Memorandum by the Royal Society of Edinburgh
1. The Royal Society of Edinburgh (RSE)
is pleased to respond to the House of Lords Economic Affairs Committee
Inquiry into the Economic Impact of Immigration on the UK Economy.
This response has been compiled with the assistance of a number
of expert Fellows of the RSE in the form of a small working party,
under the direction of the Vice-President, Professor John Mavor.
The working party has sought to answer those questions which it
feels it can most authoritatively comment upon.
2. As Dustmann and Weiss (2007) indicate,
there are distinctive types of migration, with different implications.
Thus, there are economic motives for migration and non-economic
ones, in some cases related to natural disasters or persecution
(eg asylum seekers). There are both legal and illegal immigrants.
Immigration may be temporary or permanent—perhaps the majority
of cases falling into the former category. Migration may be circulatory
(with frequent movements between host and source countries) especially
in the case of those engaged in seasonal work, or transient, with
movement to various host countries before fixing on a final destination.
There is also contract migration for a fixed term under a working
contract or residence permit. Finally, there is return migration,
whereby migrants return voluntarily to their country of origin.
The economic implications arising from each of these forms are
quite distinctive.
Q.1 What are the numbers and characteristics
of recent immigrants—age, gender, country of origin, immigration
status, duration of stay, skills and qualifications? How do the
characteristics of EU migrants differ from other migrant groups?
What are the expected future trends for immigration from within
and outside the EU?
3. The effect of immigration on the economy
depends on a number of factors including the state of the economy,
the nature of the immigration and the type of immigrants, but
the increase in the scale of immigration has certainly raised
its importance in policy terms. According to the Labour Force
Survey (LFS), over 10% of the British working age population was
born in another country, but many of these are recent arrivals.
According to Dustmann and Fabbri (2005), around 8% of all immigrants
arrived within the last year they analysed (2004) and around 40%
within the previous decade. Furthermore, the composition has changed.
While in the 1990s the largest single group were of Irish origin
by 2004 those born in the European Union (excluding Ireland, but
including those from the eight accession (A8) countries) made
up the largest group. A8 immigrants tend to be young, male and
single. They are generally in less skilled occupations. Indeed,
one in five of them are in temporary jobs, though a high proportion
of them have a qualification.[37]
Immigrants are far from homogenous. There are substantial differences
in economic outcomes depending on colour and ethnic origin. Thus,
white migrants, if anything, fare better than the indigenous white
population, whilst Pakistanis and Bangladeshis appear to be particularly
disadvantaged.
4. As the Committee has recognised, it is
difficult to obtain accurate data on the number of immigrants
in the UK and this makes it difficult to draw accurate policy
conclusions. There are four main sources of data on flow statistics.
First, the Worker Registration Scheme suggests a figure of 486,660
workers have registered to work in the UK between EU enlargement
in May 2004 and September 2006, but this figure does not include
the self-employed as well as those who fail to register. The second
is the International Passenger Survey which is designed to capture
those staying more than 12 months, which suggests a figure of
408,000 but the accuracy of the data has been questioned. The
third measure is obtained from National Insurance Number Registrations
and, although these include the self-employed, these provide a
figure of 380,000. Finally, the LFS, which is designed to capture
those staying more than six months measures the stock of migrants.
The figures quoted above represent lower bound measurements and
there is even greater uncertainty over upper bound measurements
and Blanchflower et al (2007) conclude, The number of
A8 migrants entering the UK since Accession appears quite uncertain
from the available data.
5. There is also a problem with the measurements
of return migration. There are no procedures in place to capture
immigrants who leave the UK. Dustmann and Weiss (2007) suggest,
however, that on the basis of LFS data about 60% of male and 68%
of female foreign-born still remain in the UK after five years,
but there are substantial differences among groups, with the more
educated tending to leave early. Improving the immigration statistics
should therefore be a policy priority.
6. Using the Worker Registration Scheme
data, Gilpin et al (2006) suggest that the UK attracts
migrants from countries in Eastern Europe with the lowest GDP
per head, and the highest unemployment rates. However, Blanchflower
et al (2007) suggest that the Human Development Index (HDI)
is an even better predictor of migrant flows. If the low scores
on the HDI are temporary, resulting from adjustment to post-communist
structures, then we may well expect to see a significant slowdown
of migration to the UK.
Q.2 In what sectors and occupations are immigrants
employed? How do immigrants' labour market outcomes—including
their employment rates and earnings—compare to those of
local workers? What determines migrants' performance and integration
in the UK labour market?
