Immigration: how much do we really understand about how it affects job prospects?

Gwilym Price - Professor of Urban Economics and Social Statistics, and Co-Director of the ESRC CDT Data Analytics and Society, University of Sheffield

The irony is that California has benefited enormously from the inflow of migrants over the years. If it were a country, its economy would be the fifth largest in the world. This contrast between the fear of migrants, and the reality of the economic benefits they bring, is a familiar story the world over and one that strongly resonates with much of the contemporary Brexit debate in modern Britain. Those who voted in favour of leaving the EU did so primarily because of concerns over immigration. On the other hand, those who want to remain in the EU have emphasised the economic growth and new employment that migrants can generate.

A crucial feature of these political divides is that they vary hugely across the country. There is startling regional variation, for example, both in attitudes to immigration and in the Brexit vote. It was voters in the former industrial areas of the UK, with low income, high unemployment and low levels of educational attainment, who were particularly likely to vote for Brexit. And areas with low immigration levels largely voted for Brexit – per the graphic below.

This raises a controversial but profoundly important question. Did voters in these areas make rational, informed decisions? Or were they deluded: swayed by tabloid headlines claiming migrants were taking their jobs? Such claims contrast starkly with the scientific evidence, which consistently finds little if any impact of immigration on the employment prospects of UK-born workers.

Accounting for local variation

But there is an important caveat to this interpretation. The scientific evidence to date is based on nationwide estimates. The impact of immigration at the local level and whether its effect on employment varies across the country is not actually known. Our usual data source for estimating the impact of immigration is the Labour Force Survey, which has a sample size equal to around 0.5% of the UK population. While this is fine for producing national estimates, it does not provide large enough samples at the local level to gauge regional or local variations in the impact of immigration on employment.

If immigration really does reduce the employment prospects of UK-born workers in some areas while boosting them in others, this would raise considerable social justice issues. It would, for example, provide a strong moral obligation for regions that have gained from immigration to compensate areas that have lost out. It also profoundly affects how we interpret the Brexit vote and how we make future immigration policy. If we cannot estimate the regional employment effects of immigration, we cannot know whether opposition to immigration is due to local economic ignorance, or the result of a profound injustice.

Plugging the gap

Our goal over the past three years has been to find a way to plug this gap in the evidence base. To achieve this we needed a dataset with enough observations to allow us to tease out the effects of immigration at the local level. The UK census is the obvious candidate as it includes pretty much the entire adult population. But the census only runs once every ten years and between these snapshots, local boundaries change quite significantly. So, to produce enough time points in the data, we combed through every census going back to 1971 to construct a unique dataset.

There were two further challenges. First, the fact that employment rates in one area are often affected by those of neighbouring areas. We call this the “spatial spillovers” effect and it must be controlled for. The second problem was that correlation did not necessarily mean causation. For example, if we observed that more migrants increased employment levels in an area, it may have been because migrants generated higher employment. But it might equally have been because migrants were drawn most strongly to areas that already have high employment.

We developed a model to take into account these issues and derived a set of geographical areas that are consistent across five censuses. While we do not yet have estimates for every region, we have successfully run the analysis for London. So we think we have found a way to plug the evidence gap.

In London we found that no migrant group had a statistically significant long-term negative effect on employment. EU migrants – the main source of controversy underlying Brexit – significantly boosted employment rates in London.

What this means for UK immigration policy

Our findings have important implications for post-Brexit immigration policy. If Boris Johnson is to pursue his proposed points-based immigration system, it must take into account the actual impact of EU migrants on employment. A second, more wide-reaching implication of our research is that it opens up the possibility of identifying regional variations in how immigration affects jobs. We may finally be able to confirm whether or not the concerns about immigration in some of the poorest areas of the UK have any basis in fact.

If some areas are shown to have lost out due to migration while other areas have made huge gains, it would strengthen the case for redistributing the spoils of globalisation between regions. And this would, of course, fit rather well with Johnson’s “mission to work night and day” to earn the support of voters in traditional Labour heartlands.

This article was originally posted in The Conversation.