluation schemes arises from the conflict between the.
need for simplicity and flexibility on the one hand and consistency and
accuracy on the other. Many executives stressed that it is necessary to
take into account the human and economic realities of a given situation
but at the same time that there must be " systematic, impersonal and
factual appraisal of a job if a scheme is not to gradually deteriorate into
a hit-or-ffiiss method ". Another fundamental difficulty arose from
differences of opinion between managements and trade unions regarding
the worth of job evaluation. Trade union officials tended to distrust a

JOB EVALUATION

method which " attempts in advance to limit the factors which may be
considered in rate setting ". Although management representatives were
not all convinced of the merits of job evaluation, they generally seemed
to prefer some form of it as a rate-setting tool.
The study concluded that—
. . . much study is needed in determining the best type of plan for a given situation, in solving administrative problems and in reconciling fundamentally
different attitudes . . . While differences of point of view in rate setting are
inevitable in any bargaining situation, the differences can be magnified or
lessened by the way job evaluation is applied. Both management and union
representatives have suggested the advisability of looking upon evaluation as a
practical guide and tool rather than as a rigid system. Approaching job evaluation in this way may bring about wider understanding of its worth and its
requirements among executives, supervisors, employees, and shop stewards.
A tool, it is generally recognised, should be designed with regard for the men
and women using it. In some situations a simpler tool is of more value than a
complex one, and the more complex the tool, the greater the training required
for its use. Changing conditions may call for radical changes in design. In any
case, a good tool needs continuing care to keep it useful and effective.1
SURVEYS OF EXPERIENCE IN 96 CANADIAN COMPANIES AND 75 UNITED
STATES COMPANIES

Two other surveys, one by the Industrial Relations Centre of McGill
University of the experience of 96 companies in the Montreal area, and
the other by the Dartnell Corporation of Chicago, concerning the experience of 75 companies, have also reported many cases of improved
industrial relations as reflected in smoother collective bargaining and a
reduction in wage grievances.2
SURVEY OF EXPERIENCE OF 23 UNITED STATES AIRCRAFT
MANUFACTURING COMPANIES

A survey covering 23 aircraft manufacturing companies throughout
the United States yielded a large number of favourable reports. Most of
the companies stated that they were highly satisfied with their schemes,
while slightly less than one-quarter said that their schemes were satisfactory and only one that its scheme was fairly satisfactory. None reported
that its scheme had proved unsatisfactory.
Among the advantages mentioned most frequently were that the
schemes—
1

BAKER and TRUE, op..cit., pp. 91-92.

NICOLOPOULOS, op. cit., pp. 27-28; and "Job Evaluation. A Survey of Company

Practices ", in The Management Review (New York, American Management Association), Vol. XLIII, No. 11, Nov. 1954, p. 733.

SOME EXPERIENCES WITH JOB EVALUATION SCHEMES

(1) provided a factual basis for determining the worth of jobs;
(2) promoted wage equity and the standardisation of wages;
(3) led to greater consistency and uniformity;
(4) established better policies of promotion, transfer and placement; and
(5) improved morale. 1
SURVEY OF EXPERIENCE IN THE UNITED STATES PACIFIC COAST
AIRCRAFT MANUFACTURING INDUSTRY

Another study carried out in the United States surveyed the experience
that the Pacific Coast aircraft manufacturing industry, both in the Los
Angeles and Seattle areas, had had with the Southern California Aircraft
Industry plan during and since the Second World War. It was found
that—
. . . the greatest difficulty arose from the frequent conflicts between measures of
external and internal equity
The conflict between these different standards
made for continuous obstacles to the consistent administration of the plan.
Consequently, certain manipulations had to be resorted to in order to
achieve greater consistency. In particular, some jobs had to be redescribed and re-evaluated because it became increasingly difficult to find
workers to fill the more disagreeable jobs. These were rated low in the job
structure since the scheme rewarded skill much more than job conditions
(63.2 per cent, of the total points were allotted to skill as against 11.4 to
job conditions). Thus, in the case of the job of a sandblaster, which had
already been allotted the maximum possible points for job conditions, it
was agreed by the parties concerned to increase the points allotted to the
job for all the factors so as to enable a more attractive wage rate to be
paid. Another method which was used in dealing with this situation was
that of upgrading certain workers. It was left to the plant administrations to decide how many workers should be in sub-grades A, B and C
of the various job grades. The administrations thus simply upgraded
workers within a job grade in cases where it was found necessary to pay
a higher rate without either a general wage increase or a re-evaluation of
the job.
The study also found that another important problem arose from the
fact that managements were divided among themselves about various
aspects of the scheme. This led to conflict between the different ranks
of management. In the words of the study—
1
See Elizabeth LANHAM: " Policies and Practice in Job Evaluation. A Survey ",
in Personnel, op. cit., Vol. 29, No. 6, May 1953, p. 497.

JOB EVALUATION

. . . the technical personnel of management closely associated with the formulation of the plan—engineers and wage analysts—were likely to look upon the
plan as a finely tooled machine with each of its parts interdependent and
therefore not to be tampered with. On the other hand, those of management
charged with industrial relations had to deal in more flexible concepts. They
needed to meet the give and take of collective bargaining. To function at all,
some room for maneuver was essential.
Another kind of problem in the administration of a plan is raised by the
special position of the foreman as an agent of management. He is likely to
prove unreliable for evaluation purposes, since he is only marginally a representative of management. Although he exercises supervisory functions in the
name of the company, his relationships to the work force are likely to be close
and his sympathies and even his loyalties divided. His vantage point is not high
enough to see the wage structure as a whole, and his sense of status is served by
an overevaluation of the importance of the work he supervises. For the S.C.A.I.
plan the problem was complicated by the fact that in some plants the foreman's
rate of pay was tied to the pay scale of those under him
Largely back stage, but no less acute for that reason, a conflict within the
individual management groups existed in almost every plant under the plan.
The industrial engineers wanted to preserve the integrity of their scientific
method; the industrial relations men, to adjust to the pressures and to engage
in the tactics of collective bargaining; the production supervisors, to ease their
administrative tasks; the personnel representatives, to recruit readily, reduce
turnover and grievances, and raise morale; the accountants, to contain costs
within set limits. Reconciliation of diverse points of view was not limited to
the union-management arena.
Because it was an industry-wide scheme difficulties arose from the
fact that the various plants administered the scheme differently and from
the attempt to draw up composite job descriptions covering all plants.
For example, it is stated in the study that—
. . . Lockheed and Northrop, for quite separate reasons, evidenced more
laxity or generosity (depending on the point of view) than the other companies.
. . . In an effort to include under one title description jobs in different companies,
each with its own make-up of tasks, the job descriptions became very broad
and their limits hazy. The effort of reaching agreement was exceeded only by
the effort of administering the " composite jobs ". The mass filing of " reclassification " grievances was, if not invited, at least made possible as thousands
of workers claimed to fall within the ill-defined boundaries of higher-paid jobs.
Among several other problems encountered, one is of considerable
importance, namely the fact that the collective agreement provided for
plant-wide seniority, so that in the event of a layoff the individual dropped
could lay claim to any job held by a person of lesser seniority provided
the job requirements could be met. Since employment and the type
of work fluctuated considerably, chaos resulted. The breakdown of
" skill " into small elements under the job evaluation scheme greatly
complicated and magnified the normal difficulties in administering plantwide seniority. To quote the study—

SOME EXPERIENCES WITH JOB EVALUATION SCHEMES

. . . factory workers about to be laid off " bumped " workers of less seniority
holding other positions . . . . Much resentment among the bumped employees
resulted . . . . Some employees were bumped on an average of five times a
month. The paper work for the company was immense. Ten men had to be
moved to lay off one. The company often had to rehire ten men to get the
one it wanted. Senior employees could demand a whole series of trials before
they found a job on which they were acceptable permanently. Inefficiency was
the order of the day.1
The principal findings of the authors of the study are summarised
below.
(1) The higher the wage level of the plant, both relative to the community
and in absolute terms, the more successful a scheme is likely to be.
This is because job evaluation as a method of wage determination is
more inflexible than others and, if the wage level in the plant is
relatively low, it is more difficult for the scheme to reflect changes in
employment market conditions and in the cost of living. A scheme
will also work more easily in a stable employment market and where
labour is not in short supply.
(2) A job evaluation scheme should be devised and administered with
due regard to the conditions of the employment market, which cannot
be ignored if the scheme is to be successful. It should, therefore,
reflect those forces which are important in the market, e.g. relative
supply of and demand for labour, bargaining power of the parties
and job conditions.
(3) The better the state of industrial relations the easier it is to introduce
a job evaluation scheme.
(4) The details of a scheme should be drawn up in such a way that they
do not conflict with other provisions of a collective agreement such
as, for example, seniority clauses and grievance procedures.
(5) A job evaluation scheme is likely to be more successful if it is introduced on a plant-to-plant basis than if it is applied to a whole industry.
This is because it is difficult to standardise jobs throughout an industry
unless the plants in it are so similar that they can be treated as being
virtually a single firm.
(6) If the workers in a plant are unionised, it is highly desirable that any
scheme adopted should be agreed to and, if possible, developed
jointly by the company and the trade unions.
(7) A scheme is better administered by the industrial relations staff of a
company than by the industrial engineers who may have developed it.
The essence of successful administration of a scheme is flexibility, and
1

KERR and

FISHER, op. cit.,

87.

JOB EVALUATION

this is better understood by those engaged in industrial relations work
than by industrial engineers.
(8) It is of major importance that the number of job titles and classifications be kept to a minimum. If they are not, a scheme becomes too
inflexible because of the narrow coverage of the job descriptions.
Promotions within a grade become difficult to make and errors in
description become more serious. Moreover, workers tend to feel
more insecure and to cling to their present jobs because they may not
have the qualifications for another job.
(9) Any anticipated changes in method should be carried out before a
scheme is installed and all modifications in it should be resisted until
it becomes firmly established.
(10) In preparing job descriptions it is sound practice to emphasise in
them the things which make one job diflerent from another rather
than to give a comprehensive statement of all the duties of the jobs.
(11) A scheme which provides for single rates and for definite ratios
between the rates for classes of workers (A, B, C, etc.) within a job
grade is easier to administer than one which establishes rate ranges
and has no such fixed ratios.
(12) Over a period of time it will prove difficult to maintain the heavy
weighting which is given to skill in the Southern California Aircraft
Industry plan and other job evaluation schemes. There are influences
operating which will require that greater weight be given to other
factors.1
(13) Since workers will gradually become used to a scheme and will
resist any attempts to change it—so powerful a force is custom—it
is a matter of considerable importance for the maintenance of good
industrial relations that a scheme be introduced gradually and with
great care.2
STUDY OF THE DISTRIBUTION OF EMPLOYEES IN SIX
SOUTHERN CALIFORNIA AIRFRAME MANUFACTURING PLANTS

AS mentioned above, conflicts between measures of external and
internal equity have raised difficult problems in the administration of the
1
For example, it has been contended by various writers that with the development
of automation and the increasing tendency to transfer skill requirements from the
worker to an expensive machine, responsibility will have to be given greater weight
than heretofore.

« KERR and FISHER, op. cit., pp. 91-94.

SOME EXPERIENCES WITH JOB EVALUATION SCHEMES

!»

o
o o t N T t r i K -^ o ~ - r J r i c ô « n r i — I N ©

\o — u-i NO <•* r- — o o ( N T j - t s r - - v " i m » n r o
-C3\r*wMO'or*',oo^'^,oo»i^i-i i

as
ro
o"

!
!»

q vo

oo q o \ » q v )

>n ^ —

ON V Ì

o \ --_ i

O N O r x n O N N O ^ N — VOOO^-ONVDOONOO

coONNO'tosNDNONO'Ono-'Non^«
- - • * 00 VI « W W ON O O O O ' t " ^ ^ - ^ W
—T
—T psf wT
— r i »rT —«* - Í ^ CN

o
o

i
1

es"

!»

I
u

í
1»
!
!»

I
<

1
!»

I
f
1

t o o c t N O N r - N c o v o m T j - O N - i
O C N "o — N ^ ( s m r ^ r ^ o w ^ o \ * n > o
u-i
<nnr-.vûMTtfnvDNû^riTiTjri"
-^t

i t

WÌVÌ'-'tXUOqNn^inCO^'nP*; O
rJ^rJ^r^Tfr©TfrN¿</^00rn^:f'"i©~

I 1
' '

ci"

o
o
o

OOON»û^-nOONOOX-ONO\-N i i
mi/^Of*ir-r--t^ —N n ( N o o N \ f r - o
(M -«t N • * n Tt O^Tt vo >n co m Tt n
— '

00
Os

o
r-j

—«

r - v o r-ON vo r - m ON Q "<* NÚ \ o (N \ o — íN i i
(S \ o r - r - - o o vo oo ON o - ^ O ON oo »o m r4
H
NO c i r«->
r ^ w r t f n n T i TfoOTt^r*i

NNri«rs-^O¡ON'nONON00>n^;(N¡q
Trrnr^rjt^mwSTi-NO^tirirJoNTj-'örn

Tt ON - ^ 00 ON ON r- t—, ON •<*• « O r n NO O> rn — ONNOt^(S^ co-MWNisOT'ir »rtTfíSíNON'^tr-TOOONOr-vq^tNiri
m

i
'

o
Ö
o

i
'

—

H
— NM^vNNor-oooNO^nfi^-viVor-oo

JOB EVALUATION

Southern California Aircraft Industry plan. These difficulties are reflected
in the differing distribution of workers among the various grades in
different plants, as revealed by a survey covering six airframe manufacturing plants in southern California.1 For example, table 8 shows that
the respective proportion of workers in grade 12 was 12.8, 19.3, 3.8,
35.0, 4.7 and 5.2 per cent, of the total labour force in each of the six
plants. It would have been reasonable to expect to find somewhat
similar distributions in all the plants (which used the same job evaluation
scheme) since, as the study points out—
. . . except in periods of great economic fluctuation or an extensive readjustment
which changes the normal composition of skills in aircraft plants, it is reasonable to assume that there exists a normal distribution of skills in any aircraft
p l a n t . . . . Firms producing airplanes have a like number of skilled jobs, a like
number of unskilled jobs. It should, then, be fair to expect that the various
airplane producers would have a similar percentage distribution of skills
expressed as similar percentages of jobs and employees in the various labour
grades.

STUDY OF JOB EVALUATION IN THE WEST GERMAN STEEL INDUSTRY

Interesting experiences, though conducted along lines somewhat different from the surveys mentioned above, were also reported in the study
undertaken by the High Authority of the European Coal and Steel
Community on the application of job evaluation in the steel industries of
member countries—Belgium, France, the Federal Republic of Germany,
Italy, Luxembourg and the Netherlands. The study concerning Western
Germany, mentioned earlier, includes a thorough analysis of the introduction and results of job evaluation in six plants, and especially of its
actual effects on the wage structure. The study was carried out in five
production departments of each undertaking: Thomas steelmaking,
Siemens-Martin steelmaking, electrical steelmaking, blooming and rolling.
For a number of jobs in these departments the following comparisons
were made: basic wage rates as they existed before the introduction of
job evaluation compared with rates fixed on the basis of job evaluation;
gross hourly earnings before and after the introduction of job evaluation;
and rates and gross hourly earnings after job evaluation. These comparisons were made in the form of simple correlations and summarised
•See "Distribution of Employees by Labor Grade in Some Airframe Plants",
in Research Bulletin of the International Association of Machinists (Washington),
Vol. VII, No. 12, Dec. 1953, pp. 4-5. This problem was also referred to in Chapter III;
charts III and IV illustrate these differences in the case of two plants.

SOME EXPERIENCES WITH JOB EVALUATION SCHEMES

TABLE 9. EFFECTS OF THE INTRODUCTION OF JOB EVALUATION
ON THE STRUCTURE OF WAGES AND EARNINGS IN SIX
WEST GERMAN STEEL PLANTS
Plant

Number of jobs examined 1

13 29
(13) (29)

53
(59)

67
(76)

49
(56)

15
(13)

0.97 0.78

0.74

0.81

0.85

0.40

Correlation between gross
hourly earnings before
and after job evaluation
(r2)

0.52 0.90

0.86

0.81

0.86

0.76

Correlation between rates
and gross hourly earnings
after job evaluation (r3) .

0.95 0.36

0.85

0.73

0.87

0.76

Correlation between original
basic wage rates and rates
fixed after job evaluation
(ri)

Source: Haute Autorité, op. cit., pp. 200 and 243.
1
The numbers in brackets are those used for calculating the figures in the bottom row of the
table (r s ).

in correlation coefficients (r). 1 The results of these inquiries are summarised in table 9.
As can be seen from the figures given for r1; job evaluation resulted
in a more or less drastic change in the structure of wage rates, especially
in plant F though very little in plant A. In the latter case, however, only
a small number of jobs were examined, all belonging to the same department; thus, in this case the inquiry related to a relatively narrow job
cluster within which few changes in relative rates may have been necessary or, perhaps more probably, feasible. It may also be noted that,
except in plants A and D, the change in the pattern of relative earnings
was smaller than that in the rate structure: r 2 is larger than r,. 2 The
difference is particularly striking in the case of plants B and F ; in the
former plant the pattern of relative earnings after job evaluation differed
only slightly from that obtaining before, although the pattern of wage
rates had been affected rather strongly.
1

The correlation coefficient measures the similarity in the structure of two series.
Its maximum value ( r = l ) occurs when one series can be completely derived from
the other one. This would happen, for example, if all individual wage rates after
job evaluation had been equal to the old ones plus a uniform amount of money, or if
gross earnings were a uniform percentage of basic rates. The minimum value of the
coefficient (r=0) occurs when there is no quantitative relationship at all between
the two series.
2
The reason given in the report for the low figure for plant A is that two jobs had
to be deleted from this comparison.

JOB EVALUATION

Hence, in the majority of cases changes in the structure of wage rates
did not lead to an equally large change in the structure of earnings.
This is ascribed to a number of alterations made, during and after the
introduction of job evaluation, in systems of payment by results, including both individual incentive payments and premiums based on production in the plant as a whole. These changes, according to the study,
" enabled the plants to cancel out, practically speaking, the direct effects
of the new wage rates based on job evaluation on relative earnings, in
the sense that the original relationships between gross hourly earnings
were maintained or hardly modified n.1 The figures do, indeed, show a
tendency to that effect.
These findings illustrate the important part which may be played by
wage elements other than basic rates. In the steel plants examined these
elements consisted largely of production premiums of one sort or another.
Indeed, a large proportion of earnings seems to be independent of basic
rates in the German steel industry generally. In a few other plants
covered by the study but not applying job evaluation, basic rates fixed
under collective agreements represented no more than about two-thirds
of total earnings on the average.
Whereas, however, in these plants rates and earnings in the various
jobs stood in a more or less constant relationship to each other 2, in the
plants applying job evaluation differences between rates and earnings
were found to be not only large but unevenly distributed. For the
majority of the jobs examined the size of the difference varied between
one-quarter and almost one-half of the rates fixed on the basis of job
evaluation. But while in most cases all members of a team (for example
the workers at the blast furnaces, the steel converters or rolling mills)
were paid according to the same premium systems, there were substantial
differences between the systems of wage payment applied to different
teams and in some cases also between jobs in the same department not
performed by the same team. These differences in systems of wage
payment are reflected in the bottom row of the table : the correlation
between gross hourly earnings and rates is not very high.
The exception again is plant A because the jobs examined all belonged
to the same department and had a common system of wage payment.
In plant B, on the other hand, the structure of relative earnings bore
hardly any relationship to the rate structure based on job evaluation.
In the other plants there is a clear relationship between the two but,
especially in D and F, it is not particularly strong. Another aspect of this
1

Haute Autorité, op. cit., p. 207.
' Ibid., p. 32.

SOME EXPERIENCES WITH JOB EVALUATION SCHEMES

matter is that in several cases very different earnings occurred in the
same ranges of point scores, and even among jobs scoring equal numbers
of points. On the other hand, there were cases in which the same earnings
went to jobs with widely differing point scores.1
In the study one possible explanation of this discrepancy is ruled
out, namely that of large differences in productivity as between various
groups of workers. Differences in productivity commensurate with the
reported differences in earnings are not possible.2 The wide variation in
wage components over and above the basic rates was regarded as mainly
the result of wage policy decisions arrived at on the basis of traditional
relationships and compromises.3 In the view of the employers—
. . . both the wages based on job evaluation and those fixed under collective
agreements are largely made ineffective by the traditional structure of hourly
earnings which has proved to be an almost unalterable reality in the steel
industry. The adjustment between gross hourly earnings and the wage structure
based on job evaluation has, so far, been no more achieved than that between
gross hourly earnings and the rates fixed by collective agreements. The problem
of a rational ranking of earnings, which is essential to the wage policy of
undertakings and trade unions, cannot be solved through job evaluation so
long4 as the latter has no influence on the part of wages depending on productivity.
The trade union representatives who commented on the study stated
that the group of experts who had carried out the investigation could not
agree on the question whether it was the wage structure based on job
evaluation or the rather different structure of relative earnings which
corresponded best to the needs of the undertakings.5
In their view the changes in the calculation and amounts of wage
components depending on productivity made during and after the introduction of job evaluation could be explained largely by the needs of the
undertakings and the efforts of the workers, which caused the old
relationships between gross hourly earnings to be restored or maintained,
1
These points are clearly demonstrated in some of the charts on pp. 254 ff. of the
report.
2
Haute Autorité, op. cit., p. 237. The study adds that in the steel industry it is
much more difficult than in other industries to define and measure a standard output
corresponding to a standard wage as the basis for systems of payment by results.
In most of the departments studied " productivity " was defined, for purposes of wage
payment, as output in tons per man-hour but, according to the report, this quantity
is not only, and in the majority of cases not even primarily, dependent on human
performance.
3
Ibid., pp. 236, 237 and 241.
•Ibid., pp. 260-261.
6
Ibid., p. 264. The study itself gives a description of the views of the majority
of the experts from which it would appear that the latter saw little possibility of reflecting wage relationships based on job evaluation more consistently in the pattern of
earnings (p. 245).

JOB EVALUATION

even though these diverged from the wage relationships based on job
evaluation. They seemed to think that the wage structure emerging from
job evaluation as applied in the majority of the German steel plants had
proved to be unrealistic.
They considered that the system applied in most of these firms was
wrong: it is a weighted point system and therefore, according to the trade
unions, incapable of adjustment to differences and changes in conditions
of work and production. Moreover, in their opinion the system takes no
account of numerous elements which, therefore, have to be reflected in
wage elements " so-called dependent on productivity but having nothing
to do with variations in output and human effort ".1

1
It may be recalled that the system developed by the trade union experts Hagner
and Weng is an unweighted system and more in the nature of a factor comparison
method.

CHAPTER V
ADVANTAGES AND LIMITATIONS OF JOB EVALUATION
In the long run, the extent to which job evaluation will be applied
depends on the balance of advantages and drawbacks of the method
as an aid in fixing relative wages. These differ as between different
undertakings, industries and countries. It would seem, however, on the
basis of the experience described in the preceding chapter and the
analysis contained elsewhere in this study, that the principal general
merits and limitations of the method may be summarised as follows.
ADVANTAGES

(1) Job evaluation is a logical and, to some extent, objective method
of ranking jobs relatively to each other. It may thus help in removing
inequities in existing wage structures and in maintaining sound and
consistent wage differences in a plant or industry. It may also result
in a simpler wage structure than the existing one, more easily understood and accepted by the workers.
(2) In the case of new jobs, the method often facilitates fitting them
into the existing wage structure.
(3) The method helps in removing grievances over relative wages
and thus may improve labour-management relations and workers'
morale, in addition to reducing the time that management and trade
union officers have to devote to such grievances, enabling them to deal
with other important matters. In providing a yardstick by which
workers' complaints or claims can be judged, the method simplifies
discussion of wage demands and enables differences in wages to be
explained and justified.
(4) The method replaces the many accidental factors occurring in
less systematic procedures of wage bargaining by more impersonal and
objective standards, thus establishing a clearer basis for negotiation;
this too may help in improving labour-management relations.
(5) The method may lead to greater uniformity in wage rates, thus
simplifying wage administration. If widely applied in an industry, it

100

JOB EVALUATION

also tends to facilitate wage comparisons between geographically separated
employment markets, which again may be helpful for certain bargaining
purposes.
(6) The information collected in the processes of job description
and analysis may also be used for the improvement of selection, training,
transfer and promotion procedures on the basis of comparative job
requirements.
(7) Such information also often reveals that workers are engaged
on jobs requiring less skill and other qualities than they possess, thereby
pointing to the possibility of making more efficient use of the plant's
labour force. More generally, inefficiencies in the organisation and
methods of work and possibilities of improving working conditions and
reducing job hazards may be discovered.
LIMITATIONS

When judging the advantages listed above it should, however, be
borne in mind that job evaluation is not a panacea. The scope of its
useful application may be limited by a smaller or larger number of factors
of which the following seem to be the most important.
(1) Although there are many ways of applying job evaluation in a
flexible manner, rapid changes in technology and in the supply of and
demand for particular skills raise problems of adjustment that may need
further study. Perhaps difficulties of this nature are more serious with
point rating systems than, for example, with certain versions of the factor
comparison method which do not involve predetermined degrees of the
various factors.
(2) However logical and consistent a system may be, account has
to be taken of various limits to the possibility of changing apparently
illogical elements in an existing wage structure. Certain external rates
may have to be paid for " hiring jobs " whether they are compatible
with the job evaluation system or not; traditional wage relationships,
however inconsistent they may be, cannot always be altered. It has been
pointed out that—
. . . the fundamental purpose of job evaluation is to establish a mutually
acceptable criterion of equity. If both worker and supervisor agree that, for
example, the cementers are the aristocrats of the raincoat industry, what useful
purpose is served by upsetting this scale of values in favour of some mechanistic
criterion of equity ? These traditions are every bit as important as job content.1
1
W. GOMBERG: " A Trade Unionist Looks at Job Evaluation", in Journal of
Applied Psychology (Lancaster, Pa., American Psychological Association), Vol. 35,
No. 1, Feb. 1951, p. 3.

ADVANTAGES AND LIMITATIONS OF JOB EVALUATION

101

(3) When job evaluation results in substantial changes in the existing
wage structure, the possibility of implementing these changes in a
relatively short period may be restricted by the financial limits within
which the firm has to operate.
(4) When there is a large proportion of incentive workers, and
especially when incentives have to be changed frequently, it may be very
difficult to maintain a reasonable and acceptable structure of relative
earnings.
(5) In any case, job evaluation does not provide a complete answer
to the wage problem. It has nothing to say about the absolute wage
level and little about the absolute size of the wage differentials appropriate to the evaluated job structure. It is not concerned with questions
of rewarding loyalty, seniority or merit, and its consistent application
may be very difficult in cases where payment by results is applied.
(6) In particular, the process of job rating is to some extent arbitrary
because few of the factors and degrees can be measured with great
accuracy. This process can certainly not be regarded as " scientific "
and it can be considered " objective " at best in the sense that, under
ideal conditions, it is free from deliberate bias. All that can be expected
to emerge from the process is a well-considered and honest, but basically
subjective, evaluation of jobs. 1
(7) Job evaluation takes a long time to install, requires specialised
technical personnel and may be quite costly. One Canadian estimate
put the cost of installing a scheme at $45 per job evaluated and maintenance charges at about 25 cents per employee per month. 2
(8) The drawing up of job descriptions tends to formalise job
contents, which may lead to abuses by workers who acquire a vested
interest in maintaining the requirements for certain jobs. According
to one writer there are " classic examples of this in railroading (in the
United States) where featherbedding is entrenched behind job definitions that were made to fit conditions of 50 years ago ". 3 Very precise
and detailed job descriptions may also make it impossible to assign
occasional jobs not explicitly provided for in the descriptions.4
(9) Trade unions often regard the method with suspicion and in
some cases with hostility. In such cases, job evaluation may still be
1

It may be recalled that this is not a criticism of the method, but merely a limitation of its significance; wage negotiations not based on job evaluation may be much
more defective in this respect.
2
riant Administration (Toronto), Sep. 1952, p. 56.
' F. J. KNIGHT: " Fallacies in Job Evaluation ", in Advanced Management (New
York, Society for Advancement of Management), Vol. XV, No. 6, June 1950, p. 22.
4

See KERR and FISHER, op. cit., pp. 86 and 88-89.

102

JOB EVALUATION

useful in helping management to clarify its own ideas as to what is to be
regarded as a suitable wage structure, but the possibilities of using the
results of the method in actual wage negotiations and in the settlement
of grievances will be limited.
ATTITUDES TOWARDS AND PROSPECTS FOR JOB EVALUATION

It would appear from various comments and experiences noted in
earlier parts of this study that at the present stage, although job evaluation has in many cases been applied with satisfactory results, the method
is not yet universally regarded as a useful aid in designing rational wage
structures. As was mentioned in Chapter III, some people believe that
the method is not in harmony with certain economic principles that in
their view should govern the fixing of relative wages. In the discussion
of this opinion it was suggested that this may well be an unduly pessimistic
view. Furthermore, to the extent that job evaluation may not be quite
in accordance with certain ideal economic principles of wage determination, its shortcomings may not be more serious than those of other
procedures of wage bargaining. Nevertheless, there are certainly cases
in which the implementation of job evaluation plans is made difficult
by the existence of external rates and by changes in the supply of, and
demand for, different types of skill that cannot be reconciled with any
pattern of weights and factors of the job evaluation system.
It has been suggested in this study that the main advantages of job
evaluation for purposes of wage fixing (as distinct from any other uses
to which job analysis can be put) would be that of making the discussion
of problems of relative wages more systematic, of helping in the establishment of a more or less logical and understandable wage structure and of
simplifying the settlement of grievances concerning wage differences.
If this view is correct, the main question concerning the prospects of
further development and more widespread application of the method
would seem to be whether employers and trade unions find it sufficiently
helpful in these respects to justify the rather elaborate work involved
in the preparation and maintenance of job evaluation systems.
Employers
Broadly speaking, it would seem that employers and their organisations, in so far as they have had experience with it, regard job evaluation
either favourably or more or less indifferently. For example, in the
United States the National Metal Trades Association and the National
Electrical Manufacturers' Association have developed systems of their

ADVANTAGES AND LIMITATIONS OF JOB EVALUATION

103

own which have been introduced into thousands of plants. In the Federal
Republic of Germany the Iron and Steel Industries Association has
fostered the application of a particular method of job evaluation in the
iron and steel industry. In several countries, including the United
Kingdom, individual large employers have adopted a system of job
evaluation.
During the Sixth Session of the I.L.O. Metal Trades Committee in
May 1957 the Employers' group stated, among other things—
. . . The discussions in our own group, which have gone into some detail of the
various systems in operation and the experiences of the working of these
systems, have brought a much greater appreciation of all that is involved in
job evaluation. These have also brought to light the difficulties which result
from the varying conditions in the different countries and even in the different
industries or undertakings in the same country. We believe, therefore, that it is
impracticable to lay down any agreed uniform system. On the other hand, we
would not wish to underestimate in any way the benefits which have accrued
in a number of cases on the actual operation of job evaluation as a means of
setting up an equitable pay structure.
In view of what we have said, and particularly bearing in mind that the
development of job evaluation schemes has necessarily been slow, we have
come to the conclusion that there should continue to be free scope for further
experiments and development, without laying 1down any hard and fast rules for
recommendation to individual undertakings.
Briefly, while it cannot be said that employers generally are strongly
in favour of job evaluation, there also appears to be little opposition
by employers to the method.
Trade Unions
Negative Views.
The position is different on the trade union side. In the words of the
A.F.L.-C.I.O. bulletin quoted earlier " there is no single or over-all
union attitude or policy towards job evaluation " 2, but it seems fair to
say that among trade unionists there is a great deal more in the way of
frank opposition to the method than among the employers. Sometimes
trade union criticism is directed at particular systems or procedures used.
For example, the metalworkers' union in the Federal Republic of
Germany listed the following difficulties encountered in applying job
evaluation schemes used in their country:
(1) Inadequate information on a scheme is given to shop stewards
and workers.
1
Ï.L.O., Metal Trades Committee, Sixth Session, Geneva, May 1957: Fraport of
the Subcommittee on Job Evaluation Methods in the Metal Trades (mimeographed),
pp. 2-3.
2
Collective Bargaining Report, op. cit., p. 33.

104

JOB EVALUATION

(2) The prescribed methods are not always followed.
(3) There are not sufficient safeguards for the workers.
(4) There are difficulties in grading highly skilled jobs and handicraft
and partly automated jobs. 1
The Indian National Trade Union Congress considers the scheme
used in some Indian plants to be very defective—
. . . because of the absence of consultation with the workers in the evaluation
of a scheme . . . , the failure to inform them of the details of the scheme, the
absence of union co-operation in the conduct of job evaluation, the failure on
the part of the evaluator to carry out essential foundational procedures such as
job analysis and description, the failure to evaluate working conditions and
hazards and the assumption of differentials instead of their ascertainment as a
result of job evaluation.2
In other cases trade unions have strongly criticised the method as
such. Thus, according to a manual published by the International
Association of Machinists in the United States—
I.

Basically, job evaluation tends to limit collective bargaining. This reflects
itself in the following ways:
1. It tends to freeze the wage structure and thereby creates an obstacle
to the correction of inequities. It restricts the right of negotiating on
a rate of pay for each job year after year. It usually limits negotiations to bargaining for a fixed amount or fixed percentage for all jobs,
or establishing rates of pay through some " predetermined formula "
that usually does not result in equitable treatment for all.
2. It fails to consider all forces which determine wages, such as supply
and demand, other contract or area rates, etc.
3. It tends to create a barrier between the employee and his understanding of his own job rate, because his rate is set in a manner not understood by him.
4. It tends to disregard the ability of the individual.
5. It places a ceiling upon wages which is contrary to a traditional
objective of organised labour.
6. It disregards compensation for loyalty, i.e. years of service, etc.
7. It tends to dilute traditional skills, creating many new occupations
and many new classifications and thereby reducing wages.
8. It affects the seniority of employees by the creation of additional
classifications.
9. It makes the promotion of employees into higher-paying jobs considerably more difficult because of the limiting characteristic of job
descriptions.
10. It provides the company with a tool to downgrade employees during
times of cutbacks.
1
2

Communication to the I.L.O., 8 Aug. 1956.
Communication to the I.L.O., 20 Aug. 1956.

ADVANTAGES AND LIMITATIONS OF JOB EVALUATION

105

II. Job evaluation presents a threat to the stability of our organisation because
of the following:
1. It necessitates the constant attention of additional trained representatives, thereby increasing the cost of representation to the local,
district and Grand Lodge.
2. It provides management with a tool to play one group of employees
against another.
3. It creates dissension within a local lodge (where members are from
more than one firm) where all firms do not have job evaluation. It
tends to hamper the efforts of the lodge in establishing uniform area
rates.
4. It tends to place the responsibility upon the union for inequities that
are not properly corrected since the union accepted the job evaluation
plan and must, therefore, share in its shortcomings.
5. It compels the continuing and almost impossible task of educating
job study committees and shop stewards in the many ramifications of
the job evaluation plan in effect.
6. It encourages managements of different plants to work together and
provides them with a basic method to achieve jointly desired results
in the determination of wages; it strengthens management's opposition to the wage demands of the union.
III. The effects of job evaluation upon the general welfare of our society is
discernible as it affects the supply of skilled workers. It tends to discourage
bona-fide apprenticeships and, therefore, reduces the reservoir of over-all
skilled workers so that in the event
of a future crisis a serious shortage of
skilled manpower would result.1
Strong views against job evaluation were also expressed in the
statement of Worker members of the subcommittee on job evaluation
at the meeting of the I.L.O. Metal Trades Committee in May 1957.a
The statement concluded that "job evaluation would not promote
industrial harmony ". The method was not really scientific, it did not
fully take into account all the relevant factors determining equitable
wages and it was so complex as to be largely incomprehensible to the
workers and to disturb labour-management relations. The statement
continued as follows:
. . . Basically, job evaluation tends to limit collective bargaining and to freeze
the wage structure. It seeks to substitute would-be technical standards for
market forces as reflected in collective bargaining. One of the main weaknesses
of job evaluation is that it is insufficiently adaptable to the dynamic elements of
our economy, as they affect the process of wage determination. The problem
of determining the different job difficulties and wage categories can only be
solved by collective agreements. The Workers' members therefore hold the
view that collective bargaining is the best instrument for maintaining an
equilibrium between management and workers.
'International Association of Machinists: What's Wrong with Job Evaluation
(Washington, 1954), pp. 3-5.
* I.L.O., op. cit., pp. 3-4.

106

JOB EVALUATION

Favourable Views.
There are, however, examples of less critical views on the subject.
Mention has already been made of the wholehearted participation of the
United Steelworkers of America in the far-reaching United States steel
plan which the union found to have eliminated many current and prospective wage grievances. Other examples of positive trade union attitudes
in the United States include the initiative of the Commercial Telegraphers' Union which designed a job evaluation scheme that was
subsequently adopted by the Western Union company, and joint unionmanagement participation in job evaluation for the West Coast paper
and pulp industry and for the New York newspaper trade.1
One trade union expert on management engineering, attached to
the International Ladies' Garment Workers' Union, though emphasising
that job evaluation is a subordinate tool in collective bargaining and that
its results can be only one factor in determining the wage structure, does
not oppose the method. He considers that the most useful aspect of job
evaluation is research " designed to isolate the factors that have governed
the intuitive operation of collective bargaining.. . these factors can then
be used for future guidance ".2
Reference has already been made to the critical views of the metalworkers' union in the Federal Republic of Germany on job evaluation
as applied in the West German steel industry. However, in the study
issued by the High Authority of the European Coal and Steel Community,
which has been referred to in previous chapters, the trade union experts
also mentioned certain more positive " facts or alleged advantages "
put forward by the works councils of the six undertakings surveyed.3
These were the following:
(1) Before the introduction of job evaluation it was possible, and even
customary, tofixwages for particular workers or jobs in an arbitrary
fashion. The objectivity of job evaluation makes this impossible.4
(2) The establishment of a joint evaluation committee ensures the
permanent participation of the workers' representatives in the fixing
of the firm's wages on a basis of complete equality, including those
aspects which are outside the scope of job evaluation.
1
See S. A. LEVITAN: " Union Attitudes towards Job Evaluation and Ingrade
Progression ", in Industrial and Labor Relations Review, op. cit., Vol. 27, No. 4,

Jan. 1951, p. 270; and BAKER and TRUE, op. cit., p. 103.
2
W. GOMBERG, op. cit., p. 7. See also, by the same author, A Labor Union Manual
on Job Evaluation (Chicago, Roosevelt College, 1948).
3
Haute Autorité, op. cit!, p. 269.
4
The trade union experts observed on this point that the same advantages could
also have been achieved without the introduction of job evaluation (op. cit., p. 268).

ADVANTAGES AND LIMITATIONS OF JOB EVALUATION

107

(3) When, as a result of new working conditions, additional or new
efforts are called for in a job, the workers' representatives are able
to obtain a suitable wage adjustment more quickly and effectively
than would otherwise be possible.
(4) The workers' representatives are relieved of daily discussions,
explanations and debates about certain wage questions and can give
much more attention to the settlement of wage claims concerning
various jobs or small working teams.
Certain Italian trade union organisations have also shown themselves
favourably inclined towards job evaluation. The Italian Confederation
of Trade Unions (Confederazione Italiana Sindacati Lavoratori) has
expressed the view that job evaluation will give the unions a say in
connection with " new methods of wage fixing linked to technical
progress " and ensure that wage adjustments are not left entirely to the
management, since this would tend, even if such adjustments were
favourable to the workers, to weaken the influence of the unions.1
Furthermore, the Confederation has expressed the hope that job evaluation will reduce management's power of unilateral decision in matters
of transfer and promotion and, more generally, permit equitable and
responsible participation of the workers in the life of the enterprise.
As regards wages, the Confederation considers that job evaluation can
bring about a more rational and equitable wage structure, ensure the
elimination of all discrimination between male and female wages, and
prevent conflicts among workers which might otherwise result from
individual wage increases granted by unilateral management decision.
Representatives of the Italian Workers' Union (Unione Italiana del
Lavoro) have also come out in favour of job evaluation adding, however,
that certain conditions must be met. Particularly, there must be a satisfactory state of labour-management relations. Also, although job
evaluation is recognised as an important instrument for improving
labour-management relations within firms, trade union participation
in the programme is regarded as essential. Dealings between management and the workers of the firm without union participation might
improve workers' remuneration but in an atmosphere of paternalism
which in the long run could lead to deterioration in labour-management
relations.2
In France the metalworkers' union affiliated to the General Confederation of Labour (Force ouvrière) and that affiliated to the French
1
Haute Autorité, Communauté européenne du charbon et de l'acier: La qualification du travail et ses applications dans la sidérurgie de la Communauté: Italie (provisional
mimeographed edition, Luxembourg, Sep. 1957), p. 78.
«Ibid., p. 81.

108

JOB EVALUATION

Confederation of Christian Trade Unions have both expressed willingness to accept job evaluation.1 It should, however, be borne in mind
that job evaluation is not widely applied either in Italy or in France and
that there are no indications that the large national trade union organisations (the Confederazione Generale Italiana del Lavoro and the Confédération générale du Travail) would accept or favour its application.
Mention has already been made of the views of the Netherlands
Federation of Trade Unions, namely that job evaluation is desirable
from the point of view of a firm's internal wage structure and is necessary
also for the co-ordination of wages between firms and industries.
Some Problems Involved in Trade Union Acceptance of Job Evaluation.
The primary task of trade union leaders is to safeguard and promote
the well-being of their membership. Secondly, the leaders will normally
feel responsible for the growth of the particular organisation which they
represent; this may be affected by a variety of forces, including action
by employers, rival trade unions or conflicting sectional interests within
the union itself. These and other factors are apt to influence the union
leadership, particularly in its strategy and tactics in the important field
of wage negotiations, and thereby contribute to shape its attitude towards
job evaluation as a possible issue in wage bargaining. Some of the unions'
opposition to the method can no doubt be explained in these terms.
As was mentioned above, trade union representatives have on
occasion rejected job evaluation on the ground that it limits or does
away altogether with collective bargaining. From the description of
methods and procedures contained in Chapter II, it will be clear that job
evaluation can, if both parties agree, be dealt with by joint consultation
and collective bargaining, whatever the machinery set up for these
purposes. Indeed, it is difficult to see how, in cases where such
machinery exists and operates effectively, job evaluation could ever
be applied as a means of unilateral wage-fixing by the employer.
This does not mean, however, that the method does not raise certain
general problems for the trade unions. In addition, there may be specific
conditions that make it difficult for a trade union to accept job evaluation as an aid in wage fixing.
Among the general problems arising when job evaluation is newly
introduced are its sheer novelty, its complexity, and the unpredictability
of its results. To what extent will the existing wage structure be changed
after job evaluation ? How will the membership react to these changes ?
1
Haute Autorité, Communauté européenne du charbon et de l'acier: La qualification du travail et ses applications dans la sidérurgie de la Communauté: France (provisional mimeographed edition, Luxembourg, Sep. 1957), pp. 41 ff.

