Contribution of Immigration to GDP

13 January, 2005

Summary
1. The governments economic justification for the present very large scale immigration has been remarkably sketchy, consisting primarily of two broad claims. One is that, although migrants make up only 8% of the population, they contribute 10% to the UKs Gross Domestic Product (GDP). The other claim, made by the Prime Minister in a speech to the Confederation of British Industry, is that the UKs economic growth rate would be some 0.5% lower for the next two years if net immigration were to cease.

2. Migration Watch has examined these claims in detail. They both rely on an extremely simplistic methodology which is probably not tenable. Nevertheless we have used the same methodology and have concluded that both claims are grossly misleading.

3. The claim that migrants make up 8% of the population and contribute 10% to GDP has a fatal flaw. It does not include UK-born dependent children of migrants. If they are included we find that migrants make up just over 10% of the total UK population. Since their contribution to GDP is 9 .8%, they are not adding proportionately more to GDP, as the government claims.

4. The Prime Ministers claim regarding a 0.5% economic growth rate starts with an exaggeration. Originally the Treasury claimed that net inward migration would add 0.4% to GDP growth. This was based on the simple calculation that net migration would, in one year, add 0.4% to the working age population. It took no account of the fact that migrants and, later, their dependants add to the overall population.

5. The most recent Government Actuarys Department (GAD) principal projections (130,000 a year net migration) show that in the period 2003 2031 migration will add 5.2 million to the total population and nearly 3.8 million to the working age population. This equates to an average compound growth rates of 0.30% and 0.34% respectively. So, using the governments very simplistic basis of equating growth in GDP with the growth in the working age population, the increase in GDP each year would be 0.34% but the population would increase by 0.3% each year so the benefit per head per year would be about 0.04%. This works out at 7 per head per year or 14p a week. This calculation takes no account of additional infrastructure costs nor of the costs of congestion to which immigration on this scale will add considerably.

6. On other occasions government spokesmen have claimed that approximately 15% of the UKs trend growth depends on migration. In fact, trend growth is about 2.75%. Since 0.4 is 15% of 2.75, this amounts to the same claim, expressed differently to make it look more impressive.

7. We note that major economic studies in the US, Canada and Holland have also come to the conclusion that the benefit, in terms of GDP per head, to the indigenous population is very small. Indeed, a Dutch government study published in 2003 found that:

The Gross Domestic Product will increase, but this increase will accrue largely to the immigrants in the form of wages. The overall net gain in income of residents is likely to be small and may even be negative.

8. We conclude, therefore, that neither of the governments claims stands up to examination.

Detailed analysis of the claim that migrants make up 8% of the population and contribute 10% to GDP

9. The government repeatedly claim that, although migrants make up only 8% of the population, they contribute 10% towards the UKs Gross Domestic Product (GDP). For instance the Home Office issued a press statement in November, 2003, which anticipated a major speech on migration by the Home Secretary, David Blunkett, stating: Effectively managed legal migration is vital to Britains economic and social interests, the Home Secretary, David Blunkett, will argue in a speech this evening. Legal migrants make up eight per cent of the UKs population but generate 10 per cent of our gross domestic product.[1] The claim was repeated in the House of Lords on 23 Feb 2004 by Baroness Scotland of Asthal who said: Migrants make up 8 per cent of the population but generate 10 per cent of our total wealth. Migrants are substantial net contributors to the Exchequer[2].

10. Migration Watch has tracked down the provenance of this claim to a parliamentary answer provided by Ruth Kelly as follows: Mr Allen: To ask the Chancellor of the Exchequer if he will estimate the contribution made by immigrants to United Kingdom GDP since 1972; and if he will make a statement. [54513] Ruth Kelly (Financial Secretary to the Treasury): There are no direct statistics on the proportion of UK GDP accounted for by immigrants, and the data required to produce indirect estimates do not extend back to 1972. However, in 2001 foreign-born people accounted for around 8 per cent. of total UK employment, and the average gross weekly pay of foreign-born employees was about 16 per cent. above the average for all employees. On this basis, and assuming that relative wage rates proxy relative labour productivity, it can be estimated that the foreign-born population accounted for around 10 per cent. of UK GDP in 2001. [3]

11. This is, of course, a very broad brush approach. First, the definition of migrants which the government has used for this purpose runs much wider than recent, or current, immigrants. It includes anyone who is foreign born (including people born overseas of British citizens). Secondly, government data clearly shows that the performance of immigrants from different regions of the world varies widely in terms of both average earnings and labour market participation rates. Thirdly, employee compensation (earnings) made up only 56% of GDP measured on an income basis in 2003 [4]. Nevertheless, we have examined the claim that immigrants make up 8% of the population but contribute 10% of GDP on the basis of the governments methodology.

