Untangling the net: Understanding why migrants come and go

This increase – 23,000 on the previous quarter, the year to September 2012 – is the first time net migration has increased after falling for five consecutive quarters, and suggests an interruption of progress in the government’s plans to cut net migration to the ‘tens of thousands’ by 2015, although the change is not a statistically significant rise. However, the figure for net migration is still 39,000 lower than the same estimate for 2011 and 76,000 lower than 2010.

A particular challenge for the government is that declining emigration – which fell by more than 20,000 – is the primary cause of the recent increase in net migration. This was mainly among British nationals, and the government cannot directly control the number of British people who choose to leave. Immigration increased by only 2,000 in the most recent quarterly update (measuring the change from the year to September 2012 to the year to December 2012).

Emigration by reason for immigration: Why does it matter?

But while this top-line number is undoubtedly important, today’s data also provide another important set of figures for the first time, which are set to have important ramifications for the UK’s migration policy debate: emigrants, grouped by their original main reason for coming to the UK.

This new data series will make it possible, in time, to estimate the long term contribution to net migration of students, workers and family members separately. This will alleviate what has been an important gap in UK migration statistics, one that has hampered efforts to create targeted policies to reduce overall net migration. Study, work, and family are not just a few of many ways that ONS classifies migrants; they are also the key categories in which immigration policy has been made. The government has addressed each of these categories or routes separately in policies designed to reduce net migration to the ‘tens of thousands’. In setting targets for its policies within each category, the government has had to rely on immigration figures, without emigration figures to match.

Understanding the data limitation

To understand how this data limitation was created and will be alleviated in time, one needs to consider how the UK’s migration data is collected:

The International Passenger Survey (IPS) provides the key data for the government’s net migration figures by interviewing about 800,000 people every year as they come and go through the UK’s airports and other points of entry/exit. Of these, some 5,000 are international migrants, either arriving in the UK to live (immigration – about 3,000), or leaving the UK to live elsewhere (emigration – about 2,000).

The IPS has traditionally asked those people emigrating from the UK why they are leaving – for work, family or study purposes. But just over a year ago an additional question was added to the survey. It was aimed specifically at people who had migrated to the UK in the past, but were now leaving the country (i.e. “former immigrants”). The question inquired about their main reason for immigrating to the UK at the time that they came. Today’s ONS migration data release provides us with the data from that question for the first time.

Until the IPS started asking this question, significant information was missed about the movement of people to and from the UK. For example, a large proportion of people who come to the UK from another country to study will complete their course and then leave for a job or for family reasons, rather than to study again elsewhere. On arrival in the UK such migrants would have been recorded in the “formal study” category, but on departure they would be have been categorised as work or family migrants. This means that we had estimates of how many international students arrive in Britain each year, but no estimates of how many from this group departed in the following years, making it impossible to estimate the long term impact of each migration category on official net migration estimates.

The new data

So what did we learn from this new question? The ONS figures suggest that an estimated 298,000 emigrated from the UK in 2012. From these, 89,000 are “new emigrants”, that is individuals who have never lived outside the UK long enough to establish usual residence elsewhere and are therefore leaving the UK for the first time as long-term international emigrants. The new ONS statistics also tell us that an estimated 209,000 people who have formerly immigrated to the UK left the country during the year 2012 to live somewhere else (i.e. “former immigrants”). These estimates are only available based on IPS data which excludes adjustments made to derive the official Long Term International Migration (LTIM) figures. Of those 209,000  “former immigrants”:

• 68,000 came to the UK originally for work reasons
• 67,000 came to the UK originally for study reasons
• 16,000 came to the UK originally for family reasons
• 59,000 came to the UK originally for other reasons (or not stated)

This suggests that the majority of “former immigrants” who left the UK in 2012 came originally for work reasons, with study in a close second place.

Limiting the analysis to non-EU nationals leaving the UK, however, suggests a significantly different picture. Specifically, of the 95,000 “former immigrants” who are non-EU nationals and left the UK during 2012:

• 26,000 came to the UK originally for work reasons
• 49,000 came to the UK originally for study reasons
• 6,000 came to the UK originally for family reasons
• 14,000 came to the UK originally for other reasons (or not stated)

Therefore, those who came to the UK for the main purpose of study account for about 52% of all outflows of “former immigrants” who are non-EU nationals.

Remaining limitations

It is possible to subtract the numbers of those leaving for each original immigration category (i.e. work, study, family) from the number who are arriving in that immigration category to come up with a figure for net migration by reason for immigration (see Table 1 below). This is not an incorrect way of calculating a net figure, but it is only a snapshot and does not tell us about the relative contributions to net migration of different groups in the long run.

IPS data from a single year can always be subject to short-term fluctuations or random measurement errors. But more fundamentally, net migration data for 2012 reflect a single year of immigration but multiple years of past immigration. A snapshot of net migration in a given year will reflect not only inward movements and departures rates, but also levels of immigration in past years. In general, higher immigration levels among a group in a given year should lead to higher outflows among the same group in later years, but the precise relationship is unknown and could change over time. In order to see more clearly how each group of migrants contributes to net migration, we will need additional information. It will be particularly useful to see how immigration and emigration rise and fall over a period of years, and also to see data on the year of arrival of emigrants within each category.

These data are not available at this point. We cannot yet determine, for example, whether those who came for work reasons and are now leaving the UK are recent  migrants who have come to the UK for a year but have been unable to find steady work, or settled migrants who have been working in the country for decades but who have now decided to leave. Both of these scenarios – and innumerable others – could have important ramifications for government policy making.

