Yesterday, we introduced our new “Right to reply” series, where outside writers take on some of the ideas and arguments raised on Coffee House. In that case, it was the IPPR’s Matt Cavanagh replying to Fraser’s recent post on immigration and the labour market. Here’s another reply to the same post, this time by Jonathan Portes of the National Institute of Economic and Social Research:
Myths abound when it comes to the effect of immigration on the labour market — and the most damaging of these is that most or all “new jobs” go to migrants. Although I agree with Fraser Nelson’s general views on immigration, he is misleading on this one point. It is obvious to anyone who considers it for more than a couple of minutes that immigrants don’t fill most of our new jobs. Think about where you work. How many of the last ten people who were hired were immigrants? In most workplaces, probably none, one, or two. Very roughly, about 20,000 people start a new job every working day in the UK — the vast majority were born here.
The chart below (produced by Jonathan Wadsworth of Royal Holloway College, and also a member of the Government’s own independent Migration Advisory Committee) shows, using Labour Force Survey data, that in fact about 15 per cent of new hires in 2010 were immigrants. This is somewhat, but not much, higher than their share in the workforce; not surprising, because immigrants are younger and more likely to be new entrants to the workforce, so switch jobs more. So 85 per cent of “new jobs” went to British-born workers. (“new jobs” is something of an oversimplification; this is “new hires”; there really isn’t a proper definition of “new jobs”).
So the interpretation we see in some parts of the press — that somehow native Brits aren’t getting a look in, because employers prefer to hire migrants — is just wrong. The whole argument goes nowhere.
So what did (and does) Fraser mean? Well, as he has been careful to state in recent columns, his figures refer to net changes; his, arithmetically correct, point is that of the net increase in employment over recent years, much or all can be attributed to net increases in immigrants in the labour force.
But looking at net changes and concluding that this says anything about the dynamics of the labour market is an elementary error. Groups whose share in the labour force is increasing — for whatever reason — will inevitably constitute a disproportionate share of any net change in employment.
This is easy to see, because we can repeat it for lots of groups, other than immigrants, whose overall share of the labour force is increasing. For example, we can repeat Fraser’s calculations for people born between 1980 and 1990. Guess what — they too took all the net “new jobs” created in the last decade. In fact, by my very rough calculations, well over 200 per cent of them! Why? Because hardly any of them were in the labour force before 2001 and now the vast majority are, while older workers have left. Does that mean that employers hire only twenty-somethings, or that they’ve displaced others? Of course not.
Even sillier calculations are possible. Under Labour, all the net new jobs in the UK economy went to people who own iPods and iPhones. And under the Tories, they all went to people who own dishwashers. The informational worth of any of this is, of course, nil.
And there are more problems with Fraser’s recent post on all this. He suggests, for instance, that immigrants have in some sense displaced Britons from the workforce, taking, in good economic times, jobs that could otherwise have gone to unemployed Brits. As he puts it:
“And mass immigration allowed Brown to grow the economy without fixing welfare. It broke the link between more jobs and less dole.”
The problem is that this is just assertion; there’s no evidence. Fraser is just wrong to suggest that the good economic times did not benefit British workers, if what we mean is their chances of getting and keeping a job. What matters is not the absolute numbers in employment, but employment rates — the chances of someone (of working age) having a job. Employment rates rose significantly after the early 1990s recession up until the mid-2000s, and between 1999 and 2008 the employment rate for UK-born people between the ages of 16 and 64 was above 73 per cent in every quarter. As far as I can tell (data for the UK-born don’t go back before 1992), this is the highest sustained rate in Britain’s recorded economic history.
So for your average (British-born) worker, there were indeed more jobs and less dole. And, contrary to Fraser’s assertion, this was at least in part the result of successful labour market reform. I don’t know a serious labour market economist who doesn’t think this good performance was largely the result of successive reforms, by governments of both parties, over the last thirty years.
This doesn’t directly address the question of whether, despite the overall good performance of the labour market, immigration has harmed the prospects of British workers. For this, because we can’t observe the counterfactual — what would have happened if immigrants hadn’t come — proper economic research is required. There is now quite a bit of such research. Generally, what it does is to look at areas where there are lots of immigrants, and areas where there are few, and see whether (taking account of other factors as well) British workers have done better or worse. And generally what it finds is some version of this:
This chart (taken from my own research, with Sara Lemos at the University of Leicester) shows that there is absolutely no discernible correlation between the areas where new migrants from Eastern Europe settled and changes in the claimant count. More sophisticated analysis shows pretty much the same thing; and using other data sources, and other definitions of migrant, likewise. A reasonable summary is provided by Jonathan Wadsworth:
“It is hard to find evidence of much displacement of UK workers or lower wages, on average … The less skilled may have experienced greater downward pressure on wages and greater competition for jobs than others, but these effects still appear to have been relatively modest.”
I hope the above discussion has shown just how divorced from the facts and evidence the debate on the impact of immigration on the UK labour market has become. Most economists, wherever we are on the political spectrum, think that well-functioning markets usually do a pretty good job of allocating resources. That goes for the labour market too, so it is no surprise that liberal (in the true sense of the word) immigration policies are good for the economy and the labour market.
Jonathan Portes is Director of the National Institute of Economic and Social Research and former chief economist at the Cabinet Office.