REGIONAL UNEMPLOYMENT DIFFERENTIALS
| Author | IAN GORDON |
| Date | 01 February 1980 |
| Published date | 01 February 1980 |
| DOI | http://doi.org/10.1111/j.1467-9485.1980.tb00559.x |
Scottish
Journul
ofPoIitical
Ecopomy,
Vol.
27,
No.
1,
February
1980
rg
1980
Scottish Economic Society
0036-9292/80/00060097 002.00
Notes
and
Communications
(1)
REGIONAL UNEMPLOYMENT
DIFFERENTIALS: MIGRATION
NOT REGISTRATION
IAN
GORDON
Urban and Regional Studies Unit, University
of
Kent at Canterbury
The fundamentally distinctive feature of sub-national labour markets is
their openness with respect to migration, and in the case
of
smaller spatial
units to commuting also. Any coherent approach to the analysis of such
labour markets needs therefore to incorporate some explicit assumption
about the behaviour of migrant flows in the face of changing conditions.
In our earlier comment on Elias’ (1978) note we tried firstly to show that
the absence of such an assumption, or indeed of any explicit model, removed
any logical connection between the data
on
unemployment by duration
which he presented and the hypotheses advanced about redundancy policy
and registration propensities and secondly to sketch a model
of
regional
labour supply and demand in which the cyclical behaviour
of
migration
could account for conventional observations about the varying cyclical
sensitivity of regional unemployment (Gordon, 1979).
In
his reply, Elias (1979) points out (correctly) that the model was cast in
terms
of
total (male) unemployment whereas the time series data relates only
to the proportion who register-a difference he believes to be crucial-and
questions the practical significance of the role of migration as a factor at
the regional level. In support
of
these arguments he produces two further
sets of data but makes no clearer the assumptions which would link these
data with hypotheses about the functioning of regional labour markets.
We shall show that neither set of data actually supports his argument or
refutes the hypotheses of the migration-based model.
It would be helpful firstly to spell out some
of
the possible false assumptions
within our model about the role of migration in order to see whether Elias’
data bear on them. There appear to be three such of some significance, apart
from the radical simplification and omission
of
influences
on
migration
levels
:
(a) that net migrant flows are large in relation to regional differences in
employment growth in any period;
(b) that net migrant flows respond fast enough to fluctuations in unemploy-
ment relativities to produce an approximation
to
equilibrium in the
pattern of differentials; and
Date
of
receipt
of
final manuscript
:
2
August,
1979.
97
To continue reading
Request your trial