05/04/2007

A reply from MigrationWatch!

I mentioned in 'Academic hits back - and misses in migration row' that I'd emailed MigrationWatch again with the Treasury's email that pretty much blew their 'The Impact of Immigration on GDP per head' report out of the water, because the figure MW claimed referred to immigrant contribution to GDP actually only referred to Eastern European migrants. I've now actually had a proper response.

MW have posted corrected figures on their website in the confusingly titled 'The Impact of Immigration on GDP per head' (which links to a different version of the study). But, funnily enough, they're not very good, and not very accurate. Here's how.

1. MW shift back to using total rather than net addition to the population. The NIER data they're using may well show that 200,000 migrants arrived in 2004 and 2005, but as we saw in 'MigrationWatch: Redux', that doesn't necessarily mean that 200,000 were added to the population.

2. Many of these people are only here temporarily, so don't actually add anything to the population. MW seem to have forgotten that in the very next section of the report, they quote their own study that claims 50 percent of all registered adults return to Eastern Europe. Whoops.

In all fairness to MW, the National Institute Economic Review may have made these adjustments, although that's doubtful from MW's wording - especially connected to my first point. It's incredibly difficult to check, as I'd have to shell out thirty odd quid for the report to be able to check it. Of course, my usual caveat applies - if anyone knows or can quote the relevant parts of the Review, please say so in the comments and I'll amend the post with signposts to show what I've changed.

They've also shifted from using Government figures to using the figures from the National Institute Economic Review to measure the numbers of A8 migrants - even though the Government ones are freely available and have been corrected since they made a mistake calculating the number of dependants A8 migrants bring with them - on the grounds that the numbers are unreliable. The weirdest thing about this is that the £4bn figure they're starting with is itself unreliable, as the Treasury themselves admitted to me in their email, when they said:
The Treasury has not made any official estimate of the contribution of migration; the £4bn figure is a ballpark estimate of the direct contribution of A8 migration. It is based on an internal estimate of the net inflow of A8 migrants into the country, including temporary migrants, and an estimate of their average earnings from the ONS' Labour Force Survey. The £4bn figure does not include any consideration of the economic gains from increased returns to capital or greater productivity, which are very difficult to measure.
But, of course, that doesn't stop MW from using them. Notice the bit in the middle where the Treasury says the figure is based on an internal estimate of the net inflow of A8 migrants - not the flipping National Institute Economic Review figures.

In other words, MW claim that Government figures for the number of A8 migrants are too unreliable for them to use - but the £4bn figure the Government based on those very numbers is somehow reliable. Oh dear.


Still, this revision totally invalidates the 'Mars bar' figures that appeared in all the papers earlier this year. There's no new press release that says, "You know we said that Government figures proved migrants contribute only a small Mars bar a month? Well, that was wrong."

There's only the one study left in the MW report now that says anything about the total contribution of migrants, and it's the one that misleadingly uses total immigration rather than net. The one that Professor David Coleman would sort of agree with if MW had actually used the correct measure for migration. The one that, if you used the correct measure for migration, would say that migrants don't contribute only a small Mars bar per month per head, but three small Mars bars a week per head - or twelve small Mars bars a month (if you take 16p as indicating a small Mars bar, like MW). A total of 720 million small Mars bars a month, rather than 60 million.

I have emailed MW pointing out that now their report differs significantly from Professor Coleman's figure, and asked if this is because MW disagree with him, or if there are several reports that contradict MW's findings that they haven't mentioned in their report. I didn't point out that the two figures would be exactly the same had they used net rather than total migration. They can work that out for themselves.

Whatever their explanation, either David Coleman (Professor of demographics at Oxford University) was wrong when he said there were several studies that showed net migrant contribution to GDP was 0.1% and there's in fact only one, and he was wrong and that study shows a significantly lower figure than 0.1%, or he was right and there are a bunch of other studies MW have conveniently ignored. Whatever, something somewhere is screwy.

At least MW have shown some willingness to change their findings when faced with new data, so something might happen(although I do suspect that this might have something to do with me sending my email to the Independent. I only got a reply from MW to the second email I sent - after I'd spoken to the Independent). There'll probably be some series of contortions the figures get put through, but I'm not sure if that's because MW are deliberately contorting them, or that they're so convinced that immigration is a Bad Bad Thing that the figures look screwy to them unless they support their conclusion.

In any case, it's quite fun to shoot holes in their findings, and not very difficult. The frustrating thing is that nobody else - even the 'left-wing' press seem inclined to bother.

No comments: