Skip to main content

 

Immigration

People move to a new country for many reasons. How do these individual stories of seeking fresh opportunities play out over time both for those arriving and for the communities they move into? Economists are particularly interested in how immigration affects wages, employment, housing and public services.

Work, and increasingly study, are the main reasons that people from other countries give for their move to the UK. How far would you travel to find a job? How much better would your prospects need to be to leave friends, family and familiarity to work in a new country? How do you think locals would react to your arrival?

For labour economists trying to understand the implications of these moves, immigration is an example of how simple economic theories can break down when faced with the complex world. For example, while a greater supply of workers "could" lead to a fall in wages, this is often not the case. Researchers at the Centre for Economic Performance (CEP) have been among those investigating why.

With one or two notable exceptions, when CEP was founded 30 years ago not many UK economists were writing about immigration. At the time, unemployment was raging for the second time in 10 years. While some academics were interested in whether the relocation of workers (or firms) within a country could help alleviate unemployment, not many were concerned about immigration and its effects on jobs and wages of UK-born citizens.

This made sense because immigration in the early 1990s was relatively low and stable. Since then, UK immigration has grown a lot - or at least it had until Brexit and the Covid-19 pandemic - and over the last 10 years the issue has become increasingly contentious.

It is estimated the number of immigrants in the UK has fallen by over 1 million since 2020 as the Covid-19 pandemic took hold.

The graph below shows that when CEP was founded in 1990, the number of immigrants (conventionally measured as the number of individuals born outside the UK) was around 3.5 million, about six per cent of the population (right hand panel). From the mid-1990s, immigration grew strongly, pausing in the wake of the 2008 financial crash, but reaching a peak of around 9 million (14 per cent of the population) just prior to Brexit. The onset of the pandemic has had a seemingly remarkable effect on the estimated number of immigrants, which fell by more than 1 million, a two percentage point share of the population - at least according to the Labour Force Survey, the principal source of labour market data used to generate the official unemployment count. While some of this may reflect problems with sampling households in lockdown, if it is true and, importantly, if it is sustained, then this is a significant development.

Immigration 1990-2021

Graph about immigration between 1990 and 2021 by Labour Force Survey
Chart: Immigration 1990-2021. Source: Labour Force Survey.

The effect of immigration has long been the subject of debate in academia and among the wider public. On one hand, immigration is seen as good for the economy - since it raises the population and hence the size of the economy (GDP), which means more tax receipts (if most immigrants are in work) that governments can use.

Modelling hopes and fears

If most immigrants are more skilled or better at doing a job than the native workforce or fill skill gaps, then national productivity will rise, creating more wealth overall. Immigrants will usually also benefit since, by and large, they are better off in the UK than in their country of origin.

On the downside, these immigrants may also be among the more skilled workers in their home countries, so those countries will lose out. And people in the host country may worry that the increase in population brought about by immigration will increase competition for jobs, put downward pressure on wages and raise competition for scarce resources like health, housing or education. Rising immigration is also sometimes linked to fears that crime will rise.

Sometimes economics can be invoked to justify these fears. In the simplest of all economic models - one in which all individuals are identical, and the economy does not trade - immigration raises labour supply and, with no room to expand output through trade, this puts downward pressure on wages. Jobs and wages are usually the first thing that people look to when thinking about immigration's effects and the academic research followed this pattern before branching out into other areas.

Alas, simplicity in a model often omits important features that can reverse these predictions. For example, other economic models suggest that the existence of a large enough traded sector means the economy can adapt the mix of goods produced to accommodate immigrants without the need to lay off its existing workforce or cut wages (in the traded sector prices and wages are effectively set on world markets).

In reality, there are many different types of migrants with different skill sets just as there are many different types of UK-born workers. It could be that wages of workers who are the closest "substitutes" for immigrants (and so most easily replaced by them) could fall, while there are other workers who could benefit from a rise in demand for their skills from immigrants - meaning their wages and employment could rise.

This suggests the effect of immigration on UK-born individuals depends on just how close a substitute immigrant labour is for existing non-immigrant labour and this, in turn, may depend on the skill involved in a job. So, immigration could rise without putting downward pressure on wages or jobs, if those workers are helping meet a demand for skills. And if trade expands as a result of changes in immigration affording new opportunities this will also mitigate against downward pressures on wages and jobs.

All this is important because it means that the effect of immigration in the labour market is ambiguous theoretically and needs empirical verification to try to establish evidence. CEP, along with the Centre for Research and Analysis of Migration (CReAM) at University College London has been prominent in gathering this evidence.

Using data to find out what is happening here

In 1997, CEP's Brian Bell was the first to resurrect the measurement and study of the relative performance of immigrants in the UK, an earlier study having been published in 1980. Bell pointed out that (the then) recent cohorts of immigrant in the 1980s were more educated than their UK-born counterparts, but that this did not prevent them experiencing a wage penalty on arrival - immigrants with the same set of characteristics as UK-born individuals, received lower pay. He also found this wage penalty gradually dissipated over time (economists call this an "assimilation" effect).