7. For Sweden, Aslund and Rooth (2007) find
that assimilation is strongly dependent on favourable labour market
conditions. Early outcomes affect future performance, so that
it is important that immigrants start in good locations from the
point of view of employability. Both "when" and "where"
matter from a long term perspective. Therefore, attention should
focus on the initial entry to the UK and the UK Household Longitudinal
Study extension of the British Household Panel Survey will enable
understanding of diversity within the population and will assist
with understanding the long term effects of social and economic
change.
8. There have been a number of studies which
have attempted to estimate the economic impact of immigration
in the UK. These recognise the likelihood of productivity differences
between immigrants and British born individuals, due to differences
in socio-economic characteristics, levels of education, demographic
composition and English language fluency or lack of it. The skills
acquired prior to migration may not be fully transferable to the
UK and educational qualifications may not be recognised as equivalent
to UK qualifications. Migration may result in assimilation over
a period of time and this will be more marked for second and third
generation migrants. An alternative is that minorities form enclaves
in certain locations, where there may be advantages in serving
the needs of other members of the same community, but the evidence
suggests that in the long run migrants who assimilate fare better
in economic terms. Enclaves may also inhibit migrants from improving
their understanding of the English language and adjusting to UK
culture. Indeed, the English language is a competitive advantage
to the UK in terms of attracting immigrants and should be built
upon.
9. Most studies have adopted the so-called
spatial correlation approach which focuses on the correlation
between immigrant labour inflows and changes in native or overall
labour market outcomes. However, as Dustmann et al (2005)
point out, the construction of the counterfactual (ie what would
have happened without the immigration) involves assumptions which
might be challenged. Nevertheless there are some consistent results
from the various studies suggesting that outcomes are far from
homogenous among the different immigrant groups. Thus in terms
of employment non-white immigrants from the Caribbean, Africa,
Pakistan and Bangladesh have significantly lower employment probabilities
than most white immigrants, who have a similar employment experience
to the British born white population. Employment rates for non-white
immigrants also display much greater volatility than those of
British born whites or white immigrants. The rates for non-white
immigrants also decline and recover faster over the cycle, suggesting
that they act as an inbuilt stabiliser. This is an important point
that is often overlooked as it implies that the local workforce
can be insulated from some of the labour volatility and economic
shocks.
10. Pakistani and Bangladeshi women have
substantially lower participation rates than native born white
women (with unconditional participation probabilities more than
50% lower). Immigrants have a higher probability of being self-employed
than the British born population. Finally, controlling for individual
characteristics and region of residence all non-white male immigrant
groups have average wages which are more than 10% lower than those
of the white British born population. For women the overall differential
is somewhat lower.[38]
In contrast, white migrants have higher earnings than British
born whites (Dustmann and Fabbri, 2005).
11. To what extent these differences are
due to discrimination, culture or religion or to genuine productivity
differences is less certain. Frijters et al (2005) find
that immigrant job search is less successful than that of the
native population, but improves over time. Battu and Sloane (2004)
find that ethnic minorities are more likely to be in jobs for
which they are over-educated, and for Pakistanis and Bangladeshis
this is more likely where qualifications have been obtained abroad.
Manacorda et al (2006) find evidence that native born and
immigrants are imperfect substitutes—a 10% rise in the share
of immigrants in the local population increases the native-migrant
male wage differential by just 2%, and this could explain why
there are few signs of a negative impact of immigrants on local
workers. However, an alternative explanation is that immigrants
in a specific area simply push non-immigrants out of the locality,
thereby defusing the local labour market effects. Thus, Hatton
and Tani (2005) find that an increase of 100 in net immigration
to a region from abroad generates a net outmigration to other
regions of 35. There is some anecdotal evidence in the case of
A8 migrants that productivity of immigrants might actually be
higher than that of the local employees.
Q.3 Why do employers want to hire immigrants?
Which sectors and occupations in the UK economy are particularly
dependent on migrant labour and why? What is the impact of immigration
on mechanisation and investment in technical change? What are
the alternatives to immigration to reduce labour shortages?
12. A 2006 Home Office Survey reports that
many employers find that migrant workers have a superior attitude
and work ethic and are more prepared to work longer and flexible
hours. However, there are also concerns about the conditions under
which many migrant workers operate and the Low Pay Commission
has expressed concern that some of them may be being paid below
the National Minimum Wage.
13. In Scotland and the UK as a whole, the
food, agricultural, healthcare and hospitality sectors are dependent
on migrant workers and there are considerable benefits to employers
from employing migrants in sectors which face strong competition
from abroad and where profit margins are tight or that even the
continuing existence of the industry is in question (Craig, 2007).
