The missing half million - why do UK and Polish migration statistics differ so greatly?
In a Today programme interview Home Office minister, Phil Woolas, insisted that half of the estimated 1.5 million Eastern European migrants who have come to the UK since 2004 have returned home, despite evidence from Poland which suggests this is an exaggerated figure.
Speaking on the same programme Professor Krystyna Iglicka, from the Centre for International Affairs in Warsaw, said their 2008 statistics showed only a very small decline in the number of Poles abroad. Professor Iglicka also cited Polish government statistics which showed only 22,000 Poles returned from abroad to claim benefits. Although reluctant to estimate the number of Poles still in the UK she rejected the view that several hundred thousand had returned.
There appeared to be a wide discrepency between the two findings so Full Fact decided to investigate how the UK conclusions were reached.
Background - two reports; one methodology
The UK does not currently keep comprehensive figures of the number of foreign nationals entering and exiting the country. And as Phil Woolas himself admitted, the figures used to make estimates are "not an exact science". When eight new Eastern European countries (referred to as the Accession Eight or A8 for short) were admitted to the EU in May 2004 the UK government, unlike several others in Western Europe, decided against restricting freedom of movement.
Government predictions that only around 15,000 a year would come to the UK from countries such as Poland and Lithuania turned out to be massive underestimation as several hundred thousand took the opportunity to find work in the UK economy.
However, in April 2008 the Institute of Public Policy Research (IPPR) think-tank released a report entitled 'Floodgates or Turnstiles', four years on from EU expansion, claiming that over half of the estimated one million Eastern Europeans who had come to the UK had returned home.
Earlier this month the Migration Policy Institute made a more up to date assessment. It estimated the number of eastern Europeans who had arrived in the UK at 1.5 million and stated the view that 800,000 had returned home. 'The UK's New Europeans' report was commissioned by the Equality and Human Rights Commission and used the IPPRs methods to make its estimations (see footnotes 7 and 8 in the report).
Floodgates or Turnstiles - looking at IPPR's methodology
By looking more closely at how the IPPR made its original calculations we may get closer to understanding the large difference in numbers with Polish sources.
The IPPR estimate the number of Eastern Europeans that have left based on a comparison between data from the government's Worker Registration Scheme (WRS) and the Labour Force Survey (LFS) which are explained below.
The Workers Registration Scheme
The WRS was set up in 2004 and continues to operate today. With some exceptions, those A8 nationals wishing to work in the UK have to register first. One important exception includes the self-employed who are not picked up by this measure.
The Home Office statistics therefore give a good minimum of how many people have come to the UK. Given that it is a voluntary scheme and several people may choose not to register, the IPPR consider the WRS is likely to underestimate the true number of people who arrived in the UK by around a third.
Because of this, IPPR uprate the total WRS registrations to December 2007 - around 770,000 - to an estimate of just over one million (1,018,400).
The Labour Force Survey
The Labour Force Survey is a quarterly survey of 60,000 households and gives a detailed picture of the changing nature of the UK workforce. The IPPR report acknowledged a series of limitations in using the survey data. As it is concerned with employment it only considers those over 16 and under 64. As a private household survey it does not include other forms of accommodation in which new migrants are especially likely to live such as dormitories or temporary accommodation. The IPPR report further acknowledged that "response rates tend to be lower for minority groups".
Measuring outflows
Only the International Passenger Survey (IPS) directly measures outflows of migrants. However, it only counts migrants as those who intend to stay in the country for more than a year and will therefore not record short term migration flows. IPPR does not use it in calculating their figures.
Instead, the IPPR compared the LFS figures against the numbers from the registration scheme (see Table 4). So, in 2004 estimated total arrivals are placed at 167,420. The last 2007 LFS survey recorded that only 83,000 A8 nationals reported that they arrived in the UK in 2004. Repeating this method over four years suggests that only about half of the A8 nationals in the UK were picked up by the workforce survey.
Although there is clearly a discrepancy between numbers registered to work in the UK and those measured by the LFS it does not follow that all of the missing figure have left the country.
Without any accurate way of measuring outflows from the country it would be equally valid to conclude something like "only about half of the eastern Europeans in the UK are picked up by the LFS".
A 2006 interdepartmental taskforce report on the quality of migration statistics identified a number of problems with the datasets, particularly on emigration. It stated that "migration information for outflows is not readily available from any of the major sources apart from the International Passenger Survey (IPS)."
Yet the IPPR report's authors conclude that the discrepancy between the numbers of A8 nationals in the two datasets mean that "a significant proportion" have left the country. It is through this methodology that the IPPR arrive at the figure of 541,000 migrants who are not picked up by the LFS. This comparison forms the basis for their conclusion that "half of migrants from new EU countries have now left UK"- the headline of their press release.
Surprisingly, the IPPR calculations give no consideration to the limitations of the LFS. An ONS statistician who declined to be named told Full Fact that the LFS "cannot be used to measure outflows". Similarly, the interdepartmental taskforce noted the limitations of the Labour Force Survey as it "only covers long term migrants sampled from private households". It is certainly likely that the LFS is an underestimation of the true numbers, in the same way that the WRS is acknowledged to be.
The IPPR press release stated that these were "ballpark estimations" although these estimations gave the report its media headlines and political significance.
IPPR defended its methodology when asked on the Today programme. Their Director of Strategic Communications Tim Finch said all estimates came with a "health warning" but insisted that their methodology was "about as robust as you could get it".
Full Fact keenly awaits the comments of its readers and hopes you will be able to investigate further.