ADVANTAGES AND LIMITATIONS OF JOB EVALUATION

109

Will a majority of the members understand why their representatives
agree to follow an entirely new, and probably slow, method of dealing
with their urgent and legitimate wage claims? Will reliance on job
evaluation technicians in dealing with wage problems not weaken the
authority of the trade union leaders ? Will local union officers be able
to cope with the complexities involved in some systems when negotiating
with technicians specialised in the subject, and is there not a danger
that they will be at a disadvantage on such occasions?
Another general problem is the fact that formal acceptance of rules
governing relative wages tends to restrict the scope for maneuver in negotiations—no bargaining agent will be disposed to welcome this. Even
though it may be possible to use job evaluation for adjustment of an
existing wage structure in a way that is acceptable to the large majority
of the membership, the trade union leader may fear that in the future
wage claims will be put forward by some of the membership that cannot
easily be negotiated within the framework of an accepted system or,
alternatively, that following a reorganisation of production, the system
may require adjustments that will be opposed by some of the members.
This problem is particularly important in view of the general trend
towards mechanisation and automation in modern industry. In most
cases, this leads to a reduced importance of physical effort in the jobs
affected; in some cases, it leads to a reduction in skill requirements
(though in other cases it may call for an increase in particular skills);
and often it involves, or may be construed to involve, an increase in the
worker's responsibility. In practically all cases the introduction of more
highly mechanised or automated methods of production will affect the
relative importance of one or more of the factors commonly used in
job evaluation. Its impact on the wage structure, according to the job
evaluation system used, is not easy to predict. There is, of course, no
reason to assume that the impact will be a negative one; on the contrary,
higher productivity will normally enable the trade union to negotiate
a general wage increase. Yet the new ranking of jobs may cause serious
tensions both in the plant and within the trade union.
Apart from these general problems, there may be specific conditions
making it difficult for a trade union to commit itself to job evaluation.
For example, a new trade union engaged in an organisation campaign
may not be interested in a device that settles wage grievances by more or
less automatic application of pre-established rules. It may prefer to deal
with such grievances on an ad hoc basis in order to impress workers with
the advantages of joining. This aspect is particularly important when
more than one union is trying to organise a particular group of workers.
Any union committing itself to a particular job evaluation system is

110

JOB EVALUATION

then open to attacks from other unions who may seek to exploit wage
grievances as a means of recruiting new members. Alternatively, when
job evaluation is applied with the participation of a number of rival
unions, each of these may be continually trying to obtain upward reclassification of individual jobs to show its value and fighting spirit.
For example, in the southern California aircraft industry, the International Association of Machinists and the United Auto Workers were
fighting for members at a time when the job evaluation system had already
been introduced, and efforts to " squeeze the gold " out of the system
resulted in over 4,000 wage grievances being filed at Lockheed plants
in one year.1 In other cases, a union's membership itself may include
groups with more or less conflicting interests. When an industrial
union includes large numbers of both skilled and unskilled workers, it is
quite possible that one of these groups will obtain a relative improvement
in wages; this may place the leadership in an embarrassing position
vis-à-vis the other group. Or job evaluation may involve wage rates lower
for certain jobs than those prevailing in the local employment market;
quite apart from the problem of " hiring rates " discussed in Chapter III,
acceptance of a plan involving such lower rates may expose the trade
union leadership to criticism from some of its members.
Some Conditions for Union Participation in Job Evaluation.
Some of the difficulties that may prevent a trade union from accepting
job evaluation are more serious than others. Most of them, however,
point to certain conditions that have to be met if the method is to be
actively supported by the unions.
A first condition would seem to be that the union is relatively well
established and feels reasonably secure. It will be easier for the leaders
of a union to commit themselves to any scheme when the organising
phase of its development is over; it will also be easier if the union is not
engaged in battles with rival organisations and if the leaders' authority
among the membership is not seriously disputed. All of these conditions
were fulfilled, for example, when the United States steelworkers' union
agreed to participate in the job evaluation scheme for their industry.2
Secondly, the scheme adopted should be as simple as possible so
that the individual worker can understand it; it will usually be necessary
to distribute suitable explanatory pamphlets and other documentary
material.
Thirdly, once a union has agreed to accept a job evaluation system,
this should be implemented as a truly joint undertaking. Suitable pro1

KERR and FISHER, op. cit., pp. 85-86.

* TILOVE, op. cit., pp. 972-973.

ADVANTAGES AND LIMITATIONS OF JOB EVALUATION

111

cedures should be devised and genuinely applied for consultation at all
stages, including particularly that of job rating. Unless the union agrees,
job descriptions should not be rewritten without its approval after the
scheme has been installed. In particular, it should be clearly understood
that the process of job evaluation ceases with job classification and that
the determination of wage rates for jobs or job rates remains a separate
subject of collective bargaining.
Individual workers should also be given a suitable measure and sense
of participation; depending on the circumstances, they may be asked
to participate in the preparation of job descriptions or at least be informed
of the final text of their job descriptions, and clear and effective ways of
filing reasonable claims should be open to them. Normally, no worker's
wage should be reduced and no worker downgraded as a result of the
introduction of a job evaluation scheme.
It further seems essential that systems should be designed and operated in a flexible manner enabling a large number of special considerations to be taken into account. Reference has already been made to the
view that job evaluation is a matter of industrial relations rather than of
industrial engineering.1
Flexibility is partly a matter of technical arrangements, partly a
matter of attitude. As an example of the former, technical change may
be taken into account by judicious selection of factors and weights
anticipating, to some extent, shifts in the relative importance of various
job requirements. Thus, if relatively low weights are assigned to such
factors as physical effort and manual skill, while greater weight is
attributed to responsibility for equipment, certain effects of automation
on the evaluated job structure may be mitigated. Questions of attitude
arise, for example when small changes occur which under strict application of the system would lead to downgrading of jobs. Suppose, for
example, that a firm discontinues the production of a particular type of
heavy steel shaft, the threading of which accounted for most of the
threading operations performed by an engine lathe operator. The need
for threading being thus reduced, the job description could be revised
immediately to take account of reductions in the level of training and of
physical effort required, and a lower point score and wage rate could be
fixed for the job.2 Such strict application of the system could easily give
1
Stieber, however, found that when industrial engineers have to negotiate directly
with workers' representatives, they are generally inclined towards flexibility as much
as industrial relations officers are: ".. . in fact, greater flexibility in the administration
of the manual was found in those companies where the industrial engineering department dealt directly with the union than in companies where there was a sharp division
between industrial engineering and industrial relations responsibilities in day-to-day
operation of the program." (Op. cit., p. 323.)
8
Collective Bargaining Report, op. cit., pp. 38-39.

112

JOB EVALUATION

rise to awkward problems, but it is difficult to devise technical rules for
the way in which such cases should be handled. It is, however, clear
that a great deal of irritation and suspicion in regard to a job evaluation
system can easily be avoided if the plan is administered in a spirit of
understanding and goodwill. No doubt a trade union will find it easier
to accept such a system if it feels assured that the employer is willing to
have it administered in such a spirit, avoiding hardships that fully
automatic and inflexible application may involve.
CONCLUSION

From a technical point of view, job evaluation constitutes a device
for the settlement of certain more or less troublesome questions of wage
adjustment. This is the angle from which the subject has been primarily
examined in this study. To the extent that the method is useful for such
purposes, however, this is largely because it attempts to base wage
differentials on considerations that are not purely technical, but that
have, in some degree at least, an ethical basis. Much of the political
debate and decision-making in modern society is concerned with issues
of fairness and equity in the distribution of income. These terms, however, are often used in a vague and arbitrary sense, and actual decisions
often reflect considerations of economic expediency and of political
power more than anything else. Job evaluation seeks to give practical
expression to two principles of fairness that are so widely recognised that
they cannot be regarded as mere subjective assertions inspired by group
interests: that of equal pay for equal work and that of differential reward
in accordance with discernible differences in the sacrifices that the performance of productive work requires in terms of education, training,
personal application and the endurance of adverse conditions. This
objective of job evaluation gives the method a broader interest than it
would have as a mere wage-fixing device.
At the present stage the method is rather restricted in its application.
In a highly technological society aspiring towards ever higher standards
of economic achievement, notions of equity may come into real or
apparent conflict with other criteria for the determination of relative
wages—for example an accepted or alleged need to maintain or increase
certain incentives, or to bring about a certain allocation of labour among
different industries or occupations. In this study a number of instances
have been mentioned in which the objective of superior economic performance may limit the possibilities of applying the particular rule of
justice that job evaluation seeks to put into practice. Mention has also
been made of various devices that have been used or suggested as pos-

ADVANTAGES AND LIMITATIONS OF JOB EVALUATION

113

sible ways of reconciling the real or apparent conflicts that may arise, but
it seems fair to say that some of these questions have not yet been
satisfactorily settled. Further experience and research seem needed to
discover more or better means of achieving this end.
The present limitations in the application of job evaluation are
probably also due in part to such factors as lack of knowledge about
its meaning and unwillingness to accept its results where these would
be to the disadvantage of particular groups. The increasing experience
with the method and the growing volume of literature on the subject
may be expected to lessen the former difficulty.
It is interesting, and perhaps significant, that job evaluation has so
far been applied mainly to manual jobs, that is to say to occupations
ranking relatively low in the general distribution of income in most
countries. The highest-paid occupations in the economy, which one
would expect to raise even more pertinent questions as regards fairness
and equity, have rarely been subject to comparable forms of analysis.
It is generally believed that such occupations are too complex for job
evaluation. It is true that the factor descriptions commonly used in
existing job evaluation schemes could not do full justice to the requirements of, for example, many executive and senior administrative jobs.
Yet it would be surprising if such jobs did not lend themselves at all to
the type of rational discussion and analysis which job evaluation applies
to wage-earning jobs. Here again, further research may be very fruitful
and lead to interesting results ; the nature and importance of professional
qualities that completely defy description and analysis might be found to
be much smaller than is sometimes assumed.
The present study is not intended as a plea for job evaluation. Indeed
many readers may not feel satisfied by the arguments and experiences
cited that the method can really honour its promise of providing a tool
for the establishment of an equitable and workable structure of wages
and salaries. However, rational discussion being the first necessary step
towards understanding and agreement, it seems justified to hope that
the method will provide at least a beginning of a solution for some
important problems of economic justice in remuneration. Further
progress, while it cannot be ensured, will only be possible if the problems
involved are more closely studied in broader circles than has been the
case hitherto. This report will have served its purpose if it stimulates
such discussions.

APPENDICES

APPENDIX A
I.L.O. METAL TRADES COMMITTEE
(Sixth Session, Geneva, May 1957)

Extracts from the Report of the Subcommittee on Job Evaluation Methods
in the Metal Trades

STATEMENT OF VIEWS OF EMPLOYERS CONCERNING JOB EVALUATION

6. The Employers' members submitted the following statement:
We wish to place on record our appreciation of the informative and detailed
report * prepared by the International Labour Office on this question of job evaluation
in the metal trades.
We believe that the report presents a good summary of the experience with job
evaluation in a number of countries in the Western world. The great body of information in this report seems to be predicated upon the experience of these schemes in the
United States. With the exception of a brief statement concerning the situation in
India, there is practically no information from the industrially underdeveloped
countries.
But for the International Labour Office report and the discussions we have had,
the delegates from the various countries could not have been fully aware of all the
implications and possibilities of job evaluation or of the various systems practised
and the results obtained. We would, however, wish to emphasise that this is not an
entirely new conception in the art of ranking jobs and of establishing equitable relationships between the payments for the varying jobs met with in different plants.
The discussions in our own group which have gone into some detail of the various
systems in operation and the experiences of the working of these systems have brought
a much greater appreciation of all that is involved in job evaluation. These have also
brought to light the difficulties which result from the varying conditions in the different
countries and even in the different industries or undertakings in the same country.
We believe, therefore, that it is impracticable to lay down any agreed uniform system.
On the other hand we would not wish to underestimate in any way the benefits which
have accrued in a number of cases on the actual operation of job evaluation as a means
of setting up an equitable pay structure.
In view of what we have said, and particularly bearing in mind that the development of job evaluation schemes has necessarily been slow, we have come to the conclusion that there should continue to be free scope for further experiments and development without laying down any hard and fast rules for recommendation to individual
undertakings.
We would welcome the International Labour Office continuing their studies of
this subject and making available from time to time to the member States the results
of their studies and information on the experience gained.
1
1.L.O. : Job Evaluation Methods in the Metal Trades, Report III, Metal Trades Committee, Sixth Session, Geneva, 1957 (mimeographed).

118

JOB EVALUATION
STATEMENT OF VIEWS OF WORKERS CONCERNING JOB EVALUATION

7. The Workers' members submitted the following statement:
Having discussed and very carefully examined the substantial report prepared by
the Office, and having further examined the various concepts of job evaluation in the
metal trades, the Workers' members came to the conclusion that job evaluation would
not promote industrial harmony.
The claim of job evaluation to be " scientific " is unfounded, for it is based on
value judgments which are essentially subjective. Nor does it take fully into account
every factor of significance in determining " equitable rates " for different jobs.
Moreover, the complexity of job evaluation systems is such as to render them largely
incomprehensible to the workers, thus creating dissatisfaction and disturbing labourmanagement relations.
Basically job evaluation tends to limit collective bargaining and to freeze the wage
structure. It seeks to substitute would-be technical standards for market forces as
reflected in collective bargaining. One of the main weaknesses of job evaluation is
that it is insufficiently adaptable to the dynamic elements of our economy, as they
affect the process of wage determination.
The problem of determining the different job difficulties and wage categories can
only be solved by collective agreements. The Workers' members therefore hold the
view that collective bargaining is the best instrument for maintaining an equilibrium
between management and workers.
Owing to the controversial nature of job evaluation, it is not a fit subject for recommendation at the international level.
COMMENTS OF EMPLOYERS', WORKERS' AND GOVERNMENT MEMBERS
ON JOB EVALUATION

8. At various stages during the different sittings of the subcommittee, the
Employers', Workers' and Government members, as groups and individually,
made comments on job evaluation.
Comments of Employers'' Members
9. The Employers' members indicated that many of them who followed
the developments in their own countries were not aware of the systems and
practices applied in other countries. From the exchange of views among the
Employers' members it was clear that there was no uniformity in respect of
job evaluation. On the contrary, there was a wide range of practice. It appeared,
therefore, that it was essential to learn a good deal more about job evaluation
before they would be able to arrive at conclusions concerning it. The
Employers' members said that it appeared from the statement submitted
by the Workers' members that they had arrived at conclusions somewhat
similar to those of the Employers' members who, however, felt that, in view
of the complexity of the subject, the Office should be invited to continue its
studies of job evaluation.
10. The United States Employers' member said that the discussion in the
subcommittee showed that there was a complete misunderstanding concerning
job evaluation and the role it had to play. Nevertheless, the United Steelworkers of America, a number of the local unions of the International Association of Machinists and some other trade unions or leaders were in favour of
job evaluation. However, it was obvious that there were a great number of
controversial issues regarding job evaluation, the subject being new. For
that reason the Employers' members, while congratulating the Office on the
excellent report submitted to this session of the Committee, invited the Office
to continue its studies of this subject.

APPENDIX A

119

11. The Employers' member from the Federal Republic of Germany
stated that the employers in his country were in favour of job evaluation
because they found it to be of considerable assistance in wage-setting and in
arriving at a proper basis for the calculation of production costs. It also
assured an equitable wage structure for the workers.
12. The Finnish Employers' member said that the advantages of job
evaluation could not be contested although he agreed that there were practical
difficulties in applying the different systems.

Comments of Workers' Members
14. The Workers' members indicated that they had discussed the report
submitted by the Office and, in the light of the contents of this report as well as
their own experience, had come to the conclusion that it would be difficult to
formulate a uniform concept of job evaluation methods in the metal trades.
The situation differed in the various countries. The Workers' members also
considered that there was no similarity of views between the statements of the
Employers' and Workers' members. The matter was not fit for discussion at
the international level and the workers also had serious doubts whether it
was fit for discussion at the national level. Further, it was controversial as
shown by the exchange of views which had already taken place in the subcommittee. The Workers' members considered that job evaluation tended to
limit collective bargaining and to freeze the wage structure. It was insufficiently
adaptable to a dynamic economy. Before reaching any conclusions, discussions
should continue, within each country, in respect of basic problems of job
evaluation
16. A Workers' member from the Federal Republic of Germany said that
job evaluation afforded no proper differentiation between skill and maintained
it did not eliminate differentiations based on sex or age. It affected the wage
structure because the continually changing methods of job evaluation upset
the wage scale and were often contrary to that agreed by collective bargaining.
Job evaluation was also sometimes introduced by management without consulting either the workers or the trade unions and it obviously made the
workers suspicious as to the management's intentions and the value of the
scheme. Moreover, it was impossible to lay down standards applicable to
industry as a whole or to a particular branch of industry. Workers considered
job evaluation as a one-sided method which in the long run would lead to
low wages. Further, it presented evident dangers in countries in which the
trade union movement was not yet fully developed.
17. Another Workers' member from the Federal Republic of Germany
stated that the German Metal Workers' Federation h a d . . . pointed out that
several substantial difficulties had been encountered in applying job evaluation
schemes and that they had serious doubts as to the usefulness of this method.
They considered that job evaluation fell within the category of management
techniques and that it should be treated as such. Any grading of workers and
consequently their wages should be discussed and fixed by negotiation between
management and trade unions and not by the introduction of methods which
had a doubtful scientific value. It was clear that workers should take part in
fixing their own wage structure and there was a danger that job evaluation
might eventually lead to a split between workers and management.
18. The United States Workers' member considered that job evaluation
was not suitable for international conclusions. It was the I.L.O.'s duty to

120

JOB EVALUATION

promote conclusions which would result in a higher standard of living for the
workers and lead to industrial harmony. The workers' experience showed that
job evaluation promoted neither. The International Association of Machinists,
which had collective bargaining covering over a million workers in the United
States and Canada, was opposed to job evaluation as such because it considered
it as an instrument militating against the best interests of the workers. The
National Metal Trades Association of the United States, one of the authors and
promoters of job evaluation, conceived job evaluation as a substitute for
collective bargaining. The workers considered that wages should be fixed
through collective bargaining and not by job evaluation methods. It was
obvious that if the members of this subcommittee, who had all had experience
of job evaluation, were unable to agree on the matter, it was unlikely that
ordinary workers would understand what was meant by it. Moreover, job
evaluation took human dignity away from the workers because it treated them
as if they were numbers in a scheme of a mathematical formula. The whole
system was not adaptable to the dynamic elements in the modern economy.
19. The Canadian Workers' member, referring to the United States
Employers' member's remarks regarding statements made by various trade
unions in support of job evaluation (see paragraph 10 above), said that he
wished to remind this member that both the American Federation of Labor
and the Congress of Industrial Organisations were unanimously opposed to
job evaluation. The workers did not share the view which was held by some
that job evaluation was helpful in promoting training. Moreover, they considered that many of the job evaluation schemes introduced by management
aimed at breaking down skills and that they might therefore result in the
breaking up of craft unions.
20. The United Kingdom Workers' member said that neither the United
Society of Boilermakers, Shipbuilders and Structural Workers ofwhichhewas a
representative nor other metal trades workers' unions in the United Kingdom
had any intention of accepting job evaluation. It did not seem that the Employers' members accepted the principle of collective bargaining on job evaluation. This attitude could not be reconciled with the view that job evaluation
should lead to increased benefits to the workers. Job evaluation was, in his
opinion, one of the means for breaking down the skills and consequently it
was aimed against craft unions. It should, however, be remembered that
craft unions were of great importance for the metal trades.
21. The statement submitted by the Workers' members was, in his opinion,
very objective. It pointed out the difficulty in reaching international conclusions . . . . The workers realised that job evaluation might be good for some
countries but that it was unsuitable for others. The document submitted by
the Workers' members placed the right emphasis on collective bargaining and
covered all the essential points mentioned during the discussion in the subcommittee.

Comments of Government Members
24. The United Kingdom Government member said [that] even if the
initiative for job evaluation and work study came from management the
workers should be able to make their point of view known from the very
beginning. This might be done through discussions similar to that which was
taking place in this subcommittee. Some experiments with job evaluation had

APPENDIX A

121

proved to be successful and if any further experiments were to be made they
should take advantage of what had already been done in this field. It was in
this way only that proper safeguards might be introduced so that opportunity
would be given to workers and their organisations to have their views heard
right from the very beginning
26. The Indian Government member said that during the discussion
suggestions had been made that the I.L.O. should continue its studies on job
evaluation. It might therefore be helpful to state the types of problem to which
the I.L.O. should give attention. Job evaluation could be looked at from
three different standpoints: technical, psychological and those concerning
countries which were trying to industrialise rapidly.
27. As regards the technical aspects, it would be useful to have case studies
concerning selected plants in a number of countries, which should be carefully
chosen for the purpose. Such case studies should be made both in regard to
plants which had introduced job evaluation systems and those which had no
such systems. This would make it possible to compare the wage structure
obtaining in each type of plant. As conditions varied from industry to industry,
it would be desirable to select trades where job evaluation could be introduced
with least difficulty. Further, attention should be focused on selected key
trades because job evaluation was not necessarily applicable to all trades.
The different essential factors should be taken into account, their choice being
left to the industry concerned along with the choice of weights. Consideration
should also be given to the circumstances in which job evaluation would be
introduced, so as to ascertain how far it was suitable for old establishments
as well as for new plants. The size of establishments should also be taken into
account. In this connection, attention should be paid to productivity studies
and to the question whether such studies should precede job evaluation.
28. With respect to the industrially less advanced countries, the part that
might be played by job evaluation should be ascertained and comparisons
made with progress achieved in industrially developed countries. It was
necessary to make a distinction between the two types of countries, so that the
less advanced countries could benefit from the experience of those which were
more developed. Further, consideration should be given to whether the metal
trades should wait until industry as a whole reached a higher level of development, or whether it should give a lead irrespective of the state of general
development. Attention should also be given to the influence of the wage
level and the selection of areas for job evaluation, particularly those where
wages were low.
29. As regards psychological aspects, the following points might be taken
into consideration. Regard should be given to consultation with trade unions,
taking into account the level at which such consultation should be carried out,
its frequency, timing and manner. Consideration should also be given to
conclusions that might be drawn from experience in regard to certain essential
safeguards that concerned either individuals or groups or both. There was
undoubtedly a strong opposition by trade unions to job evaluation and this
opposition could not be ignored because any scheme which would not take
into account the fears and doubts entertained by the trade unions would fail.
Schemes should also provide for properly trained personnel who would be
entrusted with the task of introducing job evaluation, which should be introduced stage by stage, depending on the atmosphere prevailing in a particular
plant. During the whole process, the state of development should be continually kept in view and due account should be taken of the desirable priorities.

122

JOB EVALUATION
EXPERIENCE WITH JOB EVALUATION SYSTEMS

Government Members' Statements
33. The Canadian Government member said t h a t . . . in Canada job
evaluation was introduced in a small sector of the Federal Public Service.
The government employees were roughly divided into three classes: (a) top
civil servants appointed by Order in Council; (b) the bulk of employees
appointed by the Civil Service Commission who were paid the same rate everywhere in Canada; and (c) the prevailing rate employees—as they were called—
paid according to the rates prevailing in the locality in which they were employed. About ten years ago dissatisfaction arose among canal employees
who were in the last group because on each canal the wage rate changes were
not uniform : some classifications might receive increases on the prevailing rate
basis and other classifications would not. Consequently, a job evaluation plan
for these employees was established on a tripartite basis; a committee consisting of one representative each of the employing department, the workers
and the Department of Labour prepared job descriptions and awarded points
for each job. This established a known and accepted wage relationship amongst
the classes on each canal. This plan had been operating successfully. A second
plan of a similar nature was set up nine years ago, at the request of the workers,
for the atomic energy plant operated by the National Research Council.
Experience had shown that unskilled and semi-skilled workers could be
recruited locally, whereas skilled workers had to be recruited in Toronto.
Further, there was need for special safeguards in view of the nature of the work.
Three years ago the Government set up a public corporation to run the atomic
energy plant. Since then collective bargaining had been introduced and the
wage structure had been established on the basis of collective agreements.
These collective agreements retained the same wage differentials that had
previously been established under the job evaluation plan. These two examples
concerned a small sector of public service. They were conducted on an experimental basis and proved to be a success. In private industry, job evaluation
was used to a small extent. The United Steel Workers of Canada used job
evaluation and union business agents said that it was satisfactory. It appeared
that job evaluation had been accepted as a part of their collective agreements.
34. The Canadian Government member further observed that the Employers' members' statement... did not take a stand either for or against job
evaluation. The Employers' members apparently felt that this matter was new
and required further study. This was true and it was therefore clear that this
question should be fully discussed in an international forum such as this Committee. The Workers' members' statement was interesting and useful. For
example, it pointed out that the claim of job evaluation to be scientific was
unfounded. In his opinion, job evaluation was a useful tool for establishing
the wage structure and had the best possibility of successful application where
it was established under the collective bargaining process.
35. The Chairman in his capacity of a Government member for the Netherlands outlined the position obtaining in his country. Following the war, it
was necessary to adopt a national wage policy which was worked out in close
co-operation between the Government and the employers' and workers'
organisations. At the beginning, the workers were classified in three categories,
i.e. unskilled, semi-skilled and skilled. This rather rough classification was
sufficient for some branches of industry. At a later stage, the better-known
systems of job evaluation had been studied and compared by an expert committee set up for this purpose. This study resulted in the elaboration of a
normalised method. Job evaluation systems were established in several indi-

APPENDIX A

123

vidual plants and led to fairly satisfactory results. There were some difficulties
which were due to the complexity of the systems and to the fact that they had
not always been understood. Job evaluation systems allowed for a more
extensive differentiation between the minimum and maximum wage. The
unemployment benefit was usually fixed as a percentage of a normal wage.
If the wage depended on the market conditions alone, it would, at certain
times, drop to a very low level and consequently unemployment benefit would
be lower still. Most employers no doubt felt that it was useful to introduce
a system of job evaluation so as to have a more objective and scientific instrument for the determination of wages and production costs. The whole problem
was of great importance
36. With the permission of the subcommittee the Chairman read the
following statement on behalf of the other Netherlands Government delegate:
Before the war distrust of job evaluation and of incentive wage systems prevailed
in the Netherlands, following a long period of unemployment and unsettled economic
conditions. During the years of the war, through secret meetings of representatives of
workers' and employers' organisations, better relations had been established which
led to the creation of the " Foundation of Labour " where all general social problems
were discussed every week
It was not possible to remove the distrust of any wage system unless there was
co-operation between employers and workers. Job evaluation was closely connected
with the worker's wage. It was linked with the continually increasing need for highly
skilled workers but it should be considered as the first step in the whole process of
wage fixing. Further, job evaluation and wage fixing should be kept apart. In the
past, there had always been some kind of job evaluation in a rough and ready way.
Usually, workers were divided into skilled, semi-skilled and unskilled and every
year there was a struggle between the employers and the trade unions which wanted
to upgrade some jobs to higher categories. It was both necessary and possible to
establish a classification which would be more stable because in a great number of
cases, though not in all, it was possible to estimate the value of a function. This was
only possible through applying a system of evaluation although not necessarily a
system which would be considered as scientific. Before the war, there was a tendency
to level out wages, whereas after the war skill differentials were introduced and consequently the social climate had greatly improved.
Job evaluation systems were useful only if the man in the street and the employer
were able to trust them. In the Netherlands job evaluation systems played an important part in collective bargaining and they were introduced by bipartite bodies on
which both employers and workers were represented (" Vakraden "). Mobility of
labour, full employment and training within industry could be dealt with in a proper
way on the basis of a job analysis which should be followed by job evaluation as a
second step. In the Netherlands, where industrialisation was proceeding very fast,
it had been possible to retrain and increase the skill of a great number of unskilled
and semi-skilled workers and also to train skilled workers for another job within a
comparatively very short time. In this process, both job analysis and job evaluation
were of considerable help. It was, however, essential that these useful tools be applied
only through close co-operation between workers and employers. They would play
their part in improving social and economic conditions only if they were a result of
mutual agreement between the two sides of the industry. In such circumstances, job
evaluation systems would lead to promoting better industrial relations, better appreciation of the individual capacities of the worker, greater output, and industrial peace.
37. The French Government member emphasised that the problem was
very complex and that in analysing it a distinction should be made between
three essential factors. The first factor, which was technical and thus represented the most objective and scientific part of the problem, concerned classification of jobs and functions on the basis of systems or criteria which might
differ to a very large extent. Discussions might be held as to the classification
of jobs so as to satisfy all the parties concerned. The second factor was completely different because it concerned the relation between the number of points
allocated to different jobs and the corresponding wage. As this matter con-

124

JOB EVALUATION

cerned the general wage policy, it might lead to discussions very similar to
those held during collective bargaining. The human element represented the
third factor. It was essential to take into account the individual merits of the
workers so that they be graded according to their abilities. This aspect should
be discussed with the supervisory staff of the undertaking. The whole matter
had two different aspects—that of a tool and that of its application. As the
two aspects differed, it was desirable to examine each aspect separately.
38. The Czechoslovak Government member emphasised that in his country
special attention was paid to wage policy. The basic principle was that the
results of the work, i.e. the national income, should be distributed between all
workers. Many years ago, a national list of jobs was established. This classification took account of the degree of education, training, aptitude, effort and
responsibility involved in each job as well as of the difficulties of the work, the
influence of environment and similar factors. Equal pay for equal work was
the basic principle of this classification. Similar lists were established for the
different branches of the national economy as well as for administrative and
managerial work. These lists were kept up to date so as to take account of
new jobs. This work was done by a special committee composed of representatives of workers and management.
Employers'1 Members' Statements
39. The Employers' member from the Federal Republic of Germany said
that there did not exist any uniform system of job evaluation in his country,
although 80 per cent, of the large and medium-sized plants used the Refa
time study system and established piecework rates on this basis. Only about
10 per cent, of these studies had led to the establishment of true job evaluation
schemes based on the allocation of points, introduced in accordance with
collective agreements between the parties concerned. Jn general, the workers
were divided into three categories: unskilled, semi-skilled and highly skilled,
with subdivisions in many cases to take account principally of the nature of the
work and job conditions. However, in Berlin for example, a uniform and
compulsory job evaluation system was applied. A catalogue of the different
wage rate categories had been established, which took account of such factors
as educational qualifications, experience, skill and responsibility. The necessary
preliminary for the introduction of a job evaluation system was lengthy and
careful preparation. It would be useful to reduce the number of systems
applied and it would be interesting to examine how the systems used in the
metal trades could be unified. The experience obtained in the Berlin region
and in other regions was a first step in this desirable direction.
40. The Finnish Employers' member said that the information given in the
Office's report on the subject might give the impression that Finland had
already realised some important progress in the field of job evaluation. This
was introduced very rapidly in the autumn of 1950 when, during a two-month
strike, the Government, employers and workers made an attempt to arrive at
an agreement on wages. The parties were forced to do this on the basis of
job evaluation. After the question had been examined by an arbitration commission the three interested parties accepted the principle. He did not consider
that this was the way to introduce a job evaluation scheme. On the employers'
side a few persons possessed some knowledge of the subject, principally theoretical, but the workers had hardly any idea of it. They had, therefore, strongly
resisted job evaluation schemes at the plant level. The trade unions had done
all they could to prevent such schemes functioning. The resistance of the
workers in the different plants was particularly strong in regard to point
rating schemes. The employers were not sufficiently prepared to be able to