In addition, migrant workers can fill skilled employment gaps,
as has been illustrated by the increasing number of new NHS dentists
that are migrants, while migrants have long been a source of supply
of doctors to the NHS. Though the data are not recent, a report
to the Scottish Executive (Elliott et al 2001) records
that in 2000 16.8% of all hospital doctors and 5% of all GPs in
NHS Scotland qualified outside the UK. The proportion of hospital
doctors had increased from 12.3% in 1990, while the proportion
of GPs was about the same as a decade earlier. This evidences
that migrants have for some time been an important source of physician
supply to the NHS. With the expansion of medical school places
in the last two years this position looks set to change.
Q.4 What impact has immigration had on the
labour market, including wages, unemployment and other employment
conditions of the UK workforce, and has it differed for skilled
and unskilled employees? How does the minimum wage affect the
impact of immigration?
14. The main market response to removing
economic disparities among countries is either labour migration
or firm migration, with the location of call centres in low wage
countries being a mixed mode. Labour will move from low wage to
high wage countries, attracted by higher wages and the greater
availability of jobs. If mobility is on a sufficiently large scale
then changes in relative supply and demand for labour should increase
wages in the low wage economy and lower wages in the high wage
economy until the wage disparities net of mobility costs are removed.
However, as Dustmann, Fabbri and Preston (2005) note, immigration
inflows will change the skill composition of the host country
labour force if the skill composition of immigrants differs from
that of the host country, so that the restoration of equilibrium
implies a need for short-run changes in both wages and employment
for different levels of skill or perhaps long-run changes in the
economy's output mix. In the UK, the main impact has thus been
to put labour market pressure on low-skilled rather than high-skilled
workers. The impact on the wage distribution has been to increase
the disparity in that distribution. As theory (Altonji and Card)
suggests, the shift in supply of low-skilled workers depresses
the wages of unskilled workers but increases the wages of skilled
workers, and may increase the return to capital. The overall impact
on the average wage should be positive. The econometric analysis
in Dustmann, Frattini and Preston (2007) appears to confirm this.
If the minimum wage were to act as a floor to this downward pressure
then like any "price control", it could result in the
emergence of a "black market" for low-skilled labour.
Perhaps the scandals of "gangs" in UK agriculture and
fishing reflect this.
15. There is little robust evidence that
immigrants have harmed native employment opportunities (See Card,
2005), and despite the introduction of, and increase to the minimum
wage in recent years, there appears to have been little impact
on employment opportunities in the UK. However, it should be noted
that since 2004 a very substantial proportion of A8 workers who
have registered have been aged under 24 and since 2004 the unemployment
rate among native 18-21 year olds has been rising. Further research
is required to investigate developments in the youth labour market
in the UK.
Q.5 What is the economic impact of illegal
immigration, including on employment, wages and the fiscal balance?
16. We are not aware of detailed research
on this in the UK, although there is work in the US looking at
illegal immigrants from Mexico. However, the impacts are primarily
likely to be felt in low-pay/low-skill jobs and if they were widespread,
one might be able to pick them up from inconsistencies in data:
eg between various measures of the value of output, or in data
on claimant counts in "low-skill" sectors as "legals"
are replaced by "illegals". This latter effect would
have fiscal implications as it would reduce income tax revenue
and increase government expenditure.
Q.6 What is the economic impact of a net change
in the UK population? If there is a net increase, does the impact
differ when this comes from higher immigration rather than from
changes in birth and death rates?
17. As is well-known, the dynamics of population
structure have significant long-term implications for savings
and pensions. Indeed, immigration can be regarded as a mechanism
for offsetting the adverse consequences of an ageing population
in many developed economies, since immigrants are typically younger
than the mean of the indigenous population and have higher fertility
rates (Drinkwater and Blackaby, 2004). It is worth noting in this
respect that the one developed economy where this is seen as less
of a problem is the US, and this is due to the impact of immigration
flows into the Hispanic community. Birth and death rate effects
are likely to be very long term whereas immigration effects would
operate more quickly.
Q.7 What has been the impact of immigration
on key macroeconomic indicators: GDP and GDP per head, unemployment,
productivity, investment, inflation and asset prices especially
housing? Do the economic effects of immigration vary over time?
