1) Emigration: The only source of information is a survey
In order to estimate the level of net-migration, one needs information on both the inflows (immigration) and outflows (emigration) of migrants. Data on emigration is particularly scarce. The International Passenger Survey (IPS) conducted by the Office for National Statistics (ONS) is the only source of emigration data, and is used in ONS estimates of net-migration. Since these estimates are the official statistics targeted by government policy, limitations on this data source are a matter of critical policy importance.
There has been considerable debate about the accuracy of IPS measures of total emigration, and there is no reliable alternative source of data against which to check its results. The IPS does not count all entries and exits, but rather interviews a sample of international passengers. It now interviews about 2000 or more emigrants annually, though prior to improvements in 2007 its emigration estimates were based on less than one thousand annual interviews. Like any survey, IPS comes with inherent uncertainty: it can only produce estimates with margins of error rather than pinpoint numbers. In 2009, IPS estimated 337,500 emigrants, but, considering its margin of error, this only means that we can be 95% confident that the real level of emigration was between 315,800 and 359,100, a range of over 43,000 (see Figure 1 below). Official net-migration figures are therefore only estimates as well, and subject to margins of error from both the inflow and outflow estimates.
In addition to margins of error, surveys also face biases that may come from numerous sources, including people refusing to be interviewed and obstacles to obtaining a truly random sample. Thus, IPS estimates depend on a series of calculations of “weights” designed to extrapolate from a few thousand survey interviews to a larger population.
Also, IPS identification of emigrants depends on people’s self-reported intentions about for how long they plan to leave the UK, which may or may not materialize. This question is critical because only people departing for at least one year meet the official definition of a migrant. The ONS attempts to adjust for people who change plans, but again there is no alternative source of information to help confirm whether its estimates of “switchers” are correct.
Another important limitation is that IPS emigration data do not allow for calculation of net-migration by categories of migrants (e.g. workers, students, family migrants). This is because when IPS collects information on migrants leaving the UK, it does not gather information about the characteristics and purpose of entry of these same migrants at the time of their initial arrival to the UK. For instance, it is possible for someone to arrive as a foreign citizen, naturalise and then leave as a UK citizen. This would exaggerate the contributions of non-British nationals to net-migration, as the same individual would raise net-migration among foreign nationals upon arrival and reduce net-migration among British nationals upon departure. In the same way, some migrants arrive in the UK for the purpose of formal study, but leave for employment elsewhere, exaggerating net-migration in the formal study category and taking away from net-migration in the work category. In early 2012, the ONS introduced a new question to the IPS which asks people who are leaving the UK about the purpose of their previous entry. The answers to this new question will provide the basis for estimates of student-net migration. The first annual data will become available in late 2013.
Figure 1
