The Seniority Model of Trade Union Behaviour: A (Partial) Defence

DOIhttp://doi.org/10.1111/j.1467-8543.1989.tb00216.x
AuthorHoward Gospel,Richard Disney
Date01 July 1989
Published date01 July 1989
British
Journal
of
Industrial Relations
2712
July
1989
0007-1080
$3.00
The Seniority Model
of
Trade
Union
Behaviour:
A
(Partial) Defence
Richard Disney and Howard Gospel*
I.
INTRODUCTION
The issue of whether the behaviour
of
a trade union can be adequately
modelled using orthodox economic analysis has long been a source
of
some
controversy both within and between the disciplines
of
economics and
industrial relations.
The debate has been heightened by the recent spate of
publications on the subject in the economics literature.*
In a contribution to this journal, Peter Turnbull
(1988)
adds to this
debate, arguing that current investigations of the behaviour
of
trades unions
by economists are conceptually flawed and
of
limited practical relevance.
The present paper does not intend to comment on all the issues raised by that
article. Indeed, we are sympathetic to many of them. Rather the intention is
to pursue one issue which seems to us
to
be
of
particular interest: the
usefulness
or
otherwise
of
the ‘seniority’ model
of
union behaviour.3 In such
a model, layoffs
or
redundancies within the union sector are determined by a
ranking
of
individual members according to accumulated experience. One
theoretical attraction
of
such
a
model,
in
which individuals have clearly
defined positions in an ordering of ‘job rights’, is that union members can
have heterogeneous preferences. This
is
clearly superior to a model
of
union
behaviour in which all members are required to have identical
preference^,^
and in which
it
is
unrealistically assumed that the union’s objective function
can be derived without an explicit discussion
of
the internal procedures by
which the union establishes a set
of
collective preferences.
Turnbull’s arguments against the seniority model are twofold. First, ‘the
seniority model says little,
if
anything, about actual union behaviour’
(Turnbull,
1988: 104)
and its ‘central fallacy is
.
.
.
the assumed congruence
between what
is
written in a collective agreement and what happens in
practice’
(ibid:
105).
The implication is that, even
if
seniority clauses are
contained in contracts, they are rarely implemented in practice. This
*
Professor
of
Economics and Senior Lecturer in Industrial Relations respectively at the
University
of
Kent at Canterbury.
180
British
Journal
of
Industrial Relations
contention is examined in Section
I1
of
the present paper. The second
criticism is that the seniority model has basic theoretical drawbacks which
so
limit the model that
it
‘can apply only
to
a perfectly homogeneous union
membership performing identical work tasks’
(ibid:
106).
The central point
of this critique is that the account
of
union decision-making procedure
contained in the seniority model is naive and may, indeed, predict that the
union will vote itself out
of
existence. Section
111,
therefore, examines these
arguments and suggests
in
contrast that the voting model underlying the
theory
of
union decision-making
is
robust and that the seniority model does
not necessarily predict a union voting itself out
of
existence.
Section
IV
argues that the seniority model can cast light on other issues
which are
not
considered by its critics. In particular it can serve as one way of
deriving a theory
of
why people join unions. Furthermore,
a
more general
model can be constructed in which the wage chosen by the union and the size
of
the union are determined simultaneously. It can be shown that such a
model is resilient even when the assumption is dropped that seniority rules
are always rigidly adopted in determining the order of layoffs or job losses.
Although we therefore argue that the seniority model is absolved from
some of the criticisms levelled against it, and indeed can be extended
fruitfully to examine other issues, it has limitations. The most important
in
our view
is
in the absence
of
any treatment of the role of the employer in the
development and application of seniority rules. This is revealed by the
somewhat ambivalent role
of
management in the actual development
of
seniority systems (described
in
Section
11)
and an ambiguity in the behaviour
of
the firm
in
the standard economic model described in Section
111.
This
issue
of
management and seniority is therefore considered
in
Section
V.
11.
ARE SENIORITY RULES IMPORTANT
IN
PRACTICE?
At the outset
it
is useful to distinguish two types
of
seniority: benefit
seniority and competitive status seniority. The former refers to the level
of
wages, fringes, holidays, sickness arrangements and other benefits awarded
to workers on the basis of length
of
service. The latter refers to rules which
regulate competition between workers in promotions, layoffs and redun-
dancies. Thus, the former is concerned with the reward
of
labour and the
latter with the allocation
of
labour;
or,
in
industrial relations terms, sub-
stantive rules and procedural rules involving worker selection respectively.
In
what follows, we are primarily concerned with competitive status senior-
ity (although we touch briefly on benefit seniority), and in particular with
Turnbull’s assertion that such seniority rules are relatively unimportant in
practice.
Turnbull focuses
in
the first place
on
the American experience. His
interpretation of the historical origins and contemporary operation
of
seniority rules in the United States may be summarised as follows. Seniority
grew up in the
1930s
and
1940s
out
of
an accommodation between large

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT