3. Net migration and immigration: An overview
Net migration is the difference between immigration (i.e. people moving to the UK for more than one year) and emigration (i.e. people leaving the UK for more than one year). Overall net migration increased from less than 100,000 in the mid-1990s to a preliminary estimate of 216,000 in 2011.
However, for the last eight years the negative net migration of British citizens (i.e. more British citizens leaving than coming to the UK) has roughly offset the positive net migration of other (i.e. non-British) EU citizens. As shown in Figure 3 below, in most years since the mid-2000s, the magnitude of positive net migration of non-British EU nationals has been very similar to the magnitude of negative net migration of British nationals.
During 2004-2011, total net migration of EU citizens (i.e. combining net migration of British and other EU nationals) was 34,000 (i.e. the difference between –660,000 British net migration and +694,000 other EU net migration). This can be seen in Figure 3 below. Overall, net migration of non-EU nationals constituted 95% of total net migration to the UK in 2011 (205,000).
There is considerable uncertainty related to the measures of immigration to and emigration from the UK. This creates considerable uncertainty about the net migration estimates upon which government policy is currently focused. The International Passenger Survey (IPS) is the main data source for the ONS estimates of Long-Term International Migration (LTIM). The survey is large but the international migration estimates are extrapolated from a few thousand interviews, and are associated with a substantial margin of error. For example, while the ONS preliminary estimate of net migration for 2011 is 216,000, the reported error associated with this estimate was +/- 35,000, making it possible to create a 95% confidence interval around a range of estimates between 181,000 to 251,000. That means, roughly, that there is a 95% probability that net migration for 2011 was somewhere in that range.
The data from the IPS is combined with information from various other sources to provide LTIM estimates. As shown in Figure 4, according to LTIM estimates, total immigration increased from about 300,000 in the early and mid-1990s to over 500,000 in the mid and late 2000s. While the increase in overall immigration in the late 1990s and early 2000s was caused by a rise in non-EU immigration (with immigration of British and other EU nationals remaining flat during that period), the rise in the mid-2000s was caused by a sharp increase in immigration of EU nationals. Since 2004, immigration flows of EU (including British) and non-EU nationals have remained relatively flat at just under 600,000. Over the past few years, about half of all immigration has been inflows of non-EU nationals.
Immigration from outside the EU – over which the government has more control – can be broken down into four main channels: work, family, study and asylum. Figure 5 below, which uses data from the International Passenger Survey, shows that students constitute by far the largest group among all non-EU immigrants coming to the UK (60% in recent years). The rapid increase in non-EU student migration (up from 50,000 in the mid-1990s to about 180,000 in 2011) also accounts for almost all of the increase in immigration of non-EU nationals over the past twenty years. Labour immigration and family immigration both increased between the mid-1990s and mid-2000s, but have since been stable (family) or on a declining trajectory (labour). The magnitudes of immigration for work and family purposes in 2011 were very close to those prevailing in the late 1990s.
Not all migration is permanent, and a proportion of the migrants coming to the UK in a given year will leave again after a few years. Consequently, a reduction in immigration today will eventually lead to a reduction in emigration in the subsequent years, while an increase in immigration today will subsequently lead to an increase in emigration. We call this the 'net migration bounce effect'.
For example, as shown in Figure 5 above, the immigration of students from outside the EU increased significantly since 2008. The rising number of students coming to the UK implies that in the longer term, as this new larger set of international students come to the end of their studies, many will return to their countries of origin or move elsewhere, and therefore contribute to a rising number of people leaving the UK.
So it is very likely that the emigration of students – which has not been directly measured, due to limitations of past IPS questionnaires - has been rising and can be expected to continue to increase in the coming years, which will help reduce net migration. This means that, in the hypothetical scenario of no change in immigration of students next year, net migration of students would start to decline.
Conversely, any decrease in immigration will lower net migration in the short term, but this reduction will be partially reversed in the longer-term due to lower emigration (i.e. fewer people leaving). Figure 6 below illustrates the difference between the short-term and longer-term effects of cutting immigration on net migration. By short-term effect we mean the effect of reducing immigration on net migration, without considering changes to emigration. By long-term effect we mean the combined effect of the initial reduction in inflows (which will reduce net migration) and the subsequent reduction in outflows (which will increase net migration). In our example, the long-term emigration effect begins in the year after the reductions in immigration are made, but increases with passing years.
Figure 6 models the short-term and long-term effects of a hypothetical reduction of immigration of just under 150,000 over three years (for a more detailed discussion of the methodology, see our commentary on the Net Migration Bounce). It shows that, in this scenario, the long-term reduction in net migration (i.e. taking account of lower emigration) will be less than half of the initial reduction achieved because of lower inflows.
Students, while by far the largest group, are proportionally unlikely to settle permanently – the Home Office study The Migrant Journey showed that 21% of students who arrived in 2004 still had leave to remain in 2009. This is an imperfect measure, since it is based on only a single cohort, and also because some unknown proportion may have overstayed their visas and remained in Britain without legal leave to remain, but it is the best means available to estimate the rate at which particular types of migrants stay in Britain for at least five years.
The impact of reducing the number of family migrants coming to the UK – while smaller in the short term – is proportionally greater in the long term than cutting the number of students. This is because family migrants are considerably more likely to settle (63 per cent of those who came in 2004 were still in the UK in 2009).
The impact on net migration of reducing labour migration is highly dependent on whether reductions are to Tier 5 (only 11% of whom are estimated to remain after 5 years) or to Tiers 1 and 2 (40% of whom are estimated to remain after 5 years.) But the majority of labour migrants do tend to leave rather than stay permanently, meaning that cuts to labour immigration will also push down emigration in the long term.
Implications for stabilising the UK population at or under 70 million: for the government to reduce net migration to a level that would stabilise the UK population below 70 million, cuts would need to be made to immigration of non-EU citizens that take into consideration the “net migration bounce”: the future decline of emigration resulting from present cuts in immigration. This means that deeper reductions of the immigration of more transient groups (such as students) are needed to deliver long-term results that equal those of reducing the inflows of less transient groups (such as family).