TRADE UNIONS AND THE POSITION OF WOMEN IN THE INDUSTRIAL WAGE STRUCTURE*
| Published date | 01 July 1977 |
| DOI | http://doi.org/10.1111/j.1467-8543.1977.tb00084.x |
| Date | 01 July 1977 |
| Author | S. J. Nickell |
British
Journal
of
Industrial Relations
Vol.
XV
No.
2
TRADE UNIONS AND THE POSITION
OF
WOMEN IN THE INDUSTRIAL
WAGE STRUCTURE*
S.
J.
N1cKsu.t
I.
INTRODUCTION
A casual glance
at
the average hourly earnings of manual workers in various
manufacturing industries’ reveals a tremendous variation from one industry to
another. Indeed this variation is
so
big that the average wage for women in some
industries is actually higher than that for men in others: in spite of the women’s
average being only about
65
per cent of the men’s (in
1974).
Given this enormous
variation, it seems likely that a detailed analysis of the industrial wages structure
will throw
a
considerable degree of light on many aspects of the labour market.3 In
particular, if we know something of the industrial structure of collective bargain-
ing and trade union coverage, we can examine the impact of unions on the wages of
men and women and also look
at
the relationship between the collective bargaining
structure and other salient aspects of
the
labour market. Following this approach,
the central piece of empirical work reported in this paper consists of
a
number of
wage regressions over
a
large cross-section of minimum list-heading industries in
the Standard Industrial Classification. In addition, we analyse several other rele-
vant variables, notably the incidence
of
sick pay and occupational pension schemes
and the structure of collective bargaining itself.
Before considering the technical details of the empirical model it is worth
commenting in general terms on the place of women in the labour market and on
the notion of discrimination. This we do in the next section.
11.
THE
POSITION
OF
WOMEN
IN
THE
LABOUR
MARKET
Much work has been done on the position of disadvantaged groups in the labour
market and a lot of it has been focused on the extent of discrimination against these
group^.^
Pure discrimination may be said to exist if two individuals who have
identicd
productive characteristics receive different rewards over their lifetime
simply because they belong
to
two different groups. It is clear from this definition
that such ‘pure’ discrimination is exceedingly difficult, if not impossible, to meas-
ure, simply because the relevant productive characteristics of individuals, includ-
ing,
as
they do, things like proneness to quit and reliability, are impossible to
identify. Furthermore, ex-post measurement of such variables
as
the turnover rate
will not provide
a
true guide to
a
characteristic such
as
pronenesg to quit since the
former is determined simultaneously with the structure of rewards.
As
an example
ofthis last point consider the following simple stylised model. There are two readily
identifiable groups of workers with identical productive characteristics. Employers
and bankers, however, think that the members of one group, (A), are more quit-
*
Many thanks are due to Orley Ashenfelter, John Gennard, James Heckman, Richard
Layard, David Metcalf, Charles Mulvey and John Pencavel
for
the ideas, stimulation and
information which they have lavishly provided either in lengthy discussions
of
this and
related topics or
in
comments on previous versions of this paper.
I
am also grateful to Nick
Foster, Louise Hamshire, Stephen King, Paul James, Cathy Pick and Joe Wilcox who at one
time or another provided considerable assistance in the production
of
this study.
The
work for
it
was partially financed by the Nuffield Foundation.
t
Lecturer in Economics, London School
of
Economics.
192
TRADE UNIONS AND WOMEN IN THE WAGE STRUCTURE
193
prone and feckless than members of the other group,
(B).
Because of this, members
of A face higher borrowing rates than members of
B,
since bankers think they are
more likely
to
default on loans. Consequently they have less of an incentive
to
undertake either specific or general training and furthermore will choose occupa-
tions with a flatter lifetime wage structure. Employers will also be less willing
to
finance specific training for group
A.
Later on in their working lives members of
group
A,
having less specific training and being on flatter wage profiles, will, in
fact, quit more readily. This equilibrium is then self-sustaining. Employers and
bankers have their views about members of group A continuously reinforced and
at
first sight there appears to be little discrimination-group
A
do less well because
they
are
less reliable.5 On the basis
of
this argument, it is clear that, even if all the
difference between the wages of two groups of individuals can be explained by
differences in
observed
relevant characteristics, the disadvantaged group may still
have considerable cause for complaint.
Turning now to the particular case of women, the central characteristic that
distinguishes the woman in the labour market from her male counterpart is her
lower level of labour force attachment. This lower level of attachment is not, of
course, the
cause
of her lower rewards in the labour market for it is determined
simultaneously with these rewards. But when she is young the expectation of this
lower level of attachment is of great consequence. It plays an important role in the
socialisation process
of
early childhood which determines self-perception and later
on it influences crucial educational choices.6 Furthermore the existing and
observed lower level of attachment of women colours the attitude of employers to
any potential female employee. The combination of the woman’s own view of
herself and her future, her educational attainments and the employers’ (correct)
view
as
to her probable commitment to the labour force then determine her current
and future level of rewards and her actual degree of labour market participation. It
is clear that these latter will turn out
to
be lower than for that group which is
expected
to
be, socialised
to
be, and is, more committed to participate in the labour
market, namely men.
It is clear from this discussion that the present position of women in the labour
market is the result
of
a complex sequence of social and economic interactions and
that a single measure, such
as
the degree of market discrimination, does not really
capture the essence of the ~ituation.~ In the light of this, our aim in this investiga-
tion is the modest one of analysing the industrial wage structure and related
variables in order to add
as
much
as
possible to the existing picture
of
the relative
positions of men and women at work and, in particular, the role of trade unions in
this regard.
Current knowledge and hypotheses about the labour market status of women
may be briefly summarised by the following list of special characteristics
which either have been or might be (on the basis of received theory) attributed
to the female labour force. Each point is followed by
some
of the supporting refer-
ences.
(a)
Women have
a
lower attachment
to
the labour force than men (Mincer,
1962;
Cain,
1966).
(b) Women receive less training than men (Age earnings profiles from
New
Earnings Survey,
1970,
cited in Chiplin and Sloane,
1974.
Skilled apprentice-
ship statistics,
New Earnings Survey,
1972,
p.
235).
(c)
Women have
a
higher turnover rate than men (D.E.
Gazette,
1973,
p.
654).
(d) Women are absent more often and take more time off than men
New Earn-
(e) Women obtain
a
higher proportion of their earnings from payment by results
ings Survey,
1970,
Table
98).
To continue reading
Request your trial