THE MEASUREMENT OF SECONDARY LABOUR MARKET EFFECTS ASSOCIATED WITH GOVERNMENT TRAINING1
| Author | JAMES J. HUGHES,IAN BRINKLEY |
| DOI | http://doi.org/10.1111/j.1467-9485.1980.tb00557.x |
| Date | 01 February 1980 |
| Published date | 01 February 1980 |
Scottish
Journal
of
Political
Economy,
Vol.
27,
No.
1,
February 1980 00369292/80:00040063
$02.00
0
1980
Scottish Economic Society
THE MEASUREMENT OF SECONDARY
LABOUR MARKET EFFECTS
ASSOCIATED
WITH GOVERNMENT TRAINING'
JAMES J.
HUGHES
AND
IAN
BRINKLEY
University
of
Kent, Canterbury and Centre for Environmental Studies
In recent years there has been a rapid expansion in government financed
training and retraining. Whether
or
not the extra investment absorbed by
these expanded manpower programmes
is
worthwhile to society-as distinct
from the individuals participating in the programmes-will be influenced by
the secondary labour market effects associated with them.
Although this paper focuses upon the secondary effects associated with
Skillcentre
(SC)
training, the conceptual issues involved, the methodological
approach adopted and the problems encountered in measurement, are all
applicable to other types
of
government training. They might also be relevant
for
analysing other labour market phenomena such as the impact of re-
dundancy upon a local labour market.
The paper divides into six sections. The first defines terms and discusses
some
of
the conceptual issues involved. The second discusses the implications
that secondary effects have for cost-benefit analysis and comments on the way
in
which the literature has dealt with them. The remainder
of
the paper is
concerned with research recently carried out in the Southampton-Portsmouth
area. Sections
I11
and
IV
define the survey area and outline the methodology
adopted. Section
V
reports the results
of
this research while section
VI
discusses how these affect the latest cost-benefit calculations that are available
to
policy makers.
I
DEFINITIONS
AND
CONCEPTUAL
ISSUES
The most important secondary effects associated with government training
are likely to be those
of
replacement
and
displacement,
although an
induced
employment
effect might also sometimes occur.
Direct replacement is said to occur if
a
job
that would have been occupied
by
a
trainee, had he not undergone training, is filled by someone who would
otherwise have been unemployed. However, replacement might not be as
simple and direct as this definition implies since an individual moving from
This paper is based upon research financed by the Training Services Agency and the
Manpower Services Commission. We are grateful
to
both organisations for financial
support and to Vernon Ward for his advice and encouragement throughout the project.
Our thanks are
also
due
to
Philip Beaumont and John Bowers, both
of
whom commented
upon
an
earlier draft of the paper.
Date of receipt of final manuscript:
25
May
1979.
63
64
J.
J.
HUGHES
AND
I.
BRINKLEY
employment to training might initiate
a
whole chain of job changes. Indirect
replacement occurs provided that an unemployed person is ultimately
drawn into the vacuum that is created, however far down the chain this
occurs. Such indirect replacement, coming at the end of
a
sequence of job
changes, will be difficult to identify since the job changes will not take place
simultaneously. Furthermore, this lack
of
simultaneity will mean that
replacement might take some time to show through, even though the economic
climate remains unchanged. But in practice the economic climate never
remains unchanged for long and therefore the scope for replacement
will
change. Although
a
trainee might have been employed immediately prioi to
training there
is
no guarantee that he would have remained
so.
Similarly,
even though he was unemployed immediately prior to training he might
have become employed at
a
later date. Since the scope for replacement only
exists during those periods in which the trainee would have been employed,
and since the probability of this employment will vary over the cycle, the
scope for replacement will likewise vary. As a consequence actual replace-
ment might also vary.
At any point in time the job vacated
by
an individual trainee will either
have been filled or not filled, assuming that it still exists. If the job has been
filled by someone who would otherwise have been unemployed then we have
full replacement.
If
it has not been filled, or has been filled at the expense
of
a
vacancy elsewhere, then we have zero replacement. In other words,
at
any
instant in time we can never observe partial replacement associated with a
particular job, although over a period of time this is obviously possible.
At any point in time the replacement associated with a group of trainees
whose jobs still exist,
Y,
is given by
Vf
--
r=
Vf
-
Vf+V,.+V,
N-S
where
Vf
=
the number of trainee jobs filled, directly and indirectly, by
Vf,
=
the number of trainee jobs filled other than by the un-
V,
=
the number of trainee jobs that still exist but are unfilled;
S
=
the number of trainee jobs that no longer exist, and
N
=
the total number of trainees in the sample.
For the group
r
can assume any value between
0
and
1.
Clearly
r
will vary
through time if
Vf,
V,.
or
V,
change.
If
trainee
jobs
that
no
longer exist are included in the denominator then
we have an alternative measure of replacement,
R.
This shows the number of
trainee jobs filled by the unemployed relative to the total number of trainees
in the sample, i.e.
the unemployed
;
employed
;
To continue reading
Request your trial