House of Lords - Economic Affairs

Select Committee on Economic Affairs Minutes of Evidence

Memorandum by Professor Robert Rowthorn, Cambridge University

ASTRACT

Since 1997 a new UK immigration policy has displaced previous policy aims, which were focused on minimizing settlement. Large-scale immigration is now seen as essential for the UK's economic well-being, and measures have been introduced to increase inflows. The benefits claimed include fiscal advantages, increased prosperity, a ready supply of labour and improvements to the age-structure. Fears that large-scale immigration might damage the interests of unskilled workers are discounted.

This submission examines these claims. It concludes that the economic consequences of large-scale immigration are mostly minor, negative or transient, that the interests of more vulnerable sections of the domestic population may well be damaged, and that any economic benefits are unlikely to bear comparison with its substantial impact on population growth. Such findings are in line with those from other developed countries.

Although it does not benefit the UK population as a whole, large-scale immigration does benefit migrants, their families and sometimes their countries of origin. It can be argued that UK migration policy should take the interests of these other parties. This issue is not addressed in the present submission.

THE ECONOMIC IMPACT OF IMMIGRATION

Immigration has been increasing for many years, but the pace has accelerated noticeably since Labour came to power in 1997. The net inflow of non-British citizens into the UK trebled in the eight years from 1997 onwards, and there was also an increase in the net outflow of British citizens (Figure 1). In 2005, 474,000 non-British citizens entered the country as long-term immigrants and 181,000 left, making a net gain of 292,000 (after rounding).[1] In the same year, 198,000 British citizens left the country as long-term emigrants and 91,000 returned from long-term emigration making a net loss of 107,000. Many people fear that cumulated over decades, these flows will have a serious impact on the sense of national identity and historical continuity, especially when viewed against the background of increasing separatist tendencies in Wales and Scotland. Opinion polls regularly show widespread public concern about these developments. However, in this document, I restrict myself to narrower economic issues.

I argue that, taken as a whole, the large-scale immigration is of minor economic benefit to the existing population of the UK as a whole, although it is certainly of benefit to the immigrants, their families and sometimes their countries of origin. Large-scale immigration will lead to a rapid and sustained growth in population with negative economic and environmental consequences in the form of overcrowding, congestion, pressure on housing and public services, and loss of environmental amenities. It also undermines the labour market position of the most vulnerable and least skilled sections of the local workforce, including many in the ethnic minority population, who must compete against the immigrants. There are, of course, domestic beneficiaries of large-scale migration. These include employers who can obtain good workers at close to the minimum wage, or even less, and the consumers of goods and services that rely heavily on migrant labour.

1. Demographic Issues

Figure 2 shows projections of future UK population under various assumptions about migration. Three of these projections are taken directly from the Government Actuary's Department (GAD) website, and the other is my own estimate based on information provided on this website.[2] All of these projections assume the same age-specific birth and death rates and they differ only in what they assume about international migration. Details are as follows:

Natural Change.

Zero Net Migration.

Principal Projection.

High Migration.

By comparing the above projections, we can get an idea of how international migration might affect the UK population in the future. Total population under the High Migration variant (D) is about 24 million larger by 2074 than under the Natural Change variant (A). This difference is entirely due to migration. Some 14 million of the total is because more people enter the country than leave, and 10 million to the fact that the immigrants are mostly young and bear children in the UK.

It is important to recognise the role of extra births in the context of migration. Estimates by GAD imply that, at the margin, for every extra 1,000 migrants who settle in the country, the eventual population increase is around 1,400. This knock-on effect is only felt after a delay and its existence is concealed in ONS publications such as press releases and Population Trends. These focus only on the immediate impact of migration and ignore the additional children that will be born in the country as a result of migration.

Predicting immigration is difficult since long-run trends are often masked by short-run fluctuations. There has been a prolonged upward trend in immigration from the New Commonwealth (mainly Africa and South Asia), although this growth has been interrupted recently. This may be merely a blip, in which case immigration from these areas will eventually resume its upward path, or perhaps immigration from the New Commonwealth has now stabilised. Either way, net immigration from this area is still very high.[4] There was also a massive, unexpected upsurge in migration from Eastern Europe following EU enlargement. As economic conditions in Eastern Europe improve, this flow should slow down as fewer people come to this country to work and some of the migrants already here return home. However, there will eventually be a new influx of immigrants when existing restrictions on migration from Bulgaria and Romania lapse, and further down the line there may be a much greater influx if Turkey joins EU. Given the uncertainties involved, it is difficult to say with any confidence what the rate of net migration is likely to be in coming decades. However, it does seem that the figure of 145,000 assumed by GAD for its Principal Projection is on the low side. In the absence of new measures to contain the flow, it seems likely that immigration will exceed this amount, perhaps by a considerable margin. Indeed, the figure of 145,000 has been exceeded every year from 1999 onwards (Figure 3).

There is also another point to consider. The projections presented in Figure 1 assume that immigrants have the same fertility as the existing UK population. This assumption is not supported by the evidence. For example, in 2001, the total fertility rates of women born in Bangladesh and Pakistan were equal to 3.9 and 4.7, respectively, as compared to 1.6 for women born in this country.[5] Most immigrant groups do not have as many children, but even so the overall total fertility rate for women born overseas was 2.2, which is significantly higher than the figure for women born in the UK. This helps to explain why the share of live births to women born abroad has risen from 12.9% in 1995 to 21.9% in 2006.[6] When such discrepancies are taken into account, the eventual impact of a given rate of immigration may be even larger than the above projections imply.

Fertility

All of the above projections assume a total fertility rate (TFR) equal to 1.73. In fact, the birth rate has been rising in recent years and provisional figures for 2006 indicate a TFR of 1.84 for the UK as a whole.[7] If this birth rate is sustained in the future it will have a substantial impact on the population. If it is combined with a high rate of immigration, such as we have observed in the recent past, the effect will be dramatic. Figure 4 shows what happens to population with a TFR = 1.84 and net migration = 205,000. By 2074, population exceeds 81 million. This is 10 million greater than under the GAD Principal Projection, which is currently taken as the benchmark for discussions about future population in the UK.[8]

Figure 4 also shows what happens to population with a TFR = 1.84 and zero net migration. This gives some idea how population would evolve with existing birth rates but a tough immigration policy that ensured that the same number of people entered the country as left. Under this projection population grows for a time slowly for a time and then starts to fall rather slowly. By the end of the period, population is virtually the same as at the beginning. Given current birth rates, if the objective of public policy were to stabilise UK population, a policy of zero net migration would be the way to achieve it.

Age Structure

Migration and fertility affect the age-structure of the population. They are often seen as a way of rejuvenating the population and thereby sharing the cost of supporting those who are too old to work. There are many ways of measuring age-structure. Here I use a simple measure which is defined as follows:

This is a mixed measure which is mainly influenced what happens to the age-structure of the population. It is also influenced by changes in the age at which men and women become eligible for a state pension. The potential support ratio (PSR) indicates how many individuals there are of official "working-age" who are potentially available to support each person of official pension-age.

Figure 5 shows what happens to the PSR under a variety of assumptions about migration. The projections correspond to those shown for population in Figure 1. In each case the PSR falls steeply for a period of about 15 years starting around 2020, after which time the curves start to flatten out. The steep fall is an echo of the post-war baby boom which created a large bulge in the age structure. Following the baby boom the birth rate fell sharply, with the result that, as the cohort of people born in the 1950s and early 1960s start to retire, there will be relatively few people of working-age to support them. The diagram also indicates how migration affects the PSR. The most interesting feature is how little difference mass migration makes to this ratio. Under the zero net migration variant the PSR in 2074 is 1.94 and under the high migration scenario it is 2.11. To achieve this rather modest change requires adding an extra 20 million persons to the national population by 2074.

2. The Macro-Economic Benefits of Immigration

Immigration is said to be a good thing because with a bigger national labour force the country can produce more output. "We need immigration to keep the economy growing." This statement raises a number of issues. Firstly, the impact of migration on national output depends on the skills and motivation of the immigrants and on their ability to gain suitable employment without displacing local workers. Highly skilled or entrepreneurial immigrants typically contribute more than less skilled immigrants, but even the latter will contribute something to GDP provided they can find useful employment and do not put local workers on the dole. Secondly, there is the issue of who gains from immigration. The fact that immigration helps the economy to grow is of no benefit to the local population if they do not share in this extra growth. If immigrants appropriate all of the extra income that they generate, their entry will be of no economic benefit to the local population. Indeed, their presence may be a burden on the local population. For example, unskilled immigrants may earn low wages and receive extensive financial support from the government for their families in the form of public services, such as health and education, together with welfare payments and tax credits. In this case, even though they help to raise national output, they take more out from the economy than they put in. The gap is filled by taxes levied on the rest of the population.

There are two conventional ways of measuring the economic benefits which immigration confers on the local population. The first is the impact of immigration on per capita income. Does immigration increase the amount of output per head of population? The second is the impact of immigration on government finances. Do the immigrants generate a fiscal surplus, so that locals have to pay fewer taxes than before? Let us consider these in turn.

Per Capita Income

Immigration is claimed to increase per capita income in the following ways:

Increasing the absolute size of the national labour force.

Increasing the share of the population that is of working age

Special skills and capabilities.

Doing the jobs that locals will not do

Flexibility.

Some types of immigration undoubtedly raise GDP per capita, but others do not. The net effect of any particular type of migration is an empirical issue that depends on the precise mix of the various types of immigration. The National Institute has examined this issue and it finds that immigration raised GDP by 3.1% between 1998 and 2005.[11] Over the same period, net immigration of non-British citizens amounted to 3.2% of the UK population. If we take into account the children born to migrants and migrant deaths during this period, the effect of migration was to add around 3.5% to the UK population. This is greater than the increase in GDP of 3.1% which the National Institute estimates was due to immigration during the period. If the National Institute estimate is correct, the net effect of immigration over this period was therefore to reduce GDP per capita slightly. However, the effect was very small and within the margin of statistical error.

Fiscal Transfers

The opponents of immigration often present immigrants as a burden on the taxpayer. They can cite many specific examples. Many immigrants do not work or receive more from the government in the form of benefits and public services than they pay in taxes. Conversely, the supporters of immigration can cite many examples where migrants pay more in taxes than is spent on them by the government. The first systematic study of the fiscal impact of immigration in the UK was by Gott and Johnston for the Home Office.[12] This study was concerned with the fiscal contribution of the migrant population as a whole in the tax year 1999-2000. Migrants were defined as foreign-born residents and UK-born dependent children who have two parents who are foreign-born or are in lone-parent households where the head of household is foreign-born. This study was updated and slightly modified by Sriskandarajah and his colleagues at the IPPR.[13]

The IPPR study estimates that in the tax year 2003-2004 migrants paid £41.2 billion in tax and consumed £41.6 billion in benefits and state services, giving a net fiscal contribution of approximately -£0.4 billion. This calculation can be questioned on a number of grounds. Table 1 illustrates how the calculations might be adjusted to accommodate some of these criticisms.

Children of mixed parentage.

Defence.

Budget Balance.

When all of these adjustments are made the fiscal balance of migrants becomes positive and is equal to £2.6 billion (Table 1). In addition to the adjustments listed above, there are other items which should be taken into account but are difficult to quantify.[18] There is no space to discuss them here, but suffice it to say that most of them would reduce the fiscal contribution of migrants. However, it is unlikely that their inclusion would greatly alter the picture.

The above observations indicate the difficulties in obtaining an accurate picture of how immigration has affected government finances. The measurable net fiscal contribution of the migrant population lies between -£5.3 billion and +£2.6 billion. These figures may seem large in absolute terms, but they should be seen in perspective. They range between -0.47 and +0.23% of GDP, between -0.83 and +0.40% of individual consumption, and between -1.16 and +0.55% of government expenditure. In comparison to the economy as a whole, the fiscal impact of the migrant population, taken as a whole, is small.