APPENDIX A

125

persuade the workers of the advantages of these schemes for them. A Finnish
collective agreement had provisions concerning the operation of job evaluation.
One of these was that there should be plant committees in which the employers
and workers could put forward their views. For various reasons the workers
had not been very co-operative. Of 130 enterprises covered by this collective
agreement, only about 20 had obtained good results. In the other plants, after a
number of compromises a job evaluation system now existed which could be
considered to be a ranking system. This experience showed that the successful
application of job evaluation required very considerable preparation by both
the employers and workers concerned. It was necessary to establish a certain
number of rules which were applicable to a whole industry.
41. The Swedish Employers' member said that the first information about
job evaluation coming from the United States reached Sweden in 1948. Management started an investigation about the subject which had been carried out with
the knowledge of the trade unions, which did not oppose it. In 1951 a committee
on job evaluation was set up, which was also concerned with foundries. A
year later, it was agreed to start experimenting. As the results were successful
the experiments were extended to other plants. In June 1956 the committee
put forward proposals for the application of job evaluation in the form of
a booklet. Information about the results of this new system was spread
by workers' organisations and this resulted in a lessening of the workers'
opposition.
42. The United States Employers' member said that in the United States
a large percentage of workers were not organised and therefore they could
not be covered by collective agreements. In his company, job evaluation was
started in 1937 by unilateral action of the management. During the past
20 years, only three grievances arising out of the company's job evaluation
scheme required arbitration. A grievance procedure was established but in the
course of the last 15 years there were less than 100 grievances. According to
the speaker's experience, a large percentage of job evaluation installations
resulted in wage increases for the workers.
Workers' Members'' Statements
43. The Canadian Workers' member said that there were a number of job
evaluation systems in Canadian industry and they varied from plant to plant.
A number of factors such as punctuality, physical demand and absenteeism
were used in different plants in a variety of ways. All job evaluation systems
greatly reduced the percentage of workers at the top of the wage scale, leaving
the vast majority in much lower grades. Neither managements nor workers
had much experience with job evaluation, which was not scientific as had been
claimed. Moreover, since jobs were continually changing in every plant job
descriptions also had to be continually revised, thus serving little purpose.
The workers' experience showed that job evaluation was not to their advantage.
It led to many complaints and grievances and a good number of them had to
be referred to arbitration although in some plants it yielded results which were
moderately satisfactory. However, job evaluation could not be used for all
jobs or be applicable to all plants in the various parts of the country. On the
whole, Canadian industry settled conditions of work through collective
bargaining.
44. A Workers' member from the Federal Republic of Germany commented on the statements of the Employers' member from his country concerning the job evaluation scheme in force in the Berlin region (see paragraph 39
above). Up to the present no true negotiations had taken place between the
interested parties regarding the introduction of the idea of job evaluation in

126

JOB EVALUATION

wage agreements. Job evaluation greatly affected the individual status of
workers. It made their wages vary, as they might be assigned to different jobs
in different wage groups. Further, a worker's wage group might also be
changed as a result of the introduction of new technical devices which in fact
was taking place fairly often. Job evaluation also resulted in the freezing of the
wage structure and consequently the workers had to be safeguarded by collective agreements against such dangers due to the introduction of job evaluation.
The usual factors such as skill, responsibility, etc., and their weighting ought
to be the subject of negotiation between the parties concerned. The necessity
for this arose from the fact that other factors in addition influenced what
should be the appropriate wage rate in a particular case. In enterprises where
most work was of a manual nature and especially in repair work where the
individual worker was important, traditional methods of collective agreements
were sufficient. But where automation had been introduced the situation was
different as the degree of responsibility of the workers was increased, often
very considerably. In other cases, such factors as rapidity of comprehension
and initiative were important. The trade unions were not in a position to
recommend job evaluation as a general basis for wage fixing. In the Federal
Republic of Germany, job evaluation was still in a preliminary stage of discussion between the employers and trade unions in order to fit it into the process
of collective bargaining. As regards the situation in West Berlin, the existing
collective agreement was a part of the wage orders established in 1942 by
the Nazis (Lohngruppenkatalog), against the will of the trade unions.

APPENDIX B
INTERNATIONAL COMMUTEE FOR SCIENTIFIC MANAGEMENT
Extracts from the Report of the International Conference on Job Evaluation '
(Geneva, May 1950)
INTRODUCTION

Aim of Conference
The purpose of the conference was to prepare for the panel discussion of
the Sixth Section of the Brussels Congress2, which will be devoted to job
evaluation. It was attended by 67 specialists in scientific management from 13
different countries.
Subject of Discussion
In view of the limited time available, the conference concentrated on the
application of job evaluation to the grading of predominantly manual jobs by
the point system. A considerable number of its findings and conclusions will
also be useful, however, in other fields and spheres of application and for other
job evaluation methods.
Scientific Preparation of Conference
A preliminary inquiry made it possible to establish general agreement on a
fairly large number of matters and to concentrate the discussion on certain
essential points, upon which a common view could be reached or at least the
divergent concepts could be clarified.
When considering these differences of conception it should be borne in
mind that job evaluation is still in the process of development. It therefore
seems preferable to leave free scope for its evolution along the lines laid down
by the Geneva conference until we have enough experience to undertake a
more complete unification of its principles and techniques.
The following sections of this report deal first with the most important
points on which there is already more or less general agreement, and then with
the points still open to debate.
(A) POINTS ON WHICH THERE WAS MORE OR LESS GENERAL AGREEMENT

(I) Purpose of Job Evaluation
In its widest sense the purpose of job evaluation is to establish a systematic
basis for comparing jobs from the point of view of their demands on workers
capable of performing them.
1
Comité national belge de l'organisation scientifique : IXe Congrès international de
l'organisation scientifique, Vol. I: Rapports présentés aux sections (Brussels, 1951), section VI
(original in French).
* The Ninth International Management Congress (Brussels, July 1951).

128

JOB EVALUATION

When applied to job grading its purpose is to establish a scale of nonmonetary values of the jobs in a given sphere. The fact that the values are not
expressed in money units does not mean that they are not economic values and
therefore are relative and subjective.
When applied to the grading of wages, job evaluation relates only to the
" job content " element in the wage. It is therefore only one of the factors in
wage determination.
Its long-term techno-economic aim is the optimum distribution of labour.
If judiciously applied, it also has a social purpose, since it can contribute
to industrial peace by eliminating wage inequalities which are unrelated to the
abilities of the worker (these inequalities being revealed by job analysis) and
which are sources of tension and conflict.
(2) Characteristics of Job Evaluation
(a) Accuracy.
Job evaluation can be regarded as being " scientific " only to the extent
that it treats its problems in " the spirit of science ", that is, in an orderly and
systematic fashion.
At the same time the method of evaluation must be worked out and applied
in a practical and sensible manner. The method must not attempt to be more
accurate than is warranted by the basis of evaluation and the powers of judgment of those using it.
It should aim at a reasonable degree of precision in objectively determining
the facts and comparing the elements on which the ratings are based, using
scientific means of measurement where necessary, but not losing sight of the
fact that the resultant job grading is finally based on value judgments, which
are inevitably subjective and relative.
(b) Intelligibility.
If the workers are to trust it, the system of job evaluation must be easy to
understand. If the worker himself takes part in the rating process (several
forms and degrees of participation can actually be worked out and are in use
in several countries) this may help to establish confidence.
(c) Flexibility.
Any method of job evaluation must be flexible enough to be adapted to the
special features of the field in which it is applied. The discussion snowed that
advanced methods fulfil this condition.
(3) Stages in Job Evaluation
Determination of the non-monetary value of a job involves
— job analysis and description;
— job specification (in descriptive, graphic or codified form);
— numerical weighting.
Job Analysis and Description.
(a) By a thorough analysis it is possible to attack the whole complex of
problems relating to human labour.
(b) For the purpose of job rating, however, job descriptions normally
deal only with those elements which are considered relevant in differentiating
jobs according to their respective demands.
(c) For psychological reasons it is nevertheless advisable to include
certain elements (including matters of secondary importance) proposed by the
workers.

APPENDIX B

129

(d) The description must be as objective as possible and easily understood
by the worker, who should find it a clearly recognisable picture of his job.
It must therefore be expressed in workshop language; the use of vague, abstract
or psychological terms (with which the worker is generally unfamiliar) or of
merely qualitative terms should be avoided; and in order to avoid pre-judgment
of the value of the job as far as possible, the description should relate solely
to the material means used and the operational components of the job, which
will be recorded on the spot by the analyst and checked by the personal observation of the rater.
(e) The amount of detail in the description depends on the intended scope
of the evaluation. The description of a job to be evaluated on a nation-wide
level will necessarily be in more general terms than one to be evaluated for a
particular undertaking. An over-detailed description leads to demands for
rating review whenever changes occur in matters of secondary importance.
On the other hand, an over-general description allows the rater too much
latitude in assessing job requirements and leads to arbitrariness.
(f) It is generally a good thing to group the items of the description in
the same way as is done in the evaluation table.
Evaluation Table.
This lists the rating factors in the method adopted, and for each factor the
field to which it applies and its subdivision.
(a) Regardless of its application, the evaluation table is always a " translation " of the job description into terms of human labour requirements.
(b) Regardless of the character and number of factors, these should be
— capable of clear definition;
— easily understood by the worker;
— present in all the jobs in the field covered;
— capable of satisfactory evaluation;
— independent of each other as far as possible so as to avoid overlapping
evaluations.
(c) Despite the fact that existing evaluation tables vary greatly as regards
the nature, definition and number of factors, the conference considered that
the factors in them could be grouped under the four following headings :
— mental abilities;
— physical abilities;
— character qualities;
— working conditions (environment).
(d) A model evaluation table grouping the factors according to this principle was put forward by Messrs. Bramesfeld, Lorenz and Dor and was accepted
by the majority as a suitable basis for comparing methods.
Distinction between Internal Factors and Factors Extraneous to Job.
In view of the fact that job evaluation has to establish a hierarchy of the
" job control " element of the jobs and attempt to create a basis of reference
reasonably independent of temporary' factors of a personal, political or economic character, the great majority of those present felt that such a distinction
should be made.
As it is the job which must be evaluated and not the worker holding it,
functional factors must be distinguished from individual (or merit) factors.
If the job grading is to be stable, a distinction must also be made between
functional factors and compensatory factors, the latter being restricted to the
grading of wages.

130

JOB EVALUATION

The conference was almost unanimous on the desirability of classing
environmental factors (where not inherent in the job) and compensatory
factors under the " extraneous " heading.
Scope of the Factors.
This is governed in principle by the statistical distribution of human
abilities over a given population
Examples of Each Degree.
In view of the difficulty of avoiding any possibility of misinterpretation of
the definitions, it is essential that each degree-step should be illustrated by an
adequate number of typical examples, which will serve as benchmarks for the
rater.
(4) Checking the Effectiveness of the Evaluation Method
When the evaluation method is applied to the grading of wages, its success
can be judged by such results as a fall in the number of disputes and wage
claims, greater regularity and stability of personnel and (other things being
equal) increased productivity of labour. Its effectiveness will therefore be
measured in the social as well as in the economic field.
It is necessary, however, to be able to measure the homogeneity of the
results. In this connection the conference accepted two definitions proposed
by the statisticians which, if applied, make it possible to determine the intrinsic
qualities of the method as well as the abilities of the rating personnel.
(a) A method is valid in so far as its results can be used as the basis for a
forecast.
(b) A method is reliable if repeated application gives coherent results
(i.e. the dispersion remains within limits which are regarded as acceptable in
practice).
(5) International Experimental Investigation
The conference recognised the desirability of international comparison
of the results of applying the evaluation methods used in the participating
countries. The rapporteur-général was therefore requested to select 20 predominantly manual jobs adequately represented in all the countries and to send
descriptions of them to the rapporteur for each country. The jobs will then be
evaluated in each country by the rapporteurs, by one or more methods in use.
The results will then be sent to Professor Matthew, University of Birmingham,
who has agreed to carry out the statistical analysis.
(B) POINTS ON WHICH IMPORTANT DIFFERENCES OF OPINION EXIST

(1) Extent to Which Evaluation Framework Should Be Generally Applicable
Although this question was not part of the general subject of the conference,
which was the application of job evaluation to job grading, it was felt desirable
that it should be discussed, as the inquiry had revealed fundamental differences
of opinion on this subject.
Should the evaluation table be planned on general lines so as to be applicable
to all human labour problems or should it be restricted to the purpose of job
grading?
The conference decided for the second alternative, the first being upheld
only by the Belgian delegation and certain French specialists in industrial

APPENDIX B

131

psychology. Lacking an adequate concrete basis of comparison, however, the
conference was unable to judge the practicability of an evaluation table capable
of general application '.
(2) Purpose of Evaluation
Should the elements of a job be evaluated or should one evaluate the job
directly?
Some, but not many, methods evaluate the elements separately and then
deduce the evaluation of the job itself. The discussion showed that the matter
is one of convenience rather than of principle; it is not always possible in
practice to assign a constant composition to each job; there are therefore
cases where the unit to be evaluated is the element rather than the job.
But one must not lose sight of the fact that the job is more than the sum
total of the elements and includes the interdependent effect of the elements on
the worker, which is the determining factor as regards the ability and effort
required of him.
(3) Weighting of Evaluation Factors
The preliminary inquiry had shown that different weighting principles
were in use, namely—
(a) a weighting system deduced from the existing wage structure by
— correlation analysis;
— positioning the rating scales according to the actual wages for key posts for
which the wage is considered by the interested parties to be reasonable
in relation to the duties of the post;
(b) weighting on the principle of relative scarcity of the required abilities
in a sample population
— without regard to the effect of supply and demand;
— allowing for the effect of supply and demand.
After full discussion, the majority of those at the conference favoured the
principle of weighting based in one way or another on the existing wage
structure, at least initially and if only in view of the fact that job evaluation is
not normally acceptable to the workers if it results in major changes in the
existing hierarchy of wage rates.
In theory, where the existing wage structure results from the free play of
supply and demand... it reflects the scarcity factor and the relative appeal
to each worker of the jobs for which he is fitted.
In practice, these conditions are only very imperfectly fulfilled and this fact
seems to justify a search for an " objective " weighting principle capable of
ensuring a long-term optimum distribution of labour.
If the principle mentioned under (b) is applied, the objective and theoretical
relation between the grading and weighting of a factor can be depicted as a
Galton curve which for practical purposes can be replaced, between the
marginal points of the field of application, by a straight line.
The actual economic value of a particular ability is not, however, necessarily
in linear proportion to its degree.
If the second principle under (b) is applied, unless one fixes a priori the
optimum relation between the trends of supply and demand (e.g. at what the
planner of a job evaluation system " considers the best in the higher interests
of the economy and society ") the relation can be derived only from statistical
analysis of the existing wage structure.
It therefore looks as if the weighting methods described under (a) and (b)
are related in practice even though different in theory.

132

JOB EVALUATION

( O IMPORTANT POINTS ON THE PROGRAMME WHICH WERE NOT FULLY
DISCUSSED THROUGH LACK OF TIME

1. Job evaluation rates the job and not the worker doing the job. It must
nevertheless take account of the demands which the job makes on a suitable
worker.
Should it consider the marginal worker (i.e. one who is still able and
willing to do the job) or should it consider the average worker?
2. Is the relative weighting of factors already a part of wage policy?
It is conceivable that a long-term wage policy may try to keep enough
workers in certain occupations which are particularly important for the quality
of the national economy by weighting certain factors heavily. Relative weighting deduced from the existing wage structure will only reflect this policy if it
corresponds to an adequately balanced distribution of the labour force in
relation to the aim envisaged.

APPENDIX C
SOME MAJOR FEATURES OF JOB EVALUATION SCHEMES
APPLIED IN VARIOUS COUNTRIES
AUSTRALIA

In Australia, the conditions of employment of workers in the metal trades
are governed by the Commonwealth Court of Conciliation and Arbitration
Metal Trades Award. With a few exceptions, this award is taken as the basis
for similar state awards.
The rates for the different jobs which are laid down in the awards are
minimum rates. Any amounts paid above these rates are almost invariably
based on a shop or establishment allowance (i.e. a flat amount above the award
rate). It is, however, the official policy of the Australian Metal Industries
Association to follow award rates and not to depart from the structure of
payments provided in the various awards. 1
In determining the minimum rates for the different job classifications, the
Commonwealth Court a chose certain key occupations in the metal trades and,
after hearing evidence and using its own knowledge of the trades acquired over
many years of inspection and hearings, fixed an amount for these occupations
and then laid down a formula for classifying all other occupations in the metal
trades on a comparative basis.
An over-all attempt at evaluating jobs in the metal trades by the Arbitration
Court may be said to have begun with a decision of J. Beeby of 18 December
1930. This judgment consolidated previously separate and independent decisions setting rates in the metal trades. 3
An award of Commissioner J. M. Galvin of 16 January 1952 contained a
more complete discussion of the basis for setting wages in the metal trades.
Skill and responsibility for working without supervision appeared to be the
main factors used by the Commissioner for job evaluation purposes. 4
With respect to skill, in some cases the amount of skill required for a job
was estimated by inspecting the actual work, particularly if the skill required
in the job could be compared by some objective method with that required for
other jobs. With respect to spray painters, for example, a margin of four
shillings per week was allowed between work involving one coat and work on
both prime and finishing coats. 6 In a large number of jobs such as welding,
coremaking and metal polishing, however, the Commissioner appeared to
consider that skill would depend to a great extent on the amount of time
required to learn the job. It was held in a few cases that technological advances
had reduced the skill required for a job and that the margin paid for skill
ought therefore to be reduced. For example, the margin payable to sectional
1

Information supplied by the I.L.O. Correspondent in Australia, 31 July 1956.
Now the Commonwealth Commission of Conciliation and Arbitration.
:
Commonwealth Arbitration Reports (Sydney, Melbourne and Brisbane; the Law Book
Company of Australasia, Pty., Ltd.), Vol. 28, pp. 923-1225.
' Ibid., Vol. 73, pp. 325-477.
• Ibid., p. 384.
J

134

JOB EVALUATION

welders (welders trained to d o only one type of welding) was reduced although
the higher rate was maintained for welders with more general training.
Responsibility of the worker for carrying out his j o b without supervision
appeared to be the second main factor considered by the Arbitration Court
in determining rates. Commissioner Galvin held, for example, that a separate
and higher classification should be set up for " a heat treater not subject to
plant metallurgical supervision " because it is beyond doubt that " he carries a
greater responsibility than the heat treater working in a plant where the metallurgical department determines his methods for h i m " . 1 A similar argument was
applied in raising the margin for the j o b of " rigger and splicer " because it
necessarily involved work without supervision.*
BELGIUM

In Belgium, a n elaborate job classification system which is described in a
series o f three m a n u a l s 3 has been worked out by an official tripartite body, the
General Technical Commission, which was set up by the Minister of Labour
and Social Security in 1945 4 and finished its work in M a y 1951.
TABLE A.

BELGIUM: FACTORS USED IN THE JOB CLASSIFICATION SCHEMB
DRAWN UP BY THE GENERAL TECHNICAL COMMISSION
Factor
Maximum
Factor
Maximum
points
points
I. Vocational training and knowC. Muscular:
ledge:
15. Muscular effort . . 250
16. Position
100
A. Theoretical training and
IH. Intellectual and mental qualities :
knowledge:
1. School training . . 500
17. Concentration . . . 300
2. Supplementary train18. Judgment
200
ing
500
19. Memory
200
20. Inventiveness . . . 200
B. Practical training:
21. Aesthetic taste . . . 200
3. Formal apprenticeship and other training 388
IV. Character and conduct :
4. Informal apprenticeA. Character:
ship and other train22. Authority of coming
500
mand
150
23. Moral authority . . 150
IT. Physical qualities :
24. Presence of mind . . 150
25. Orderliness and cleanA. Sensory:
liness
150
5. Sight
100
26. Appearance . . . .
150
6. Touch
100
B. Conduct:
7. Hearing
100
27. Towards others . . 400
8. Smell
100
28. Towards the enter9. Taste
100
prise
400
B. Motor:
29. Measurable
conse10. Speed of movements 150
quences of conduct . 400
11. Precision of moveV. Job conditions :
ments
150
30. Influence
of
sur12. Co-ordination
of
roundings
300
movements . . . .
150
31. Risks of accidents . 300
13. Speed of reaction
150
32. Risks of illness . . .
300
14. Agility
100
1

Commonwealth Arbitration Reports, op. cit., p. 370.
•Ibid., p. 387.
* Ministère du Travail et de la Prévoyance sociale: Méthode de qualification du travail
de la Commission technique générale (Brussels, 1956).
• Decree of the Regent, 28 Feb. 1947 (Moniteur belge of 18 May 1947). The Commission
consisted of representatives of the Government and employers' and workers' organisations
as well as technical advisers.

135

APPENDIX C

Thirty-two factors are used. For each factor there are two classifications
indicating on the one hand the degree of the particular quality required or
job conditions involved (degree of effort, risk, etc.) and on the other hand the
extent of utilisation of the quality. 1 The factors and maximum points are given
in table A.
With regard to wages, the Belgian job evaluation scheme does not imply
the establishment of a new and inelastic system of wage determination. The
main objective of this method is to establish the relative values of various jobs
according to their job content with a view to determining adequate wage
differentials. For example, if job A is allotted 2,000 points and job B 4,000 points
this does not necessarily mean that the wages of the worker performing job B
will be 100 per cent, higher than those of the worker performing job A. The
number of points arrived at by the job evaluation scheme is only one factor,
though an important one, in wage determination. Other factors which affect
the wage level, such as the minimum living wage, profit sharing, productivity
increases, seniority and personal merit, are taken into account.
FRANCE

All the jobs in the Télémécanique électrique company plant at Nanterre,
from the highest to the lowest, have been evaluated according to job evaluation
principles since February 1948. Up to 1955 two separate job evaluation schemes
were used for jobs of a predominantly manual and intellectual character
respectively. Since then a much more elaborate scheme has been used for all
jobs, as it was found by the company that the two original schemes were not
precise enough in some respects. These are, however, still used for checking
purposes. 2
Table B lists the factors and points used for manual jobs under the original
scheme.
TABLE B. FRANCE: FACTORS AND POINTS UNDER THE JOB EVALUATION
SCHEME FOR MANUAL WORKERS AT THE TÉLÉMÉCANIQUE ÉLECTRIQUE PLANT,
NANTERRE
Point scores for each degree
Factor

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

Training
Practical experience
Adaptation
Complexity of the work . . . .
Physical effort
Mental or visual concentration .
Responsibility for tools and
finished products
Responsibility for materials and
work in progress
Responsibility for the security of
others
Responsibility for the work of
others
Conditions of work
Risks in performing the task . .

11
11
10
7
7
7

14
14
12
12
10
10

17
17
14
17
13
13

22
22
17
22
16
16

28
28
20
28
20
20

5
7
5

7
10

15
16
10

20
20
12

See above, p. 38.
Information supplied to the I.L.O. by the company.

13
8

36
23
34

136

JOB EVALUATION

The factors used under the present scheme are the following:
I. Factors common to all jobs :
1. Training.
2. Adaptation.
3. Difficulties of the job.
4. Physical effort.
5. Mental tension.
6. Job risks.
7. Discretion.
8. Human contacts within the company.
9. Human contacts outside the company.
II. Factors special to jobs predominantly manual :
10. Responsibility for tools.
11. Responsibility for products.
12. Responsibility for the safety of others.
III. Factors special to jobs predominantly intellectual :
13. Responsibility for mistakes.
14. Responsibility for judgment.
IV. Supplementary factors for management jobs :
15. Responsibility of authority.
16. Human responsibility.
V. Factors relating to environmental conditions:
17. Factors of environment.
Under the present scheme jobs are evaluated by allotting points to the
appropriate factors according to detailed instructions which are contained in
a manual. For example, table C shows the points which are allotted for
different degrees of responsibility for the safety of others.
TABLE C. FRANCE: POINTS FOR RESPONSIBILITY FOR THE SAFETY OF OTHERS
UNDER THE JOB EVALUATION SCHEME FOR ALL EMPLOYEES AT THE
TÉLÉMÉCANIQUE ÉLECTRIQUE PLANT, NANTERRE
Duration of
exposure to the risk
Nature of the risk

1. Accident to others improbable.
Work done either in isolation or without machinery
and with light equipment only.
2. Work involving a risk only of slight accidents to
others; avoidance of this risk demands only a
reasonable modicum of attention.
3. Inattention during work on a machine or during
the execution of an order could result in more
serious accidents to others. Examples: foot or
finger crushed, accident to the eyes.
4. Sustained attention necessary to avoid serious
accidents during the performance of the job. However, others can avoid being hurt by paying attention.
5. The safety of others depends exclusively on care
in carrying out the job and inattention in the slightest degree could have fatal consequences.

Less than
30 per cent.
of the time

More than
30 per cent.
of the time

5
6

137

APPENDIX C
GERMANY (DEMOCRATIC REPUBLIC)

As was mentioned in Chapter III, it is considered desirable in the German
Democratic Republic to apply a uniform system of job evaluation in all
industries. One suggested scheme 1 is described below.
The most important factor under this scheme is knowledge and experience,
for which degrees are distinguished in table D , which also gives the number of
points corresponding to each degree.
TABLE D. GERMAN DEMOCRATIC REPUBLIC: SUGGESTED POINT VALUES
FOR DIFFERENT DEGREES OF KNOWLEDGE AND EXPERIENCE
Point
Point
Degree of knowledge
Degree of knowledge
value
or experience
or experience
value
0
None
More than 12 months' in-plant
Up to three weeks' instruction . .
training
8
1
Up to three months' in-plant trainPrevious specialised training reing
2
quired
10
Up to three months' in-plant trainAdvanced specialised training reing plus considerable experience 3
quired
12
Very advanced previous specialised
More than three months' in-plant
training required
15
training
4
Six to 12 months' in-plant training 6
Table E gives the various factors and corresponding point values for the
scheme as a whole.
TABLE E. GERMAN DEMOCRATIC REPUBLIC: SUGGESTED SCHEME
OF FACTORS, DEGREES AND POINT VALUES
Point scores for each degree
Factor

1.
2.
3.
4.
5.

Knowledge and experience
Mental effort
Physical effort
Responsibility for others .
Responsibility for equipment
6. Job conditions:
(a) Temperature . . . .
(b) Water, oil, grease,
dust, dirt, etc. . .
(c) Gas, acid, poison, etc.
(d) Noise, glare, etc. . .
(e) Risk of accident . .

0
0
0
0

1
1
1
1

2
2
2
2

3
3
3
3

4
4
4
4

0
0
0
0

1
1
1
1

2
2
2
2

3
3
3

GERMANY (FEDERAL REPUBLIC)

Tables F and G illustrate a scheme established by agreement between the
management of the Volkswagen works and the West German metalworkers'
union. The criteria used in allotting points to the various factors are set out in
table F ; the point ranges corresponding to each factor are recapitulated in
table G.
1
See E. HAMPICKE: Grundfragen der qualitativen Arbeitsbewertung (Berlin, Verlag Die
Wirtschaft, 1957), Appendix IV.

TABLE F. FEDERAL REPUBLIC OF GERMANY: CRITERIA FOR THE ALLOTMENT OF POINTS TO THE DIFFERENT FACTORS
UNDER THE VOLKSWAGEN SCHEME
Number
of points

Skill,
dexterity

Bodily
strain

Mental
strain

Nervous
strain

Responsibility

Temperature

Oil,
grease, dirt,
dampness

Dust

Accident
hazards

Gas, noise,
cold, darkness, etc.

none

very little

0.5

small

occasionally light
strain

little

sometimes
a little

little

medium

little to
medium

1.0

light
occasionally medium strain

medium

little

sometimes
medium

high

medium

1.5

medium

occasionoccasionally medium ally high

occasionally high

sometimes
medium

medium

very high

sometimes
high

2.0

sometimes
high

medium
strain

high

medium

sometimes
high

—

high

3.0

very high

heavy

very high

high

very high

—

very high

4.0

highest

very high

—

sometimes
very high

—

5.0

—

highest

—

very high

—

6.0

—

highest

—

U)
00

o
55

139

APPENDIX C
TABLE G. FEDERAL REPUBLIC OF GERMANY: FACTORS AND RANGE
OF POINTS UNDER THE VOLKSWAGEN SCHEME
Factor

A.
B.
C.
D.
E.
F.
G.

Range of points

Training, experience
Skill, dexterity
Bodily strain .
Mental strain .
Nervous strain
Responsibility .
Job conditions:
(a) Temperature

Factor

0-7
0-4
0-5
0-3
0-3
0-6

Range of points

(b) Oil, grease, dirt,
dampness
(c) Dust
(d) Accident hazards .
(e) Gas, noise, cold,
darkness, etc. . . .

0-3.5

H. Unsupervised work

0-1.75
0-3.5
0-3.5
0-3.5

. .

The relation between the points for the different job grades and wages is
indicated by table H.
TABLE H. FEDERAL REPUBLIC OF GERMANY: RELATION BETWEEN JOB
EVALUATION POINTS AND WAGES AT THE VOLKSWAGEN PLANT
Points

21.1 and above
18.1-21
15.1-18
12.1-15
9.1-12
6.1- 9
3.1- 6
0-3

. . . .

Wage group

Wage as percentage
of the average wage
(average» 100)

8
7
6
5
4
3
2
1

131
117
107
100
95
91
87
83

Reference has already been made in the body of the study to the EulerStevens method of job evaluation 1 : the 16 factors used and the maximum
numbers of points allotted to them under this method are given in table I.
TABLE I. FEDERAL REPUBLIC OF GERMANY: FACTORS USED
AND MAXIMUM POINT SCORE ALLOTTED TO EACH UNDER THE
EULER-STEVENS METHOD
Factor

Maximum
point score

Knowledge
Skill
Responsibility for equipment and
quality of production . . . .
Responsibility for the safety of
others
Responsibility for the progress of
work
Muscular effort
Attention

7.0
4.0
3.0
3.0
3.0
8.0
2.0

Factor

Maximum
point score

Mental effort
Temperature
Water, acids
Dirt, dust
Gas, fumes, humidity
Noise, vibration
Glare, darkness
Risk of catching cold, work out
of doors
Risk of accidents

4.0
3.0
2.5
3.0
3.0
2.5
1.5
1.5
2.0

Thus, out of a total of 53 points 19 at the most can be scored for job conditions, 9 for responsibility, 11 for skill and knowledge, and 14 for various
kinds of effort. In allotting point values, both the nature or level of the requirements and the relative duration of the conditions are taken into account.
1

See above, Chapter II.

140

JOB EVALUATION
INDIA

According to information supplied by the Indian National Trade Union
Congress a job evaluation scheme is applied in some Indian plants.1 The
factors and points used in the scheme and the points which are allotted to
various jobs are set out in tables J and K.
TABLE J. INDIA: FACTORS AND POINTS UNDER A JOB EVALUATION SCHEME
USED IN SEVERAL PLANTS
Factor
(a)

(bj
(bj
(c)
(d)
(e)

(f)

Points

Skill requiring manual dexterity .
Physical effort
Mental effort
Intelligence
Knowledge of precision instruments
Responsibility
Consistency

30-75
5-30
5-30
10-30
5-30
5-20
5-10

TABLE K. INDIA: POINTS FOR VARIOUS JOBS UNDER A JOB EVALUATION
SCHEME USED IN SEVERAL PLANTS
Points for each factor
Job
(a)

Armature winder, simple
jobs
Armature winder, more
difficult work . . . .
Turner " C "
Turner " A "
Grinder " C "
Grinder " A "
Grinder, special grade .
Miller " C "
Miller " A "

(bj

(b,)

(d)

<*) (f)

Total
number
of points

75
40
75
30
50
60
40
65

5
5
10
10
10
10
10
10

30
5
10
5
10
20
10
25

30
10
30
10
20
25
10
30

10
5
25
5
20
30
10
30

20
5
15
5
20
20
5
15

10
5
10
5
10
10
5
10

180
75
175
70
140
175
90
185

NETHERLANDS

The Netherlands standardised method2 distinguishes ten factors, one
of which (job conditions) is divided into five sub-factors. More or less detailed
definitions of the factors and sub-factors have been published by the Netherlands National Standards Commission.3 Points are allotted to each of the
14 factors and sub-factors on the basis of a further subdivision according to (a)
the degree in which a factor or sub-factor is called for or has to be endured,
and (b) the specific nature of the factor as it occurs in a job. Thus, for the
" heaviness " sub-factor there are four degrees (measured in kilos), but also
three classes of frequency and duration. In this respect the method is similar
to the Euler-Stevens and Belgian methods referred to above. As regards the
1

Communication to the I,L.O., 21 Aug. 1956.
• It may be recalled that this method .is applied in many different industries. See Chapter II above.
* Hoofdcommissie voor de Normalisatie in Nederland: Norm V 3000 (Delft, 1952).

141

APPENDIX C

" knowledge " factor, six degrees are distinguished but at the same time the
factor itself is subdivided into knowledge of isolated facts and of mutually
dependent questions, the latter being further subdivided into simple, less simple
and involved relationships. Through these subdivisions the total number of
factors becomes rather larger than the 14 in respect of which points are allotted
separately. The minimum number of points which may be allotted to any
factor or sub-factor is zero; the maximum generally is eight, although according
to the official description of the method " one may, however, go as high as, for
example, ten points in exceptional cases ". 1 Thus the method is more flexible
than other systems in which the number of maximum possible points is more
or less rigidly fixed.
The Netherlands system is " weighted ", the points allotted for each factor
being multiplied by a coefficient. The factors, their maximum number of
points and their weights are shown in table L.
TABLE L.

NETHERLANDS: FACTORS, POINT VALUES AND WEIGHTS
UNDER THE STANDARDISED METHOD
Normal
maximum
number
of points
per factor *

Factor

Knowledge
Independence3
Contact with others
Authority exercised
Articulateness *
Dexterity
Adaptability with respect to materials
and equipment
Job conditions:
Heavy lifting
Fatiguing positions
Burdensome attention
Environmental factors
Personal risk
Unusual gifts
Responsibility (risk of causing damage)

Weight •

Normal total
maximum
points

40
42
14
14
7
16

8
8
8
32

1
Hoofdcommissie voor de Normalisatie, loc. cit. As indicated above, the normal maximum may
be exceeded in exceptional cases. ' Stichting van de Arbeid: Werkclassificatìe ah kuìpmiddel bij de
Loonpolitiek (The Hague, 19S2), p. 14. * This factor, unusual in other systems, is denned as the nature
and degree of detail of the instructions under which the job is carried out. * Ability to express more
or less involved ideas either orally or in writing with more or less time for thought. * No specific point
scores are assigned to this factor.

SWEDEN

At the beginning of 1951, the three parties to the collective agreement for
the Swedish metal trades (the metal trades employers' association, the metalworkers' union and the foundry workers' union) set up a committee to draft
proposals for a job evaluation scheme. About a year later such a scheme was
ready and was introduced on an experimental basis in a plant employing about
200 workers. On the basis of this experiment and of others carried out inde1

Hoofdcommissie voor de Normalisatie, op. cit., p. 57.

142

JOB EVALUATION

pendently in two large Swedish engineering firms, another committee which
was set up was able in June 1956 to draw up a final scheme and to publish a
booklet about it for the guidance of managements and workers. This scheme
has now been adopted by both of the firms referred to above.1 Table M shows
the factors and points used under this scheme.
TABLE M. SWEDEN: FACTORS AND POINTS UNDER THE
METAL TRADES SCHEME
Grade
Factor
1

Skill:
1. Theoretical
2. Practical
3. Judgment and initiative
4. Dexterity and precision

0
0-1
1
0-1

1
2-4
2-4
2-4

2-3
5-7
5-7
5-7

4-6
8-11
8-11
8-11

7
12
12
12

Responsibility :
5. Economic
6. For others' safety . . .
7. For directing work . .

0-1
0
0-1

2-4
1
2-4

5-7
2-3
5-7

8-11
4-6
8-11

12
7
12

Effort :
8. Physical
9. Mental

0-1
0

2-4
1

5-8
2

9-13 14-18
4-5
3

0-1
0

2-4
1

5-8
2-3

Working conditions :
10. Workplace
11. Risk of injury

. . . .

Maximum
point
score

Percentage
of
total

35.2

25.4

} "

18.9

9-13 14-18
7
4-6
) *

20.5

Total points . . .

122

100.0

Criteria are laid down which are to be used in allotting points to the
various factors in the actual process of evaluating jobs. For example, in the
case of factor No. 8 (physical effort) a job will be placed in a particular grade
(and be allotted points) in accordance with the degree of strain involved, as
illustrated by table N.
TABLE N. SWEDEN: CLASSIFICATION OF JOBS BY DEGREE
OF PHYSICAL EFFORT UNDER THE METAL TRADES SCHEME
Degree of effort

Job grade

Insignificant
Little or moderate; some handling of lighter objects
Moderate strain
Great strain, as from carrying heavy objects or working in an uncomfortable
position
Very great strain

1
2
3
4
5

> Communication to the I.L.O. from the Swedish metal trades employers' association,
17 Sep. 1956.

143

APPENDIX C

Thus, a job involving little or moderate strain would be placed in grade 2
and would receive 2 to 4 points.
UNITED KINGDOM

So far as is known, about 20 companies in the metal trades in the United
Kingdom use job evaluation schemes. Two of the best-known schemes, which
have been widely applied, are those used respectively by a well-known electrical
engineering firm and by Imperial Chemical Industries, Limited, in its metalworking and other plants. Table O shows the factors and points used under the
former, which was based on a study of eight key jobs in the industry and at the
company's plants. Points were allotted to these jobs on the basis of standard
job descriptions. The weights allotted to the different factors are intended to
reflect supply of and demand for particular qualities in a worker. The scheme
is applied to all manual jobs and covers about 2,000 workers,
TABLE O. UNITED KINGDOM: FACTORS AND POINTS UNDER
A JOB EVALUATION SCHEME USED BY A LARGE ELECTRICAL
ENGINEERING COMPANY
Points allotted according to degree
Factor

1. Skill, training and experience
(a) Learning period * a. .
(b) Previous experience .
(c) Mechanical ability *
(d) Complexity of process *
(e) Dexterity and motor accuracy *
2. Responsibility :
(a) Material or equipment . . .
(b) Effect upon subsequent operations
3. Mental requirements:
(a) Teamwork
(b) Alertness to details
(c) Attention to orders, drawings,
etc
4. Physical requirements:
(a) Monotony
(b) Abnormal position
(c) Abnormally heavy
(d) Disagreeableness
5. Hazards:
(a) Disease
(b) Accident
(c) Internal injury

Very low
to medium

High

Very high
to maximum

0-6
0-14
0-11
0-5
0-7

7-10
15-22
12-18
6-9
8-11

11-12
23-28
19-23
10-11
12-14

0-8

9-13

14-16

0-4

5-6

7-8

0-1
0-2

2
3

3
4

0-6

7-10

11-12

0-3
0-3
0-5
0-4

4
4
6-8
5-6

5
5
9-10
7-8

0-3
0-5
0-3

4-5
6-8
4-5

6
9-10
6

•The points are allotted as follows: 0-6 points for 0-6 weeks; 7-10 points for 7-13 weeks; and
11-12 points for 14 weeks. 'The points are allotted as follows: 0-14 points for 0-1 year; 15-22 points
for 2-4 years; and 23-28 points for 5-7 years. * " Mechanical ability " is the ability of a man to understand mechanical processes. - «^onipiexity ot process ** is concerned with the number ot different
elements in a job which have to be acquired by the operator before the job can be done satisfactorily.
The more the elements vary and the more complex they are, the higher the number of points that will
be awarded. B M Dexterity " is interpreted as meaning deftness or quickness of fingers and the
co-ordination of muscular control. Included in this is the co-ordination of hands, eyes and feet.

144

JOB EVALUATION

An example of a job rating card used under the scheme is given below.
Job Rating Card Used by an Electrical Engineering Company
in the United Kingdom

Job Name

Ç"™!?..??:...1.:

Alt. title

Des. of machine and equip.

Wage group No.

(1) SKILL, TRAINING AND EXPERIENCE

(2) RESPONSIBILITY

2
19
6
7
9

A. Learning period.
B. Previous experience.
C. Mechanical ability.
D. Complexity of processes.
E. Dexterity.

8 A. Material and equipment.
4 B. Effect on subsequent operations.
(3) MENTAL REQUIREMENTS

— A. Teamwork.
3 B. Alertness.
7 C. Attention to orders, etc.
(4) PHYSICAL REQUIREMENTS

(5) HAZARDS

1
1
—
5

4 A. Disease.
2 B. Accident.
— C. Internal injury.

A. Monotony.
B. Abnormal position.
C. Abnormally heavy.
D. Disagreeableness.

Sch.
Total points allowed: 78

Paid

Pence per hour:

QUALIFICATIONS NEEDED BY APPLICANT

Age

Weight: Light/Medium/Heavy

Remarks or other qualifications
Description of work
Summary of job
The Imperial Chemical Industries scheme, which is described in a booklet
published by that firm ', assesses jobs under four main headings as shown in
table P.
'Imperial Chemical Industries, Limited: The I.C.I. Method of Job Appraisement for
General Worker Jobs (London, 1948).

145

APPENDIX C
TABLE P.

UNITED KINGDOM: FACTORS AND WEIGHTS UNDER THE
IMPERIAL CHEMICAL INDUSTRIES SCHEME
Factor

Weight

Mental requirements
Physical requirements
Acquired skills and knowledge
Working conditions

0-4
0-4
0-10
0-7

The assessments of jobs are expressed in points which reflect, in direct
proportion, their relative values as compared with the job requirements for the
" basic labourer ".1
The matters which are taken into account in allotting points to each of
the four factors are set out as follows in an explanatory leafleta included
in the booklet:

MENTAL
REQUIREMENTS

Good memory
Ability to reason
Speed of reaction
Even temperament
Co-opera tiveness
Perseverance
Mechanical sense
Initiative
Disparate attention
Ability to visualise

PHYSICAL
REQUIREMENTS

Muscular strength
Stamina
Agility
Sensory accuracy

ACQUIRED
SKILLS AND
KNOWLEDGE

Education
Training
Experience

WORKING
CONDITIONS

Physical :
Vibration
Position
Fumes, smell
Dust, dirt
Heat, cold, changes
Wetness
Clothing and Equipment
Poor light
Exposure
Mental :
Noise
Height
Below ground
Isolation
Monotony
Nervous tension
Accident risk
Disease risk

Sense of responsibility

U N I T E D STATES

Table Q, appearing on the following page, gives a comparison of the
factors and point values or weights used in four major job evaluation schemes
in the United States.

1
A basic labourer is regarded as performing such operations as sand-barrowing, floorsweeping, taking messages, reading a thermometer and recording simple figures.
* Imperial Chemical Industries, Limited, op. cit. (explanatory leaflet, p. 3).

146

JOB EVALUATION

TABLE Q. UNITED STATES: FACTORS AND MAXIMUM POINTS OR WEIGHTS
UNDER THE SOUTHERN CALIFORNIA AIRCRAFT INDUSTRY, GENERAL ELECTRIC,
WESTINGHOUSE AND U.S. STEEL SCHEMES
Southern California
Aircraft Industry '

General Electric •

Westinghousc.

Factors

Maximum
points

Factors

Maximum
points

1. Skill . . .

400

1. Mentality .

100

2. Mentality .

3. Responsibility for
material
and equipment . . .
4. Mental application .
5. Physical
application
6. Job
conditions.
7. Unavoidable
hazards .

100

2. Skill . . .

3. Mental
effort . .

400

Factors

1. Education

Maximum
points

100

2. Experience . .

100

3. Aptitude.

125

Factors

1. Preemployment
training .
2 Employment
training
and experience .
3. Mental
skill . .

Weights

1.0

4.0
3.5

100

4. Physical
effort . .
5. Responsibility . . .

6. Working
conditions.

100

4. Physical
effort . .
5. Mental
effort . .

100

6. Visual
effort . .

7. Responsibility for
safety of
others . .
8. Responsibility for
product .

9. Responsibility for
equipment
10. Job
conditions

Total . . .

U.S. Steel s

790

Total . . .

800

Total . . .

40
40
40

25
25

4. Manual
skill . .
5. Responsibility for
materials.
6. Responsibility for
tools and
equipment
7. Responsibility for
operations . .
8. Responsibility for
safety of
others . .
9. Mental
effort . .

2.0
10.0

4.0

6.5

2.0
2.5

10. Physical
effort . .

2.5

540

11. Surroundings . . .
12. Hazards .

3.0
2.0

•International Association of Machinists: What's Wrong with Job Evaluation (Washington, 1954), p. 23.
•United Electrical, Radio and Machine Workers of America: U.E. Cuide to Wage Payment Plans, Time Study and Job
Evaluation (New York, 1943), pp. 74-75. • Monthly Labor Review (Washington, U.S. Department of Labor, Bureau
of Labor Statistics), Vol. 64, No. 6, June-1947, p. 978.