HOW MUCH UNEMPLOYMENT IS STRUCTURAL?
| Published date | 01 February 1988 |
| Date | 01 February 1988 |
| Author | Adrian Wood |
| DOI | http://doi.org/10.1111/j.1468-0084.1988.mp50001005.x |
OXFORD BULLETIN OF ECONOMICS AND STATISTICS, 50, 1(1988)
0305-9049 $3.00
HOW MUCH UNEMPLOYMENT
IS STRUCTURAL?
Adrian Wood*
In principle, everyone agrees (a) that unemployment can be caused by mis-
matches between the composition of the demand for labour and the composi-
tion of the labour supply, and (b) that such unemployment will persist if the
composition of labour demand and/or supply is unresponsive to such mis-
matches, for example because relative wages are insufficiently flexible. Many
people also believe, largely on the basis of casual empiricism, that persistent
labour demand/supply mismatches are responsible for much of the con-
temporary high levels of unemployment in the industrial market economies,
and especially in Europe. The types of structural mismatch usually
mentioned are with regard to skills and location (or region), but industrial,
gender (or full versus part time), and age mismatches are often also portrayed
as important.
Richard Jackman and Stephen Roper (henceforward J and R), in an origi-
nal and important article in the February 1987 issue of this B ULLE TIN,
undertake some statistical analysis which points to very different conclusions.
They discover (a) that in the UK structural unemployment is currently a small
proportion of total unemployment, and (b) that it has not risen as a propor-
tion of total unemployment over the past 20 years, apart from a jump in the
industrial dimension between 1979 and 1982. (The absolute amount of
structural unemployment has however risen in line with total unemployment.)
Much the same seems to be true of other European countries. These conclu-
sions, which J and R are careful to qualify in several respects, have been
extensively quoted and relied upon by other economists (for example,
Johnson and Layard, 1984; Layard, 1986; Layard and Nickell, 1986. Earlier
versions of the statistical work in J and R appeared in Blanchard, Dornbusch
and Layard, 1986 (apparently written in 1983); Jackman, Lyard and
Pissarides, 1984; Jackman and Roper, 1985.)
Are J and R's conclusions in fact robust enough to bear the weight which
others have put on them? The following paragraphs consider several reasons
(some already mentioned by J and R) why structural unemployment - or,
more precisely, unemployment due to demand/supply compositional mis-
matches - may be much more important than J and R's calculations imply.
At the end, possible implications for the analysis and policy recommenda-
tions of the other economists mentioned above are briefly outlined.
* I am very grateful to Richard Jackman, Trevor King and Carol van der Ploeg for conmients
and suggestions.
71
To continue reading
Request your trial