However, end of year data for 2012, to be published by ONS in November 2013, is planned to include emigration parsed by both reason for immigration and year of arrival.

Finally, it is important to keep in mind that IPS data are gathered by means of a sample survey. In this case, the survey question asks emigrants to remember the main reason why they immigrated to Britain, perhaps a year earlier or perhaps many years earlier. This might be straightforward for some, but for other emigrants who may have had multiple reasons for coming to Britain in the first place (for example, to work and get married), it is not a simple question. Thus, we see that 59,000 out of 209,000 departing “former immigrants” did not give a reason for their initial immigration, or provided a reason that could not be classified in the typical classification scheme of work, family, and study.

Difference between the current inflow and cumulative outflow per migration category

Table 1 shows the difference between inflows to the UK under a certain migration category in 2012 and outflows in 2012 among those who came to the UK for that same reason (at some earlier point). So, the “difference” column for the student category is the difference between 2012 student inflow and 2012 outflow by those who said that they originally migrated to the UK for the main purpose of study.

These figures do not include all components of net migration: some emigrants are “new emigrants” who never immigrated to the UK in the first place, and so  cannot be incorporated into column (1) of the table.

Table 1 – Current inflow and cumulative outflow per migration category

Main purpose for immigration to the UK	2012 inflow of those coming to UK for this purpose	2012 outflow of those who said that they originally arrived for this purpose	Difference
	-1	-2	-3
(A) All nationalities
Work	175000	68000	107000
Study	175000	67000	108000
Family	61000	16000	45000
Other reasons/not stated	51000	59000	-8000
(B) EU nationals (not including British)
Work	95000	38000	57000
Study	28000	15000	13000
Family	12000	5000	7000
Other reasons/not stated	13000	10000	3000
(C) Non-EU nationals
Work	44000	26000	18000
Study	139000	49000	90000
Family	39000	6000	33000
Other reasons/not stated	13000	14000	-1000

Furthermore, these data provide only a snapshot. The “difference” column includes the 2012 departure of migrants who came over the course of a number of years for, say, formal study, while measuring only a single year of immigration among those who arrived for the purpose of study. The snapshot has heuristic value, as it allows us to understand more about the composition of net migration flows in a single year. But it combines different sets of people and is affected not only by present day activity but also by traces of the past, as it includes emigrants who arrived in the UK at many different points in time. Examining immigration and emigration by category will be most useful with additional data that includes emigrants’ time of arrival, and further analysis to reveal the long-term contribution of each group to population growth.

These data nonetheless show some interesting preliminary results. First, the outflows among student immigrants were lower than one might have expected, which suggests that students may make up a larger share of net migration than anticipated based on prior evidence. At 67,000, outflows among student immigrants are about the same as outflows among work immigrants, even though prior studies have suggested a higher rate of departure among students. We cannot draw strong conclusions from this snapshot view of the data, but this point deserves further analysis when new data arrive. If this finding holds up under further analysis, it would imply a larger role for students in long-term net migration than previous work has suggested.

Secondly, the data show a contrast between EU and non-EU migrants. The “difference” measure suggests that work immigrants may make up a larger proportion among EU nationals while student immigrants are more prevalent in non-EU migration flows.

Other ways to explore the relation between immigration and emigration by category?

At present, efforts to undertake this sort of calculation are based on a Home Office report called The Migrant Journey, which takes groups of migrants who arrived in the UK in specific years (cohorts) and looks at the legal immigration status of those migrants after each of the next five years.

The latest version of this report shows that, of the various categories of non-EU migrants who came to the UK in 2004, 2005 and 2006, family migrants are the group most likely to still have legal leave to remain in the UK after five years (66% of the 2006 group) followed by skilled work migrants (40% of the 2006 group), and then student migrants (18% of the 2006 group).

However, these data cannot be used to calculate net migration by reason for immigration. Migrant Journey data are based on visas, rather than IPS, and so are not scaled to the official estimates of net migration used for policy purposes. Moreover, the visa data do not show migrants’ actual departures from Britain, but only whether or not their legal permission to remain in the UK has expired. Finally, visa data do not cover EU nationals.

However, with additional information from the IPS, adding years over time and providing data on the actual length of stay of departing emigrants, it will be possible to apply similar calculations to official estimates of immigration and emigration. The resulting estimates of net migration by category will still be subject to the same limitations as all IPS estimates (notably, margins of error). But they will mark an advance from Migrant Journey data by providing information on actual departures from the UK rather than just the expiration of legal leave to remain.

A valuable addition to migration data

At this stage the new data provide a snapshot of emigration by reason for immigration, and one that has limited practical application. But over time this will change. Whatever the limitations of the new data, having more detailed data about emigration – broken down by original reason for immigration – is a vast improvement. Data from the new IPS question will become a rich and useful source of information to help us better understand how different groups of migrants affect overall net migration to the UK over the long-term.

On the other hand, even these improved data have limitations, and it will be important to continue to use multiple approaches to measuring immigration and emigration flows.

Definitions

• Former immigrants = those emigrating from the UK who in the past have immigrated to the UK.
• Previous main reason for immigration = refers to the main reason that a former immigrant came to the UK.
• New emigrants = those who have never lived outside the UK long enough to establish usual residence elsewhere and are therefore leaving the UK for the first time as long-term international emigrants.

Related material

• Migration Observatory report – Top Ten Problems in the Evidence Base for Public Debate and Policy-making on Immigration in the UK
• Migration Observatory commentary – Pulling students out of the net: Should they be excluded from the net migration target?
• Home Office report – Migrant Journey 3rd Report