REGIONAL UNEMPLOYMENT DIFFERENTIALS
| Author | B. ELIAS,D. PETER |
| DOI | http://doi.org/10.1111/j.1467-9485.1980.tb00560.x |
| Published date | 01 February 1980 |
| Date | 01 February 1980 |
Scoffish
Journal
of
Political Economy,
Vol.
27,
No.
1,
February
1980
0
1980
Scottish Economic Society
003&9292!80/00070
103/$02.00
Notes and Communications
(2)
REGIONAL UNEMPLOYMENT
DIFFERENTIALS: A RESTATEMENT OF
THE EVIDENCE
D.
PETER
B.
ELIAS
Manpower Research Group, University
of
Warwick
In an earlier note (Elias,
1978)
an attempt was made to disaggregate registered
unemployment levels by region into a flow component (the incidence
of
unemployment) and a duration component (the mean completed spell of
unemployment). Cyclical variation in proxy variables for these components
was analysed, from which
I
concluded that regional differences in the cyclical
variation of unemployment levels were associated with variations in the flow
onto the register rather than duration. This
I
attributed to regional variation
in the cyclical response
of
the propensity
of
unemployed persons to register
as unemployed upon being made redundant, voluntarily leaving an employer
or
upon joining the labour force. Gordon
(1979)
criticised my analysis for its
neglect
of
the role
of
migration. In my reply (Elias,
1979)
I
agreed with Gordon
that interregional labour migration must have an influence upon regional
unemployment
levels,
but pointed out that the cyclical variation in the
$ow
onto the register in all regions is large with respect to the net migrant flows,
with the possible exception
of
the South West. Furthermore,
I
gave some
crude indicators
of
regional differences in the propensity to register as
unemployed.
In his latest note, Gordon
(1980)
re-emphasises his earlier points. He
stresses that interregional migration can maintain differences in regional
unemployment levels. On this point we are agreed,
so
I
am unclear as to
his
reasons for further elaborating his model
of
regional unemployment
differences. The point at issue though is whether or not cyclical variation in
interregional migrant flows can account
for
the regional differences in the
cyclical variation
of
flow onto the unemployment register. The evidence
I
presented earlier (Elias,
1979)
showed that there is a considerable degree
of
variation in this flow. The proxy variable
I
chose to measure the flow per
two week period (the number of males
on
the register who have an uncom-
pleted spell
of
unemployment
of
2
weeks or less) underestimates this flow,
because
of
the loss of persons completing
a
spell
of
unemployment within the
two week period. Nevertheless the flow is highly variable. At the national
level, unemployment flow statistics show that a variation in the inflow to
the register of
60-70,000
from one month to the next is not uncommon.
Date
of
receipt
of
final
manuscript:
15
October
1979.
103
To continue reading
Request your trial