Transcript London`s Place in the UK economy

Some Impacts of Recent Immigration
on the London Economy
Ian Gordon and Christine Whitehead
LSE London
Seminar, March 17th 2007
Outline
• Selective review/discussion of some
(only) of the key issues, focusing on:
– Nature of flows and LT effects on London
population
– Labour market position of migrants and
relation to high worklessness in London
– Evidence of impacts in bottom quintile of
jobs
– Implications for housing
London and the New International Migration
• This city has very long history of immigration, by groups who have
played a significant role in its economy, via:
– Flows linked to trade and imperial role; and
– Periodic waves of refugees
• But never sustained migration on scale of the last 10-20 years
• Part of a new (post-Fordist?) international era of migration:
– qualitatively distinct from that of 1950s/60s, when flows largely
shaped by importing major employers/states
– followed by major cut-backs, in context of recession or political
opposition
– and from 1980s, a combination of: highly skilled movers; new
sets of refugees; speculative/clandestine migrants; trafficking
• Diverse, autonomous, and more strongly urban in character
• London
Immigration and London’s Turnaround
• The new immigration was directly responsible for the the
reversal of London’s 50 year population decline, from the late
1980s
• As such it is one of two key planks on which the Mayor’s growthoriented London Plan rests
• The other being the continuing growth of employment in the
business service sector
• Logically, international immigration might be seen as providing
the elastic labour supply needed to accommodate this
exponential growth
• But how far the two phenomena are – or can be – linked is very
unclear
– Particularly since main immigrant groups are not heavily
represented in the sector
London’s overseas-born
population 1986
• 1.17 million – 17.6% of total
• Of whom 76% from former British
territories
• Majority from just 6 countries
– Ireland, India, Kenya, Jamaica, Cyprus,
Bangladesh
London’s overseas-born
population 2006
• 2.23 million – 30.5% of total
• Of whom 59% from former British territories
– Most of the increase (60%) among other groups,
who grew by 225% over 20 years
– 25% less Irish but 200% more French
• Majority now from 15 countries:
– Previous 6 + Nigeria, Poland, Sri Lanka, Ghana,
South Africa, Pakistan, Somalia, USA, Turkey
Sources of Growth
• Worldwide trend to higher levels of mobility, associated with
enhanced communication of opportunities, air transport, growth
of skilled workforce – and proliferation of refugee-generating
crises/conflicts
• Particularly affecting areas which are:
– Economically attractive and ‘urban’
– With existing concentrations of non-native groups
– Close to ports of entry
 Large share of expanded UK immigration in London
• Trend is not London specific, but:
• It gets much above average share of inflows (40%+, cf. 30% of
outflows); and
• Inflow rates reflect not only strong time trend, but also previous
year’s employment growth in the city – e.g. flat trend since 1999,
as for jobs, after late 1990s growth in both
National Growth in Migration
but with distinct London elements
Gross International Migration 1985-2004 (000s)
400
350
300
250
LONDON IN
RUK IN
LONDON-OUT
RUK OUT
200
150
100
50
0
1985
1990
1995
2000
2005
Who Counts, and Who Gets
Counted, as Migrants ?
•
Formal UK definition = those entrants who had been out of the country
for > 1 year and now intend to stay for > than 1 year (plus those
‘visitors’ who later apply to extend their stay beyond a year)
– Does not relate directly to the degree of involvement in local labour/housing
markets (e.g. long term students versus working visitors)
– Includes as immigrants many returning natives (c. 25%), and as emigrants
many returning non-natives (c. 50%)
•
Official data comes from sample surveys of international passengers;
plus Home Office (HO) admin data on asylum seekers and status
switchers
– A general issue is that illegals don’t get counted: HO assumption that 80100% are in Census → estimate of 10-20% understatement.
– A London issue is significant overcounting in survey of numbers with a final
destination within London
•
Making a similar assumption about residential survey coverage (i.e.
migrants of all kinds equally likely to be included – except for new
asylum seekers, and some students, in institutional accommodation),
we rely heavily on LFS data on non-natives and their dates of arrival.
Long Term Impacts on London Population Change
• Conditioned by three main factors:
– Probabilities of return/onward migration
• In steady state boost to population = annual rate of in-migration/prob. of
exit (p.a.)
• Flows of temporary migrants generate one-off increase (and change
population composition) but do not produce continuing growth
• Flows of permanent migrants (only) produce continuing growth
– Displacement effects on other Londoners
• May include (i.e. select) a ‘white flight’ element
• But driven by growing gap between space demands and availability
• Displacement i– estimated at 40-45% by Hatton and Tani - is less than
100% primarily because (effective) space demands of new immigrants
are less than of average Londoner :
– If this doesn’t persist then impact on London population would be step
increase, not continuing growth
– Child-bearing by a younger migrant population (20-44)
Contrasting Patterns by Area of Origin
• High wage countries
–
–
–
–
•
predominantly white
38% of recent inflow but only 17% of stock growth post-1986
Large temporary element (1996 entry down 50% between 1997 and 2005)
Sensitive to cyclical developments in London economy
Asylum countries (from where asylum seekers are a large proportion of the
UK inflow)
–
–
–
–
–
•
(EU, USA, Old Commonwealth, Japan / Korea)
ethnically mixed
12% of recent inflow but 22% of stock growth post-1986
Delays in entering mainstream housing/labour markets – and surveys
Smaller temporary element (1995 entry down 37% between 1997 and 2005)
Flows driven by push factors, plus perceptions of control tightness
Other low wage origins (developing and transition countries)
–
–
–
–
–
also ethnically mixed
51% of recent inflow but 62% of stock growth post-1986
Small temporary element (1996 entry down 16% between 1997 and 2005)
Not evidently sensitive to state of the London economy
A8 countries (inc. Poland) as a sub-set – but not particularly concentrated in London
• All three groups show more education than the average Londoner – but
comparable with similarly young natives
Potential Economic Gains
from Immigration
• A more elastic labour supply
• Specific types of labour in short supply
– Highly skilled labour
• To compensate for short-fall in home supply
• To fill specialist niches
– To fill jobs rejected by domestic workers
• Facilitating upward mobility of natives
• Facilitating trade relations with migrants’ home
countries
• Benefits from cultural diversity
– Contributing to supply of exotic products/services
• Increasing attraction to tourists and mobile ‘talent’
– Stimulating product or process innovation
Some Evidence from Wider Literature
•
•
Surveys of international literature (Longhi et al, 2004/6) show diverse
results but overall point to:
– Very small negative effects on both native employment (at least in
Europe and for women) and wages
For UK, Dustmann et al (2005) show:
– No overall effect on native employment, though negative effects in
intermediate education group (compensated by probably pos. effect for most
educated)
– A wage effect, which is (if anything) positive
•
At City level, for US, Ottaviano and Peri (2005/6):
– report strong positive relations between change in foreign-born population
and both wages and rents
– And argue this as consequence of positive effects of cultural diversity on
productivity levels
•
In relation to A8 migrants to UK, Blanchflower et al (2007) argue:
– No apparent effect on regional unemployment trends
– Probably tended to lower natural rate of unemployment
– And higher supply outweighing demand effects should moderate inflation
Economic Activity of 25-44 year olds
Among London Immigrant Groups
NonMigrant
Migrant 0-3 years
Migrant >3 years
Status
Rich
Asylum
Countries Countries
Employed 80%
75%
26%
Other
Rich
Asylum
Other
Poor
Countries Countries
Poor
Countries
Countries
64%
79%
54%
65%
5%
8%
20%
9%
5%
8%
7%
4%
4%
7%
6%
5%
9%
7%
10%
13%
47%
21%
11%
29%
21%
Seeking
Work
Other
wanting
work
Not
wanting
work
Continuing Differentials in Employment
Outcomes
• Even after 3 years, migrants from poor countries
retain significantly lower employment rates
– about half of this differential is associated with racial and
religious composition
– reflecting broader patterns of disadvantage (and some life
choices) among resident population
– Poles remain exceptional with very high employment rates
• And significantly lower hourly earnings
– About 30% lower (controlling for years education)
– Reducing to 22% after control for ethnicity and gender
– Polish earnings differential seems larger (for 0-3 yrs. only)
Worklessness in London – is there
an Immigration Connection?
•
•
London has lowest employment rate of any UK region
Attracted attention recently (e.g. GLA, HMT), because of:
– shift in focus away from unemployment
– relative improvement in rates for north and west versus south (as whole)
– Paradoxical position for reputedly successful economy
•
•
•
Raised question about whether new immigration could explain – with
large supply increases still in post-2000 years when job growth stops
However, growth in workless gap (against GSE) has been solely in the
group saying they don’t want jobs, which has grown steadily since
1984, independent of demand/supply shifts
If immigration plays a role, it seems to be only via migrants’ own lower
employment rates (whether reflecting choices or competitive
failure/discrimination)
Immigration and the Bottom Tier of
the London Labour Market
• Part of Sassen’s (1991) ‘global city’ thesis was that
globalcityisation polarised the employment structure:
– With growth among worst paid job types as well as at the top end
• European researchers challenged this in relation to London,
Paris and Amsterdam
– suggesting welfare state model made a difference (Hamnett)
– or that polarisation in 1980s New York reflected much higher
levels of immigration there, offering an elastic labour supply
for low paid service jobs (Working Capital hypothesis)
• Possibly relevant that consumer services have shown much
more growth in London during 1990s than in 1980s, alongside
the upsurge of immigration here
– Visible concentration of some migrant groups in low paid
occupations suggested as cause of growth + downward pressure
on pay (May, Mills et al, QMUL)
Employment of London Immigrant Groups
by High/Low Paid Occupations
Years
in the
UK
Bottom
quintile
2nd
quintile
3rd
quintile
4th
quintile
Top
quintile
(<£6.4 ph)
(£6.4-£7.5)
(£7.5-£9.9)
(£9.9-£13.5)
(> £13.5 ph)
..