One of the first of the modern studies to look at the impact of immigration on the labour market prospects of UK-born individuals was a Home Office report The local labour market effects of immigration in the UK co-authored by CEP's Jonathan Wadsworth and colleagues at UCL. The idea here - and this is the basis for much empirical work on immigration - was to compare changes in the labour market prospects of areas or groups with large immigrant inflows with changes among a "counterfactual" group that is similar but has experienced lower immigrant inflows.

Choosing the appropriate counterfactual group is something that has bedevilled the economics profession for a long time, since unlike the medical profession, there are rarely randomised trials surrounding national changes to an economy.

This report, as happens often in economics research, used skill groups and regions with differing immigrant shares as the basis for comparison. In a sign of what was to come from later research, the report found little evidence of large wage or employment effects on UK-born workers using data from the 1990s.

In 2012, CEP's Marco Manacorda, Alan Manning and Jonathan Wadsworth compared immigrant and UK-born workers in the UK to see how similar, or substitutable, they were. They found evidence of "imperfect substitution", more so among higher skilled groups, meaning that there was limited downward pressure on wages and jobs (and indeed more downward pressure on the wages of existing immigrants who are typically closer substitutes for new migrant labour).

The effects of immigration are hard to find

Since then, CEP researchers have been involved in the search for evidence of immigration's effects across the economy. And many studies reach largely the same conclusion: in the main, any effects of immigration are small.

These include:

In short, it is hard to find substantial economic effects of immigration - either positive or negative - in many sectors of the economy.

This is not to say there are no effects. For example, the study which found small positive effects of immigration on productivity also found large positive effects of immigration on the trade balance, as a firm offering services to a country which employs more immigrants from that country is likely to both increase its sales to that country and reduce the amount spent buying intermediate services from that country, as they can now be done in house. One implication of this is that the UK's trade balance will suffer due to curbs on immigration from the EU.

The explanation of many of the findings in these studies (and the methodology employed) can be summarised in the graph below. The graph uses unemployment, but any of the outcomes listed above would give a similar pattern. Each dot maps the change in the unemployment rates of the UK-born in a local area (county) with the change in the immigration share in that county over the same period between 2004 and 2015 (one dot for each of 67 areas). The solid red line summarises the relationship between immigration and UK-born unemployment rates.

If immigration increased unemployment, we would expect a strong upward sloping line: more immigrants would mean more unemployment for local workers. In fact, the line is, if anything, negative, suggesting that unemployment of UK-born workers went down (or rose less) in areas that had more immigration. It is just as easy to find areas where immigration went up (a lot) and unemployment of UK-born workers fell as it is to find areas where immigration and unemployment rose (compare areas 50 and 60 on the graph, for example). This pattern could be explained if immigrants were more likely to go to better performing local areas - where jobs are plentiful for both immigrants and UK-born workers, or if some UK-born workers move away from areas where immigration is rising. But even when economists use econometric techniques to address these and other problems, the effects of immigration on unemployment are generally small. Something else must explain the poor performance of areas with rising unemployment.

Change in Unemployment v Change in Immigration

Graph about change in unemployment v change in immigration by CEP Election Analysis: Immigration (2017)
Chart: Change in Unemployment v Change in Immigration. Source: CEP Election Analysis - Immigration (2017).

What questions are economists asking now?

On the academic front, it is a lot easier to explain the small immigration effects found by researchers if we think about supply and demand in the following way: while the supply of workers has increased, meaning there is more competition for jobs, the increase in workers also means an increase in the demand for housing, education, food and other things - which in turn creates jobs. Even if that demand was met entirely by immigrants, no-one would be worse off, and it is likely that the increased demand would be met by both immigrants and non-immigrant workers. This is a very fruitful area of research going forward.

It is also too simplistic to assume that immigration has the same effect in all sectors of the economy. Economists tend to focus on average effects and there is much interesting work to be done looking at immigration's effects in a disaggregated (sector by sector) way.

It is conceivable that the downward trends in immigration that first began after Brexit in 2017, and were apparently accelerated by the Covid-19 downturn, will continue - though we need more hard evidence to back up this latter finding. If 1 million people (two percentage points of the population) have left the UK and this is not just survey sampling problems during the pandemic but a real emigration response to the combination of Covid and Brexit, then firms affected will need to address the new reality, as will schools, hospitals, the housing market and the wider economy.

But if immigration had little effect on wages, employment and the wider economy when it was rising rapidly then should we expect little effect when immigration falls? Not necessarily. Here we need to think about the economist's mantra about "holding other things equal". It is reasonable to think that the effect of dramatic outmigration is not necessarily the same as in the effect of more gradual immigration in the past.

The restrictions on trade with the EU that accompany Brexit, the change in UK growth path that will emerge after the pandemic, and the relative performance of other economies are all known unknowns that mean the impact of immigration going forward may not be the same as going back.

Public debate about immigration's effects is unlikely to disappear despite the academic evidence that it doesn't seem to matter much, on average. So, there will still be a need for continued observation of the economy and its interaction with immigration.

By Jonathan Wadsworth