18. To put the current increase in immigration
in context, the UK population grew at just 7% over the period
1971 to 2004, which is a much slower rate than in all other EU
15 countries, apart from Germany. Thus, the effect of immigration
since 2000 has been to accelerate this growth from a relatively
low level. Blanchflower et al (2007) suggest that for this
reason, despite the fact that half of the immigrants may have
returned home, inflationary pressures are likely to have been
reduced and the natural rate of unemployment lowered, with overall
benefits to the economy. These potential benefits are, however,
dependent on the type of immigrants, the timing of entry and their
location in the UK.
19. It is likely that immigration has generally
had beneficial effects on key macroeconomic indicators. Immigration
of higher skilled workers could raise productivity growth and
reduce the natural rate of unemployment by reducing upward pressures
on wages. And the beneficial effects of migration on supply are
not matched by equivalent increases in demand, since a significant
proportion of migrant workers' earnings are remitted to the home
country. The net effect is a reduction in inflationary pressure.
It is also true that recent A8 migrants have been less likely
to claim benefits, with beneficial effects on government finances.
Finally, if higher skilled workers are complementary to capital
as previous literature (eg Griliches 1969) has suggested, then
skilled immigration may have a beneficial impact on investment.
Q.8 How does immigration affect public finances?
Do immigrants contribute more in taxes than they use in public
services? As the UK population ages, does immigration affect the
shortfall in pension funding?
20. This is a complex area of study, both
in terms of the economic impacts as well as wider political and
social impacts, and much can depend upon the methodology employed.
As we have highlighted in previous answers, immigrants are far
from homogenous and while studies undertaken to-date have been
able to focus upon the aggregate contribution of immigrants on
public finances, decision makers need to be aware of and fully
consider the diversity of the immigrant population and their relative
performance.
21. Sriskandarajah, Cooley and Reed (2005)
found that in terms of tax revenue the foreign-born population
makes a relatively higher contribution than the UK-born. This
is due to the foreign-born having higher average earnings than
the UK-born and being over represented at the upper end of the
income spectrum, thereby paying proportionally higher amounts
of tax. The key finding from the paper is that the difference
between the net annual fiscal contribution (NAFI), expressed as
a ratio of revenue to expenditure, for foreign-born and UK-born
has been getting larger, in favour of the foreign-born, over recent
years. Indeed in 2001-02, despite the UK-born NAFI being net negative,
the foreign-born NAFI remained positive. Therefore, the analysis
suggests that the relative net fiscal contribution of immigrants
is stronger than that of the UK-born, and has been getting even
stronger in recent years.
22. Ultimately, there is no evidence to
suggest that recent immigrants from the A8 countries have come
to claim or receive benefits. Rather, they have come to work and
similar conclusions have been arrived at for both Sweden and Ireland.
Q.9 How has immigration affected public services
such as health care, education and social housing? How has this
varied across the country?
23. In addition to the answer to question
eight, the contribution that immigrants make to the provision
of public services in the UK has to be considered (Sriskandarajah
et al). Immigrants are disproportionately employed in those
sectors involving public service provision. This can be emphasised
by the fact that in 2003 29.4% of the total number of doctors
employed in the UK were foreign-born (Sriskandarajah et al),
though the proportion is evidently much higher in England than
in Scotland.
24. Immigrants face many hurdles when they
enter the UK but it could be argued that it is the housing system
and its consequences, rather than employment issues, which pose
the greatest difficulties. The question focuses upon social housing
but a very significant proportion of migrants rely on the private
rented sector. Housing issues in the UK, such as availability,
affordability, and conditions may negatively impact upon the UK's
competitive position in attracting migrants in the future.
Q.10 How does the impact of immigration vary
across different regions of the UK?
25. Immigrants do not distribute themselves
randomly throughout the UK, but tend to be concentrated in certain
locations. Home Office figures show that London and the South-East
attract the greatest proportion of immigrants. Although, there
is little correlation between the relative distribution of migrants
and the growth of unemployment.
Q.11 Are there any relevant parallels and
lessons for the UK from the economic impact of immigration on
other OECD countries?
26. Overwhelmingly, the available data does
confirm the view that we have taken, that immigration has had
positive impacts.
Q.12 How do differences in migrants' skills
affect the economic impact of immigration? Does immigration fill
skill gaps? What impact, if any, has immigration had on education
and training? What is the relationship between the Government's
migration policies and labour market policies?
27. Migrant workers are generally in less
skilled occupations, though a high proportion of them have a qualification.
However, there are few data on the precise qualifications of immigrants.