21%
20%
21%
19%
20%
0-3
18%
14%
12%
21%
35%
>3
19%
15%
16%
22%
28%
Asylum
Countries
AC
0-3
31%
24%
14%
13%
18%
>3
23%
20%
14%
21%
22%
Other Low
Wage
countries
OLWC
0-3
46%
20%
10%
14%
11%
>3
25%
19%
16%
17%
22%
Migrant
Origin
NonMigrant
High Wage
countries
HWC
Trends in Employment Share and
Wages for Jobs in the Bottom Quintile
(occupations with lowest hourly pay nationally)
1.4
1.2
1
Bottom Quintile Wage
ratio of London:RUK
0.8
Bottom Quintile
Employment Share,
ratio of London:RUK
0.6
0.4
0.2
0
1990
1992
1994
1996
1998
Source: analysis by Ioannis Kaplanis of NES microdata
2000
Causal Links
• Recent migrants from low wage economies are
concentrated in the lowest quintile of jobs:
– Though concentration reduces with longer stays
• High rates of immigration to London in the later
1990s depress wages in this segment in London
(relative to rest of country)
– Nominal wage up 5.5% 1995-9 in London
• cf. 14.5% in RUK
• Lower wages in this segment contribute to its
employment growth in London after 1994:
– Quintile’s share in London employment up from 16.2% in
1995 to 19.2% in 1999
• cf. from 19.4% to 20.6% in RUK
– Elasticity of employment with respect to wage in this quintile
= –1.2 (from regional panel 1991-9)
Implications for Housing
• Inmigrants impact on the housing system
immediate
• London’s projected growth in households
heavily concentrated on inmigration
• But inmigrants have lower headship rates
than indigenous population – so direct impact
less
• Direct effects on house prices, access to
social housing, utilisation of the stock and
homelessness
Table 1: Components of the Projected Increase in Households
2001-21: Regional Analysis
(thousands per year)
Longevity
External
Migration
Separation
of Couples
North East
2
1
3
2
1
-3
6
North West
7
4
7
6
2
-3
23
Yorkshire and
Humber
5
4
5
4
1
-1
18
East Midlands
4
2
5
4
1
4
20
West Midlands
5
4
5
5
2
-2
19
East of England
5
4
6
5
2
6
28
London
6
25
6
6
2
-9
36
South East
8
8
9
7
2
2
36
South West
6
3
6
5
1
5
26
48
55
52
44
14
0
213
England
Household
representative
rates
Other
changes
(A)
Residual
Total
Table 2: Ratios of Household Reference Person to Total
Private Household Population: England 2002 to 2005
(per cent)
Age
16-24
25-29
30-44
45-64
65 and over
Born outside the UK
Entered less than 5 years
ago
23.8
37.9
49.1
57.8
29.4
Entered UK 5 years but less
than 10 years ago
19.6
43.3
52.9
57.9
43.6
Entered UK 10 years but less
than 20 years ago
10.5
45.9
55.6
60.5
50.5
Entered UK 20 years or more
ago
23.5
41.4
58.1
60.0
69.8
Born in the UK
13.7
45.9
57.3
59.1
70.5
Whole private household
population (a)
13.7
43.5
56.3
59.0
70.1
Competition for Housing
• Need to distinguish
– higher income employed migrants (often
shorter term) which affect house prices
– Lower income employed migrants who usually
enter private renting – often in HMOs initially
using less space; relatively low income
elasticity?
– Refugees and asylum seekers who are mainly
dispersed but often return to London
– Increased pressure on social housing sector
both direct and indirect
Average Property Prices
(2005/06)
£170,618 - £229,874
£229,874 - £272,959
£272,959 - £332,073
£332,073 - £782,118
Ratio HA/OO
(2005/06)
0.16 - 0.28
0.28 - 0.29
0.29 - 0.31
0.31 - 0.36
Ratio PRS/OO
(2004/05)
0.48 - 0.58
0.58 - 0.60
0.60 - 0.64
0.64 - 0.68
Second Round Effects
• Continued movement out of more
established family households to South
East and further afield – different
household composition of capital
• Effect on quality and stability of local
services but also filling important gaps
• Higher house prices implies less
movement out of social sector  even
greater pressure on system
• Competition for services and its impact on
social cohesion
Some Conclusions
• Major demographic shift in London population – both in scale
and diversity:
– In long run, principally involving migration from low wage
economies – though more flexibility/variety as result of inflow from
high wage areas
– As housing aspirations among new migrants converge on those of
rest of London population, overall growth rates should slow down
• No evidence yet of effects on overall London employment
growth, earnings, productivity or unemployment, but:
– Extended education among the migrants from low wage economies
is not being reflected in short-medium term by absorption into
appropriate job types
– Levels of worklessness (principally inactivity) are high among this
group of migrants (though partly for cultural reasons, among
married Muslim women as well as barriers to accessing jobs)
– Concentration of this group in bottom quintile jobs – especially
during first 3 years – seems to have significantly depressed London
wages at the bottom end, and consequentially boosted employment
from mid-late 1990s