Newsnight: “Immigration report held back by Downing Street”: some background and analysis

Two years ago, the Government’s independent Migration Advisory Committee published an econometric analysis of the impact of migration on the employment of native British workers. It concluded – contrary to a number of other analyses, by me and other labour economists – that there was, at least for some immigrants and in some circumstances, a negative impact:

an extra 100 non-EEA migrants could lead to 23 fewer British residents being employed.

Not surprisingly, the government jumped on this new analysis. It seemed both to contradict the arguments of those, like me, who had argued that immigration had benign impacts on the UK labour market; and to support government policy to reduce non-EEA migration by making it more difficult for skilled workers, students and spouses to move here or stay here. Theresa May claimed

And we must be honest about the fact that, in some cases, uncontrolled mass immigration can displace local workers and undercut wages…every 100 non-European working age immigrants were associated with 23 fewer British-born people in work.

The Daily Mail described this as

hard evidence that cannot be ignored..The MAC has provided the evidence he needs to get a grip on immigration.

However, some of us were more sceptical. Using different data, I and others at NIESR had published a report at almost exactly the same time which appeared to show the opposite – no apparent negative impact on British workers.   How to reconcile these conflicting findings? I wrote (apologies for the lengthy self-quotation):

What should we make of this?  For me – and I think for most professional economists – this is a suggestive result, but even if you take it at face value, given that it is only one of many, and applies only to some immigrants for some of the time, not more than that.  As they [the MAC] say, with commendable honesty

 “the results are statistically insignificant when outliers are removed from the data”

“our results are not robust to alterations to the model specification”

“our results are not robust to substituting ..the same ratios lagged..this suggests our results may not be robust to endogeneity bias.”

In addition, it also appears that when the period 1995-2009 is used (ie dropping just 2010) again the results become insignificant.  In other words, tweak the data just a a little, and the result is no longer statistically significant.  Most economists would place rather little weight on estimates which seemed to fail these fairly standard tests.

Matt Cavanagh, then at IPPR, was equally sceptical, and wrote an article on similar lines.

So what does Christopher Cook’s Newsnight report today tell us?  Briefly, the documents he has obtained suggest that further analysis, done by civil servants and reviewed by outside academics, confirms our scepticism. The MAC results were not robust, and the 23 for every 100 number is, if not necessarily wrong, definitely no longer valid, and should not be used. However, according to Newsnight, despite the fact that the further analysis was completed some months ago, the government – and Number 10 in particular – don’t want to publish it, because it would force them to admit that the evidence on which they claimed to be basing policy can no longer be relied on – and indeed that the evidence suggests pretty much the opposite.

Obviously I can’t critique the research, and work out why the bottom line was so different from the original MAC analysis, until it is actually published – although given what I said at the time about the methodological issues above, it’s hardly surprising.  However, there are two fairly clear conclusions we can draw from this.  

The first is on substance. It confirms that the strong consensus among labour economists that that there is, as Jonathan Wadsworth (himself a member of the MAC)  puts it “little or no evidence of an adverse impact on jobs and wages” is just that – a consensus, and that it extends to economists and analysts within government as well as outside. 

The second is on process. There is simply no excuse for the government not to publish a report which was completed some months ago, is purely analytical, and is clearly of wide public interest.  There is no shame in relying on independent expert evidence which at the time appeared credible, even if (not unusually for any government) Ministers grossly overstated the strength of that evidence.  What is shameful is suppressing subsequent evidence, just because Ministers don’t want to have to eat their words. .