TIME SERIES ANALYSIS OF STRIKE FREQUENCY*
| DOI | http://doi.org/10.1111/j.1467-8543.1977.tb00073.x |
| Author | John Shorey |
| Date | 01 March 1977 |
| Published date | 01 March 1977 |
British Journal oj’lndustrial Relations
Vol.
XV
No.
1
TIME SERIES ANALYSIS OF STRIKE FREQUENCY*
JOHN SHOREYt
I
INTRODUCTION
SEVERAL time series statistical studies of strike causation have appeared recently
in
the economics literature’ (Pencavel, 1970; Vanderkamp, 1970; Knight, 1972;
Bean and Peel,
1974).
Yet another, therefore, perhaps warrants some justification.
In
the first place the study outlined here is valuable simply as an empirical exer-
cise. It covered a much longer time period than have others of its genre. This both
extended the data base upon which the statistical techniques operate and in-
troduced additional historical periods into the analysis. The study is also valuable
in
that it discussed certain hypotheses that have not been fully explored in the re-
cent literature. The main justification for this paper, however, is that it uses the
new study to evaluate the theories and findings of the various statistical exercises
carried out to date. By concentrating initially
on
purely theoretical, rather than
empirical, explanatory relationships and variables, the analysis provides a logical
framework within which to interpret statistical results, and thereby resolve ap-
parently conflicting pieces of evidence. In particular it facilitates the distinction
between demand side (worker) and supply side (employer) influences upon the
strikes decision.
The hypotheses suggesting a relationship between strike frequency and
economic forces are set out in Section
11.
An
operational model is developed in
Section
I11
and subjected
to
empirical scrutiny in Section IV. Section
V
considers
the theoretical and statistical similarities and dissimilarities between the present
study and previous studies.
I1
A THEORETICAL MODEL
This section develops a model to explain variations through time
in
the
probability, P(S):, that a single bargaining unit,
i,
will experience a strike. It is
assumed that all strikes result from the breakdown of negotiations which directly
or
indirectly are concerned with wages.* Furthermore,
no
distinction is made
between the interests
of
workers and trade unions, nor between the interests of
owners and managers.
The strike probability will, on the demand side, vary positively with the fre-
quency of wage claims and the size
of
the claims workers submit.
On
the supply
side it
will
vary negatively with the size of the offers employers make during
negotiations. It will also vary negatively with the costs of strike action for either
side and positively with the extent of recent external militancy.
The three principal determinants
of
the frequency of negotiations and the size
of workers’ wage claims at any point in time are: recent changes in the amount of
effort required of workers, the extent to which established wage differentials have
recently been disturbed, and the current rate of change of prices.
On
the supply
*
This
study was originally carried out in
1973
as
part
of
the doctoral programme at the London
School
of
Economics.
I
am very grateful
for
the help given, then and since, by David Metcalf, Ray
Richardson, and David Hendry.
t
Department
of
Economics, University College, Cardiff.
63
To continue reading
Request your trial