The evidence suggests that on entering the labour market, migrants
"trade down" in terms of the kinds of posts that their
qualifications suit them for and, therefore, they may improve
the overall level of skills in the economy. It is clear that in
certain sectors of the UK economy, such as health and higher education,
important skills can only be effectively filled by immigration,
particularly in the short term, until there is satisfactory workforce
planning. As touched upon in the answer to question three, migrant
workers account for almost half of new NHS dentists.
Q.13 How can data on immigration be improved?
What improvements are already being put in place? To what extent
have "inadequate data" affected public policy? How confident
can we be in forecasts of future immigration and how important
is it that such forecasts are accurate?
28. As we mention in our answer to question
one, there are a number of data sources containing information
on the number of immigrants in the UK, but none of these is entirely
satisfactory and, therefore, improving the accuracy of immigration
statistics must be a priority. More detail in the data collection
in terms of the characteristics of immigrants is needed. This
brings us on to the important issue of the recognition of foreign
qualifications. As we refer to in our answer above, the statistics
do not contain information on the precise level of these qualifications.
It may be the case that we in the UK are simply undervaluing foreign
qualifications, or there may be good reasons for doing so. Whatever
the reasons, more precise data on the qualifications of immigrants
should be collected so that migration can be more effectively
managed. Public policy as well as public confidence in this area
will founder unless this is the case. Anecdotal evidence suggests
that the lack of accurate statistics has put increasing pressure
on local authorities and the services that they provide.
29. There is an important question as to
whether survey or administrative routes should be followed to
collect better information on immigration. Given that it is often
difficult to sample adequately from migrant groups, it is perhaps
true that the administrative route is the preferable one. It might
be interesting to explore whether the development of ID cards
might help in some way to rectify what is currently an unsatisfactory
situation.
Q.14 How do the Government's policies, including
immigration and labour market policies, affect the scale, composition
and impacts of migration? How will the points system for immigrants
from outside the EU operate? How will the Government decide where
there are skill shortages in the economy as the basis for its
points system? What has been the international experience, eg
in Australia and Canada, of such a points system? How will the
Government respond to employers asking for non-EU workers to fill
low-skilled jobs?
30. What then does all this mean in policy
terms? First, the evidence in general supports the view that immigration
has had a beneficial effect on the economy overall. Second, some
immigrants, notably some of those from ethnic minorities, are
relatively disadvantaged in terms of labour market outcomes and
may need protection in terms of equal opportunities legislation.
Third, experience differs markedly across different immigrant
groups, so blanket approaches seem inappropriate. Given the free
mobility of labour within the European Union it is not possible
to control all immigration. It may, however, be possible to control
non-EU immigration if this is thought to be desirable. Thus, a
number of countries have guest worker programmes which enable
them to obtain a permanent increase in their labour forces without
a permanent increase in their population and also to restrict
employment of migrant workers to certain sectors or occupations.
There are also two types of policy to reduce the incentive for
temporary immigrants to overstay. First, defer payment until the
end of the contract and deposit payment in an appropriate foreign
bank, accessible only when the migrant has left. Second, one can
make employers purchase a bond which is forfeited if the worker
overstays. Alternatively, one can utilise Mode IV contracts which
allow for the temporary movement of service providers between
a host country employer and a foreign firm.[39]
31. Based upon the experience of other countries,
notably Canada, it seems appropriate that the UK adopts a points
system. However, for the system to operate effectively, up-to-date
labour market statistics are needed and the operation of the system
will have to be periodically appraised and "re-calibrated",
if need be, to ensure that skills shortages are effectively managed.
32. Thus, there are a number of devices
that can be utilised if immigration is regarded as too large but
there is no hard evidence that this is currently the case in the
UK, and we are not yet in a position to be able to define the
optimal degree of immigration.
Q.15 Should more be done to help immigrants
boost their productivity in the UK?
33. It is not clear that immigrants are
"low productivity" workers. Although, as has been made
clear in previous answers, immigrants are not a homogenous group
and some immigrants, particularly those from ethnic minorities,
are disadvantaged and need assistance and protection in terms
of labour market outcomes.
34. Again, competency in the English language
is an issue here. There could be negative impacts on productivity
in some sectors if migrant workers struggle with English and co-workers
have to assist them. We encourage the Committee to examine the
language constraint and to recommend arrangements to improve the
language situation, which would in turn aid assimilation as well
as labour market outcomes.
27 September 2007
37 The LFS classifies foreign qualifications as "other",
so we have no information on the precise level of these qualifications.
But on average immigrants have higher levels of education than
the native population. Back
38
In the case of Polish immigrants differentials are even larger.
See Drinkwater et al (2006). Back
39
On this see Schiff (2007). Back
| 
