Faulty figures or misinterpretation? EU migration predictions under Labour

"You were predicting figures of between 5,000 and 13,000 immigrants a year from the expansion of the EU in 2004 and actually something like 400,000 people came in" — Jeremy Paxman, 26 March 2015

Labour has been repeatedly criticised in the last decade for its handling of the expansion of the EU in 2004. Net migration from most of the countries joining that year was initially estimated at an average of 50,000 from 2004 to 2010, although these may significantly understate the actual numbers due to more recent revisions in the data.

It was supposedly estimated that fewer than 13,000 would come over the same period.

This criticism has been based on the assumption that the report which gave Labour its estimate was accounting for all possible scenarios.

But the report was actually based on the crucial assumption that few or no EU countries would temporarily restrict the migratory rights of new EU citizens, when the vast majority ended up doing so.

The EU's enlargement 

Back in 2004, the EU was set to expand from 15 states to 25. It was expected that most (if not all) of the EU's current states would open their borders and not adopt a seven year transitional period. This period would have phased-in the ability of migrants from eight of the ten new EU countries, including Poland, to freely move to other EU states for work.

In this context, it was estimated that net migration of EU work migrants would be between 5,000 and 13,000 annually until 2010 from these eight countries in an influential report published by the Home Office.

But most EU states did adopt the transitional period, with the UK only being joined by Ireland and Sweden in opening its borders fully to immigrant EU workers. This meant migrants who may have intended to go to Germany or France for work could have come to the UK, dramatically outstripping the prediction.

The author's take

While Labour has been criticised for relying on this 5,000 to 13,000 estimate, one of the academics behind the report has disputed the criticism.

When we got in touch with Professor Dustmann, he told us that the report "was commissioned and written to cover only the case in which all countries opened their borders".

There was a paragraph in the report which dealt with the eventuality that Germany introduced a transitional period. This section stated that only a "small fraction" of those who had intended to migrate to Germany would come to Britain as "the UK is not a very popular migration destination".

While his calculation wasn't in the report, Dustmann has subsequently claimed that a rough guide for working out how many migrants from the new EU states would come to the UK was in this paragraph.

Dustmann clarified for us that he reached his rough 46,000 estimate when combining the top end estimate for the UK - 13,000 - with a third of the average annual new EU migration to Germany - 33,000 - as it was estimated that about a third of those intending to migrate to Germany could come to the UK.

Is it close?

This 46,000 annual net migration prediction is indeed close to the ONS official estimate of 50,000 for eight of the ten states to join the EU in 2004 (the A8, analysing between December 2004 and December 2010).

However due to revisions in the numbers, the actual level of net migration from those countries is likely to have been significantly higher than the estimates available, so even this estimate looks too low.

But as Professor Dustmann told us, "the figure of a third was a ball-park suggestion, not in any sense a precise figure".

Update (20 April 2015)

Net migration estimates from 2001 to 2011 were revised upwards following a review by the ONS. While it isn't possible to apply these revisions to migration from the A8 countries specifically, we've added in a caveat that the current published figures suggesting 50,000 net migration a year are likely to be underestimates.