12. Looking first at the 10% of GDP part of the claim:

  a) The percentage of the UK working age population (defined as people between the age of 16 and state retirement age 60 for women and 65 for men) who were foreign born in 2001 was 10.11%[5] (3.65 million people).
  b) The labour market participation rate of the foreign born was about 63.6% in 2001[6] so the total number of foreign-born people who were working was about 2.32 million.
  c) The percentage of the UK working age population who were born in the UK was 89.89% (32.47 million people) and their labour market participation rate was approximately 74.6% in 2001[7] so the total number of UK-born working age population in work was about 24.22 million.
  d) The foreign born therefore make up about 8.74% of the working population (2.32 million out of a total of 26.54 million workers).
  e) Sources differ as to how average earnings of those in employment vary between the UK-born and the foreign-born. In the document Migrants in the UK, by Russell Haque of the Department for Work and Pensions, migrants are calculated to have earnings of 19% above the UK-born population [8]. By contrast, the Home Office document The migrant population in the UK: fiscal effects[9] states that Data comparing the wages of UK-born residents and migrants in employment suggests that, overall, migrants perform better than the UK-born in aggregate migrants receive 12 per cent more in wage income but it is clear that this result disguises highly varied performance within the labour market. Ruth Kelly uses a differential of 16% (see paragraph 10 above). In view of this confusion, Migration Watch commissioned an analysis of gross weekly earnings for the foreign-born and UK-born population from the ONS Labour Force Survey. This indicated that the average gross weekly pay of the UK-born (in their main job) was 415 in 2002 compared with 471 for the foreign-born a differential of 13.5%.
  f) Based on this data total earnings of foreign born migrants would be 56.8 billion (471*52*2.32 million) and of the UK-born 522.7 billion making total earnings of 579.5 billion. The foreign born therefore contribute about 9.8% to the UKs GDP on this basis and this part of the governments claim is correct.

13. The other part of the governments claim was that the foreign born make up 8% of the population. The actual figure is 8.34% - so at first sight this part of their claim also looks broadly correct. However, the following breakdown of the UK-born and foreign-born population by age is revealing:

The key discrepancy is that the percentage of the population who are children is 8% for foreign born but 21% for the UK born. This is because children born in the UK of foreign-born parents are counted in the UK population. This seriously distorts the governments statistics. This was recognised in the Home Office paper dealing with fiscal effects of migration[10] which noted that migrants receipts of family-related benefits, and thus fertility rates, are high for the foreign-born relative to the UK-born population (take up of child benefit is virtually universal). It goes on to include family-related benefits against the fiscal costs of the migrant population. In fact to do otherwise would be a nonsense. We can illustrate this by considering two families, one with two foreign-born parents, the other with two UK-born parents and each family having two UK-born dependent children. If each family had identical incomes we would expect them both to contribute equally to GDP per head. However, the governments calculation would show the UK-born population as 6 and the foreign born-population as 2 and the governments calculation would lead us to believe that, across these two families, the foreign born make 3 times the contribution to GDP per head as the UK-born.

14. There is no readily available data to tell us the number of dependent children of foreign-born parents. So Migration Watch commissioned a further analysis from the ONS, Labour Force Survey to give us this figure. This indicates that when dependent children are included the foreign born population make up 10.05% of the total UK-born population (see appendix A)

15. So, a more reasonable comparison would be that 10.05% of the population who are foreign-born (including their UK-born minor dependants) produce about 9.8% of the UKs GDP. Put another way, there is no basis for the governments claim.

Detailed analysis of the claim that GDP would reduce by 0.5% for the next two years if net migration were to cease

16. In his first major speech on immigration (to the Confederation of British Industry (CBI). on 27 April 2004) the Prime Ministers main economic justification for the present record levels of immigration was a Treasury assessment which concluded that the UKs economic growth rate would be some 0.5% lower for the next two years if net immigration were to cease.

17. On examination, it turns out that this claim is based on an extraordinarily crude calculation. Furthermore, the calculation is not supported by the evidence.

18. On 29 May 2004, in response to a Parliamentary Question, Lord McIntosh of Haringey explained the basis for this calculation as follows: The basis of the Prime Ministers statement lies in the Treasurys trend growth assessment, as explained in recent pre-Budget and Budget reports and the paper Trend Growth: Recent Development and Prospects, HM Treasury 2002. Net inward migration contributes to economic growth by adding to the population of working age and hence labour supply. Treasury assumptions about growth in the population of working age are informed by recent data and the Government Actuarys Department projections. Latest data of net inward migration show that it contributed 0.4 percentage points to growth in the population of working age between 2001 and 2002.

19. So, the Prime Ministers claim first of all conveniently rounds up the Treasurys 0.4% calculation to 0.5% - a 25% increase. Secondly, it is not, in itself, a meaningful measure. Migrants, if they are involved in any economic activity, will add to GDP but it is GDP per head that is the relevant measure. India has a larger overall GDP than Denmark[11] but no one would claim that the average Indian is better off than the average Dane.

