APPENDIX 9
Memorandum from Andrew Glyn and Esra Erdem,
Corpus Christi College, Oxford and Oxford University Department
of Economics (JG 15)
THE UK JOBS GAP—LACK OF QUALIFICATIONS
AND THE REGIONAL DIMENSION
1. It is well known that the unemployment
rate, even when measured from labour force surveys, seriously
understates lack of work—many of those facing joblessness
drop out of the labour force (particularly men) or delay entering
it (particularly women). Such "economic inactivity"
is very unevenly spread over the UK regions—much more unevenly
than unemployment. Moreover inactivity tends to be higher where
there is more unemployment. The result is that the "non-employment
rate"—the proportion of the relevant population which
is not working (the unemployment rate plus the inactivity rate)—shows
a very wide degree of variation across regions. Figure 1 demonstrates
this for men aged 25-64. In 1997 both the national level of unemployment
and regional differences in unemployment were low by the standards
of the past two decades. However, inactivity was both high and
with wide regional variations so that a number of the older industrial
areas were still experiencing a very serious problem of joblessness.
Hardly more than 70 per cent of men aged 25-64 were in work in
Merseyside, South Yorkshire or Tyne and Wear (the table attached
provides the data for some of the most important series used in
the charts).
2. Lack of work is also particularly concentrated
amongst the least qualified; again differences in inactivity amplify
the differences in unemployment rates experienced by groups with
different educational qualifications. Those with no educational
qualifications show much the highest level of joblessness (figure
2). In 1997 around 40 per cent of men aged 25-64 without qualifications
were not working, compared to around 10 per cent of those with
degrees.
3. Regional and educational disadvantage
are not separate and independent. On the contrary they interact
with each other so that differences in joblessness between regions
are far larger for those with poor educational qualifications
and they are far higher for non-employment as a whole than for
unemployment as registered in the surveys. Figure 3 shows that
regional variations in non-employment are comparatively moderate
for those with qualifications ranging from university degree to
O levels; but regional differences are much higher for those with
the lower categories of vocational qualifications and they are
absolutely huge for those with no qualifications. Even in the
most prosperous regions (South East excluding London and East
Anglia) the non-employment of men without qualifications was nearly
30 per cent in 1997. Obviously some non-employment reflects genuinely
voluntary early retirement, or other causes such as illness unconnected
with the availability of work. However, a good deal of this joblessness
must have been involuntary in origin and suggests that the disadvantage
suffered by the least qualified is not confined to the less prosperous
regions. However, in the older industrial areas the situation
of the least qualified is truly catastrophic: a majority of men
without qualifications do not have work.
4. Joblessness for the least qualified men
rose sharply in the recession of the early 1980s (see figure 4,
though it should be noted that the rise between 1981 and 1983
is exaggerated by a change in the way the Labour Force Survey
was conducted). Since then non-employment has fluctuated with
the economic cycle, but around a seemingly inexorable upward trend.
Differences between regions in non-employment have remained very
large (in contrast to the dispersion of unemployment rates which
has declined).
5. Those without qualifications are a declining
smaller proportion of the labour force. More than 45 per cent
of men aged 25-64 had no qualifications in 1979, whereas in 1997
the proportion was 15.8 per cent. The least prosperous regions
have the highest proportions (over 20 per cent in four cases—see
the table). This reflects both the tendency for school students
to gain fewer qualifications in these areas and for outward migration
from these areas to affect mainly the more qualified sections
of the labour force. It would be absolutely wrong to conclude
from the declining number of those without any qualifications
that problems in the labour market for the less qualified will
soon vanish. Figure 5 shows the regional evolution of non-employment
for the first educational quartile of men aged 25-64 (the least
educated quarter); up till the early nineties all the bottom quartile
had no qualifications, so the chart exactly mirrors figure 4.
Thereafter the bottom quarter includes some of those with some
"other" vocational qualifications. They also have low
employment rates with wide regional variations, although their
position is better than those with no qualifications. Moreover
the higher proportion of the unqualified in the poorer regions
means greater disadvantage in the first quartile as compared to
that in the higher employment areas. Thus looking at the bottom
quarter, non-employment rates are still extremely high and with
a very large regional variation. The problem cannot be written
off as confined to a small and diminishing minority.
6. Nor should the problem of very high regional
joblessness for the less qualified be seen mainly as a problem
for older workers. Figure 6 shows the regional variation in non-employment
for the bottom one quarter of the educational distribution in
each 10 year age group between 25 and 64. For the oldest age group
these are all without qualifications; for those aged 25-34 the
bottom quarter includes a substantial number with "other"
vocational qualifications. Joblessness is highest of course for
those over 55, but it is both extremely high and very regionally
dispersed across all the age groups. This pattern—a very
high and regionally very diverse pattern of non-employment for
the least qualified, extending across all age groups—presents
an extremely sombre and depressing picture of the UK labour market
and one that should be at the centre of concern.
7. If the position of regions fluctuated
quite rapidly in the league table of joblessness, this would suggest
that the regional differences tend to right themselves (by job
creation or migration) and thus do not merit special attention.
This is far from the truth. Figure 7 plots the non-employment
rate for the unqualified in 1979 against the corresponding figure
for 1997. In general the low employment regions in 1979 were in
a similar position in 1997 (the correlation between their rankings
in the two years is highly significant). The fit is not perfect
of course; Northern Ireland's relative position has improved,
whilst joblessness in South Yorkshire has increased more than
would have been expected on the basis of the position in 1979.
8. The analysis thus far has concentrated
on men. Their experiences exemplify the educational and regional
dimensions to the problem of joblessness, and it is simpler to
follow through the argument in relation to one group of the labour
force at a time. However, it would be quite wrong to suppose that
the underlying upward trend in women's participation in the labour
force means that less qualified women have escaped the joblessness
described above. Figure 8 shows the very high level of non-employment
for women without qualifications, substantial regional variations
and the fact that this group has singularly failed to benefit
from the increased job opportunities for women generally. Figure
9 confirms that, for women as well as men, the high level of,
and wide regional variations in, joblessness for the least qualified
is spread across all age groups.
9. Why are the least qualified far less
likely to have work in some regions than in others? It seems obvious
that the most important influence is the general level of employment.
Figure 10 illustrates this relationship by comparing non-employment
for men without qualifications in 1997 with the average level
of non-employment for men from all the educational groups. Less
obviously, the least qualified are disproportionately affected
by lack of work generally; a 1 per cent higher non-employment
rate generally in a region is associated with almost a 2 per cent
higher employment rate for the least qualified and this relationship
is highly significant statistically, whilst those with degrees
have regional employment rates virtually independent of average
employment rates in the region. This pattern is consistent with
the general idea that in areas with low employment many of the
least qualified get "bumped down" the jobs ladder and
out of work.
10. The very fact that the regions with
particularly low employment for the least qualified are universally
thought of as the old industrial areas suggests that the problem
is that decline of traditional industries has deprived the least
qualified of work. This is of course exactly what happened. However,
it would be wrong to conclude that only industry provides work
for the least qualified (especially men) and that services jobs
cannot fill the gap. Simply in terms of providing work (and leaving
aside pay and conditions) a statistical analysis reveals that
an extra 1 per cent of people employed in services brings as many
jobs for the unqualified as the same amount of extra employment
in industry (true for men as well as women). The problem of the
old industrial areas is that, having lost very large numbers of
industrial jobs (especially in the most urban parts, see Turok
and Edge 1999), they have not replaced them with additional service
jobs. Whilst the majority of service jobs depend on regional incomes
generally (like retail trade) or on population (like health and
education jobs largely financed by central government), the finance
and business services sector in particular produces "regional
exports" just like manufacturing. Preserving and extending
what remains of the industrial base, and extending it into service
sectors with the potential to bring income into the regions, is
essential for meeting the jobs crisis for the least qualified.
11. The degree of regional joblessness market
poses severe problems for the government's "welfare to work"
programme, as has been forcefully pointed out (Turok and Webster
1998). This note has documented how regional employment problems
are overwhelmingly concentrated on the least qualified. Though
the regions considered here are large, some of them have astoundingly
high levels of joblessness amongst the least qualified and the
problem extends through the full range of age groups. Local concentrations
of joblessness are far worse still. It is very hard to see how
the government's strategy centred around Welfare to Work can be
adequate to meet these problems. All the focus is on the "employability"
of targeted groups of individuals—their attitudes, experience
and the response of employers to them. Even if this programme
was very successful, and large numbers of people are induced to
search more actively for work, the process by which this is supposed
to create more jobs is essentially macroeconomic. It is presumed
that a larger labour force looking actively for work, a bigger
reserve army of labour, will hold down wage pressure generally;
this in itself is a debatable assumption given that those involved
in Welfare to Work policies (long-term unemployed, people on sickness
benefit, single parents, workless households and so forth) are
disproportionately concentrated in the low employment areas where
the labour market is already very weak especially for the least
qualified. If there was a significant effect on wage pressure
generally, the argument is that the Monetary Policy Committee
would respond by running a more expansionary policy, which in
turn would tend to increase employment generally. However, there
is no way that this would have a disproportionately large impact
in creating work in the areas where joblessness is disproportionately
concentrated. Even on very optimistic assumptions about the chain
of effects from policies to increase the effective labour supply
to the macroeconomic creation of jobs, there is no mechanism for
bringing the jobs to the areas where they are most needed.
12. This memorandum has attempted to highlight
the extreme regional concentration of joblessness amongst the
less qualified. The conclusion is that analysis of policies to
actively steer demand for labour into the low employment regions
(Gudgin 1995) should be brought back to the centre of policy discussion.
Andrew Glyn and Ersa Erdem
Corpus Christi College, Oxford and Oxford
Institute of Economics and Statistics
October 1999
| 