Past immigration into the UK has not in aggregate led to a significant fiscal burden on the rest of society, nor has it provided a significant surplus. It has been broadly neutral. In this respect, Britain is similar to other advanced economies. In countries where there has been large scale immigration over a fairly long period of time, the stock of migrants and their descendants normally contains a wide spread of different types and age groups. This explains why estimates of the fiscal contribution of the immigrant population as a whole are typically quite small. The positive contribution of some migrants is largely or wholly offset by the negative contribution of others. This finding holds across a variety of countries and methodologies. Estimates of the net fiscal contribution of past immigration normally lie within the range ±1% of GDP.[19]

4. Impact on the Labour Market

In addition to its overall impact on the receiving country, immigration may also affect the operation of the labour market. The nature of this impact depends on the type of immigrant concerned. Provided their talents are fully utilised, the immigration of skilled workers or entrepreneurs will raise the demand for unskilled labour in the recipient country and will thereby benefit existing unskilled workers. The immigration of unskilled workers may have the opposite effect. If the immigrants are able to compete freely with unskilled local workers, the economic position of the latter may get worse. The competition from cheap immigrant labour may push down the wages of unskilled locals to the benefit of employers and other locals who utilise their services. Alternatively, competition from immigrant labour may deprive some locals of employment. An indication of the potential scale of this competition in the UK is given by the statistics on the operation of Eastern European migrants who are registered under the worker registration scheme. Between May 2004 and March 2007, the number of national insurance certificates granted to such migrants was 623,575 and 77% earned between £4.50 and £5.99 per hour.[20] Over this period, the statutory minimum wage was in the range £4.50 to £5.35 per hour for adults. It is hard to believe that competition on this scale has no effect on the economic prospects of local workers, as many advocates of immigration claim. Indeed, such a claim is inconsistent with the widely accepted argument that modern technology and structural change in the economy have destroyed many low-skill jobs and undermined the labour market position of low-skilled workers. If workers at the bottom end of the skill ladder are losing out because of adverse shifts in the demand for unskilled labour, then it seems obvious that their situation can only be made worse by making them compete with migrant labour. Those who point out this out are often accused of a racial preference in favour of white locals against non-white immigrants. In the case of the UK, many of the locals who compete with low-skilled migrants are from ethnic minorities, for example black Caribbeans, whereas nowadays the immigrants are often white. This situation is similar in the United States, where unskilled black workers face competition from Hispanic migrants.

The above are merely general observations. Before reviewing the evidence, it will be useful to consider briefly some of the difficulties which face researchers who seek to quantify the impact of immigration on the labour market.

Estimation

The following are some of the main difficulties:

Causality.

Induced Migration.

Measurement Error.

All of the above factors lead to a downward bias in the estimated impact of migration on the labour market. Econometricians normally seek to correct for this bias but it is often very difficult.

Findings

The following is a very brief survey of the evidence about the labour market impacts of migration in this country and abroad.

Historical Impact on the Average Worker

It is widely agreed that, taken as a whole, past immigration in most countries has not had much affect on the average wage of native workers. There is disagreement about its impact on employment. Most studies find that in aggregate the impact on native employment has been small. Many of these studies may have a downward bias because of measurement error and a failure to allow for induced migration. In their article on the impact of immigration on European employment, Angrist and Kugler avoid the latter problem because their geographical cells are individual EU countries, between which induced migration is likely to be small.[24] In some of their formulations, they find that immigration has a large and statistically significant effect on male employment. For each 100 male immigrants, they estimate that between 35 and 83 male native jobs will be lost.[25] Using Census data and the "difference in differences" method, Dustmann and his colleagues estimate that 23-60 native jobs will be lost for each 100 immigrants.[26] These estimates are also based on statistically significant coefficients. Using less accurate LFS data, they get a smaller and less statistically significant effect. The same is true in a later paper of theirs which is also based on the LFS.[27]

A more recent study by Gilpin et al gets statistically insignificant, but sometimes very large estimates for the impact of recent immigration from Central and Eastern Europe on local unemployment.[28] Most of the "long-run" coefficients are equal to at least 0.6, which implies that in the "long-run" 60 or more local workers will become unemployed for each 100 immigrants that enter the region. It must be stressed that these coefficients are not statistically significant, but this does not mean that they are "small" as the authors claim.[29] It simply means that there is too much noise to estimate them accurately.

Unskilled workers—overall impact

Economic theory would suggest that unskilled workers lose where they have to compete with immigrants, but benefit from the demand for their labour created by the immigration of entrepreneurs and highly educated workers. The overall impact of immigration depends on the relative strength of these opposing effects. In a recent report, Dustmann et al find that most types of worker in the UK have gained a modest amount from immigration, whereas workers in the bottom layer have suffered a loss.[30] The following are some extracts from the press release summarising this report:

"The research looks at the period from 1997 to 2005 and finds evidence of an overall positive impact on immigration on the wages of native born workers, although the magnitude of the effect is modest. Immigration during these years contributed about one twentieth of the average 3% annual growth in real wages ... The report goes on to say that although the arrival of economic migrants has benefited workers in the middle and upper part of the wage distribution, immigration has placed downward pressure on the wages in receipt of lower levels of pay. Over the period considered, wages at all points of the wages distribution increased in real terms, but wages in the lowest quarter would have increased quicker and wages further up the distribution would have risen more slowly if it were not for the effect of immigration."

Ottaviano and Peri reach a similar conclusion for the USA.[31]

Unskilled workers—competition with immigrants

In a well-known paper, Borjas estimates that competition from Hispanic immigrants into the USA, mainly from Mexico, has reduced the wages of unskilled native workers, especially blacks, by a substantial amount.[32] Ottaviano and Peri have criticised this finding on the grounds that unskilled natives and unskilled immigrants are not very good substitutes for each other.[33] They conclude that the effect of immigration is smaller than Borjas claims. Using a similar methodology, Manacorda et al find that the immigration of less-educated male workers has had quite a large impact on the existing less-educated male workforce in the UK.[34] However, they also find that this effect is confined mainly to previous immigrants. The native workforce is largely unaffected. Many of these earlier immigrants are, presumably, from ethnic minorities and many of them are now UK citizens. These findings would suggest that it is unskilled members of ethnic minorities who have been most affected by recent immigration.

Labour discipline

The IMF argues that globalisation through trade, investment and migration has undermined the bargaining power of workers in the advanced economies.[35] The threat of competition from workers in low wage economies has made workers in the advanced economies less militant and has reduced the need to restrain wages through unemployment. This means there is less inflationary pressure and that monetary policy can be more expansionary, creating additional jobs for local workers in the advanced economies. It is possible that the additional jobs created from a more expansionary policy will outweigh the loss of jobs to immigrants, in which case immigration will eventually cause total employment for local workers to rise. In other words, the "natural" rate of unemployment will fall.[36] This is a plausible argument. Note, however, that the disciplining effect of immigration (and other forms of globalisation) depends on the perceived threat of job loss, not on the actual loss of jobs. It is not clear how much immigration is actually needed to achieve this effect.

Source: International Migration, Series MN nos. 28 and 32, ONS.

Source: Population change: UK population increases by 349,000, ONS Press Release, 22 August 2007.

Table 1

ALTERNATIVE ESTIMATES OF THE FISCAL IMPACT OF MIGRANTS IN THE UK 2003-04


	Tax £ billion	Expenditure £ billion	Balance £ billion	% GDP	% Individual consumption

Original (IPPR)	41.2	41.6	-0.4	-0.04	-0.07
Adjustment
Children of mixed parentage		4.9
After first adjustment	41.2	46.5	-5.3	-0.47	-0.83
Adjustment
Defence		-3.0
After second adjustment	41.2	43.5	-2.3	-0.20	-0.36
Adjustment
Budget balance	4.9
After third adjustment	46.0	43.5	+2.6	+0.23	+0.40

Note: rows and columns may not add because of rounding errors.