20. On other occasions government spokesmen have claimed that approximately 15% of the UKs trend growth depends on migration. In fact, trend growth is about 2.75%. Since 0.4 is 15% of 2.75, this amounts to the same claim, expressed differently to make it look more impressive.

21. Both versions of this claim rely on the premise that net migration of people of working age will cause GDP to grow in line with the increase in working age population. The nature of net migration into the UK is likely to mean that this is not the case. In particular two groups, asylum seekers and family formation migrants, who have contributed significantly (we estimate about two-thirds) to the net inwards flow of migrants to the UK have substantial sub-groups whose contribution to the UK economy will be small or nil. In the case of asylum seekers, those awaiting their asylum decisions and those whose asylum claims have failed but who are still in the UK are not allowed to work. The family formation category contains a significant number of female spouses from the Indian Sub-Continent who typically have a much lower employment rate than their fellow citizens and lower average earnings when they do work. Thus the premise that net migration of working age migrants adds to GDP in proportion to their addition to the working age population is highly unlikely.

22. The next flaw in the argument is that earnings are a proxy for overall GDP. As noted in paragraph 11 above employee compensation (earnings) made up only 56% of GDP measured on an income basis in 2003 [12].

23. Even if we put aside these doubts about the use of the growth in the working age population as a measure of the growth in GDP, would the Treasury claim of a 0.4% rise in GDP through net migration be valid?

24. In 2002, of the 153,400 net migrants into the UK, 140,800 were of working age. The corresponding total and working age populations of the UK were 59,229,000 and 36,565,000 respectively. Net migration therefore represented an addition to the total population of 0.26% and an addition to the working age population of 0.39%.

25. However, this calculation takes no account of the fact that migrants themselves age and have children therefore adding to the dependant population. A much more accurate measure is provided by the Government Actuarys latest population projections, which project a net migration rate of 130,000 p.a. (as their principal projection). Comparing this projection with the natural change (no net immigration) projection shows that in the period from mid-2003 to mid-2031 migrants will add 5.2 million to the overall population and 3.8 million to the working age population. This equates to annual compound growth rates of 0.3% and 0.34% respectively [13].

26. If one accepts the Treasurys argument that an increase in the working age population gives a corresponding increase in GDP, the average annual growth in GDP through net migration will be 0.34% but net migration will cause an average annual growth in the overall population of 0.30% p.a.. In terms of GDP per head therefore this comparison would give a compound growth through migration of 0.34% minus 0.3%, that is 0.04%. This equates to about 440 million[14] each year or approximately 7 per head of population 14p a week.

27. These results are in line with economic studies carried out in the US, Canada and Holland which also conclude that the benefit in terms of GDP per head, to the indigenous population is very small. Indeed, a Dutch government study published in 2003 found that:

The Gross Domestic Product will increase, but this increase will accrue largely to the immigrants in the form of wages. The overall net gain in income of residents is likely to be small and may even be negative.

Further details of these studies are contained in Appendix B.

Appendix A

We requested that the ONS provided us with an estimate of the number of children (under 16) in 2001 split between:

Those whose parents were both foreign-born
Those whose parents were both UK-born
Those whose had a foreign-born and UK-born parent.

For single parent households we requested that they should be classified under the origin of their parent.

ONS provided data for the Spring and Autumn quarters of 2001 which showed:

Both
parents
UK-born
 
Two-parent households
 
One parent
UK - born - the
other foreign-born
Both parents foreign
- born
No. of children - Spring quarter
7,396,244
762,951
720,289
No. of children - Autumn quarter
7,333,599
721,339
736,902
Average over the two quarters
7,364,921
742,145
728,595
Percentage of total
children
83.4%
8.4%
8.2%
 
Single parent households
  UK-born Foreign-born
No.of children Spring quarter 2,346,762 250,888
No. of children Autumn quarter 2,403,635 273,454
Average over the two quarters 2,375,198 262,171
Percentage of total children 90.1% 9.9%

Over all families, and apportioning those with foreign-born and UK-born parents between the two categories, the percentage of dependent children who have foreign-born parents is therefore 11.9%.

The 2001 census gave the following figures for the UK-born and foreign-born under 16 populations and total populations:

  Total UK-born Foreign-Born
       
0-15 11,858,857 11,458,261 400,596
Total 58,789,199 53,892,620 4,896,579

Adjusting this table by applying the percentages calculated above gives the following results:

  Total UK-born Foreign-born (including UK- born dependant children)
Dependent children 11,858,857 10,447,654 1,411,204
Total 58,789,199 52,882,013 5,907,187
Percentage of total 100% 89.95% 10.05%

Appendix B

The Findings of other studies

1. In terms of wealth per head, all serious studies have found that the effect of large scale immigration is minimal.

2. In the United States a study by the National Research Council in 1997 found that the benefit of immigration added about one tenth of 1% of GDP per head per year. In the period studied, immigration to the US was comparable to that which is now taking place in the UK.

3. The United States government committee which commissioned the report recommended that immigration rates to the US should be halved (but the Clinton administration did not act on it).

4. In Canada the Economic Council of Canada reported in 1991. They stated that: "A historical perspective gives little or no support to the view that immigration is needed for economic prosperity. In the 19th and early 20th centuries, the fastest growth in per capita real incomes occurred at times when net immigration was nil or negative. Later in the 20th century, the opposite linkage is seen but, clearly, there is no long-term correlation." (page 19)

They continued:
"With respect to per capita disposable incomes, an increase in immigration has a positive effect, but it is very small." (page 131)

5. In the United Kingdom there have been no major studies, partly because relevant statistics are hard to obtain. However, Professor Mark Kleinman, writing in the Political Quarterly [15] concluded the following:

"In broad terms, the economic impact of migration is positive for the destination country; but the size of the impact is not great, and there are distributional consequences to consider. . the economic arguments alone will not be (and should not be) decisive."

6. The Netherlands Bureau for Economic Policy Analysis, part of the Ministry of Economic Affairs, has produced a wide-ranging study of the impact of immigration on the economy of the Netherlands. The web site is www.cpb.nl

7. The main results confirm findings in the US, Canada, and the UK that the benefit of large scale immigration to the resident population is very small and can sometimes be negative. The study, published in June 2003, concluded that immigration of labour has the following effects:

a) the gross domestic product will increase, but this increase will accrue largely to the immigrants in the form of wages;
b) the overall net gain in income of residents is likely to be small and maybe even negative;
c) the amount of redistribution between residents is substantial;
d) the more the skill distribution of immigrants differs from that of residents, the larger the amount of redistribution will be;
e) residents will skills comparable to those of immigrants will lose;
f) residents will skills complementary to those of immigrants will win in the long run;
g) capital owners will win in the short run, but in the long run their gains will disappear;
h) due to labour market imperfections, part of the income effect for resident workers will be replaced by employment effects (unemployment instead of a wage decrease).


Footnotes

  1. Home Office Press Release Reference 309/2003 12/11/2003
  2. Baroness Scotland of Asthal speaking in House of Lords on 23 Feb 2004. Lords Hansard 23/02/2004 Column 35.
  3. Commons Hansard 7 May 2002 Column 33W
  4. Source: Office for National Statistics: National accounts: GDP by category of income.
  5. Source: 2001 census combined results for England and Wales, Scotland and Northern Ireland.
  6. Source: Migrants in the UK: by Russell Haque: Department for Work and Pensions, December 2002
  7. Show -6 more...
  1. Home Office Press Release Reference 309/2003 12/11/2003
  2. Baroness Scotland of Asthal speaking in House of Lords on 23 Feb 2004. Lords Hansard 23/02/2004 Column 35.
  3. Commons Hansard 7 May 2002 Column 33W
  4. Source: Office for National Statistics: National accounts: GDP by category of income.
  5. Source: 2001 census combined results for England and Wales, Scotland and Northern Ireland.
  6. Source: Migrants in the UK: by Russell Haque: Department for Work and Pensions, December 2002
  7. Ibid
  8. Reference as footnote 7 paragraph 3.8
  9. The migrant population in the UK: fiscal effects Ceri Gott and Karl Johnston RDS occasional paper No.77.
  10. The migrant population in the UK: fiscal effects Ceri Gott and Karl Johnston RDS occasional paper No.77.
  11. GDP of India in 2003 = $598 billion; GDP of Denmark in 2003 = $212 billion Source: World Bank at www.worldbank.org/data/databytopic/GDP.pdf. GDP per capita = $570 and $39,260 respectively based on populations of 1.05 billion and 5.4 million respectively (Source of population data US census bureau).
  12. Source: Office for National Statistics: National accounts: GDP by category of income.
  13. The principal population projection shows the population rising from 59.554 million in 2003 to 65.700 million in 2031 based on net migration of 130,000 a year a rise of 6.146 m. The natural change (no migration) projection shows the population rising to 60.519m a rise of 0.965m. The rise due to migration is therefore 5.181m a compound annual growth rate of just over 0.3%. The corresponding comparison for the working age population shows that migration will cause this to grow by 3.793 million on a base population of 38.432 million an annual growth rate of 0.34%. (The base working age population has been calculated as the number of people between ages 16 and 64 inclusive to take out the effect of the change to female retirement age in the next decade)
  14. Based on total GDP of 1,099 in 2003 Source ONS. Total GDP at market prices.
  15. The Politics of Migration, edited by Sarah Spencer, published in 2003 by Blackwell. Page 59