As the UK’s referendum on EU membership approaches, the debate over the pros and cons of leaving has intensified, write the Migration Observatory team at the University of Oxford. After more than a decade of elevated migration from EU countries, the free movement of people has been a defining issue in this debate. But while there has been much speculation about how Brexit might affect migration to and from the UK and what the social and economic consequences of such changes might be, the outcomes of Brexit are uncertain.
Uncertainty about the impact a Brexit would have on migration falls into two categories: uncertainty about policy, and uncertainty about the impacts of policy. First, we do not know what migration policies would follow an out vote; this could depend on a complex negotiation with the EU and its Member States about the terms of the future UK-EU relationship.
Second, if we did know what migration policies would be implemented after Brexit, we could not be certain about their effects. Migration levels themselves are difficult to predict regardless of the policy framework in place, and not all of the post-Brexit policy scenarios involve lower EU migration to the UK. While existing evidence suggests that the overall economic impacts of migration tend to be relatively small, the impacts of a potential reduction in EU migration would be unevenly distributed and there is still insufficient data or evidence in some domains to draw strong conclusions.
Uncertainty about policy
EU membership affects migration-related policies in different ways. The most significant is work and residence rights: EU citizens’ ability to live and work legally in the UK without having to meet the kind of criteria that non-EU nationals currently face, such as having a skilled job or qualifying family relationship. Others include visa-free travel, enforcement, cooperation on asylum, or the recognition of EU citizens’ qualifications, among many others.
The degree to which the UK would be expected to change any of these policies after Brexit depends on what kind of relationship would replace EU membership.
If the UK negotiated an association agreement with the EU in order to retain full access to the single market, this could involve an arrangement similar to those of non-EU members Norway and Switzerland, which have both implemented free movement. In this scenario, work and residence rights for EU citizens would remain in place and there would not be new restrictions on EU migration.
If the relationship that replaced EU membership did not involve free movement, the UK would be free to impose selection criteria on EU citizens. These criteria might be the same as current rules for non-EU citizens, which require most applicants to have a skilled job, a spouse or partner in the UK, or a place to study with a registered university or college. Many EU citizens would not meet these requirements, since they are working in jobs that do not meet the skill criteria. However, it is also possible that a broader overhaul of the UK immigration system would take place, leading to different rules.
Similarly, we cannot yet know how and whether other EU countries would change their policies towards British citizens and who would be affected by an end to free movement in the absence of an EU-wide agreement. If British citizens had to meet other EU countries’ normal immigration criteria for work, study, family union or retirement, the policies would vary widely across the continent. Bilateral immigration agreements with specific EU countries that are popular with British citizens, such as Spain or France, might also be possible.
Other migration-related policy decisions that could affect the UK are similarly uncertain. These include whether citizens of some EU countries would be required to apply for visas in order to travel to the UK for short-term business or tourism. While this is primarily a border management question rather than an immigration one, it could have implications for enforcement of new immigration laws.
Uncertainty about the impacts of policy
Focusing on EU migration for work or residence in the UK, the impacts of Brexit on the openness of UK policies towards EU migration range from very small (if free movement remains in place) to quite significant (if selection criteria substantially restrict the circumstances under which EU citizens can move to the UK).
If new policies did significantly reduce EU migration to the UK, what would be the socio-economic impacts? Would this feed through into significant impacts on the UK, its labour market, its public finances or public services?
An exit from the EU would affect the UK economy in many different ways, and these effects interact with each other. For example, if Brexit increased or reduced overall UK economic growth, this could lead to either higher or lower demand for migrant workers and this would shape both the expected levels and impacts of migration. It is therefore difficult to talk about the impacts of lower EU migration as if it would be taking place in isolation.
Putting aside this very important caveat, this commentary gives an overview of what we do and do not know about the potential economic impacts of reducing EU migration. There is now a significant body of research on the impacts of overall migration to the UK, which the Migration Observatory summarises in its regularly updated briefings (see, for example, our briefings on the labour market effects of immigration, its fiscal impacts, and its impacts on local public services).
This evidence base is not perfect and, importantly, does not specifically address the impacts of the kinds of reductions in migration that could potentially be the outcome of Brexit. Similarly, much of the research has been conducted looking at periods when migration was increasing, and it is not guaranteed that the impacts of reduced migration would be a mirror image of the impacts of increased migration, as the Migration Advisory Committee has pointed out. However, some general observations are possible.
If Brexit led to significant restrictions on EU migration but everything else remained the same, on average the economic impacts on the UK would probably be relatively small in the medium to long term…
One consistent finding from past research is that many of the measurable aggregate economic impacts of migration tend to be relatively small. In the UK labour market, for example, the overall effect of migration is thought to be limited. Workers in higher-skilled jobs are more likely to benefit from migration and workers in low-skilled jobs more likely to lose out, but in either case increases in migration have not been a major factor driving wages or employment prospects. One analysis, by the Migration Advisory Committee, also examined regions and periods when the number of migrants in the UK population decreased or remained constant between, and did not find any effect on the employment of people born in the UK.
If Brexit reduced migration to the UK, it follows that the effects on UK workers across the labour market as a whole would also be small.
Similarly, the overall fiscal impacts of migration do not appear to be large. For example, a major 2014 analysis found that EEA migrants who arrived from 2001 to 2011 made net contributions to public finances of approximately £20 billion or £1.8bn per year, after accounting for tax contributions and use of public services and other costs to government. This was equivalent to about 0.25% of total government expenditure in the 2011-12 fiscal year.
…but there could be disruption and the impacts would be concentrated in particular parts of the economy…
Despite the overall smallness of the main economic impacts when considered in the aggregate, the impacts of reducing EU migration would be much more important for some groups than others. Some areas of the economy have come to rely heavily on EU migration and their business models could be profoundly affected by lower net migration over sustained periods of time.
Figure 1 (below) shows the percentage of workers in different industries who were born in EEA countries from 2006 to 2014. The share of the workforce drawn from EEA countries has increased considerably in several industries. For example, in 2014 12% of workers in accommodation and food services were born in EEA countries, up from 7% in 2006. The representation of EEA workers in admin and support services increased from 4% to 9% during this period, and in manufacturing it increased from 3% to 9%.
Many of these industries rely on low-wage workers who would be least likely to qualify to migrate to the UK if free movement were eliminated. In 2015, 13% of people in the lowest skilled jobs, those known as ‘elementary occupations’, were born in EEA countries, up from 6% in 2006 (Figure 2, below). Significant increases also took place among machine operators and skilled tradespeople.
People born in EEA countries still make up a minority of all foreign-born workers. By 2014, 6% of workers were from EEA countries, compared to 9% from Non-EEA countries. In certain occupational groups, however, this was no longer the case. For example, EEA-born workers exceeded those from non-EEA countries in elementary occupations in 2014 (Figure 3, below).
A reduction in the availability of workers in these jobs would require employers to adapt in various ways. How they would respond in practice would depend on their circumstances and the constraints of the market in which they operate. For example, some employers might mechanise processes (such as harvesting or warehouse distribution) to reduce the need for staff, or raise wages to try to make the jobs more attractive. Both of these options are likely to raise costs, which may be passed on to customers through higher prices.
In other cases, such adjustments might not be possible. For example, automation might not be technically feasible or consumers might be unwilling to pay higher prices, instead reducing their consumption or switching to cheaper imports from abroad. In these cases, companies might reduce their output or, potentially, go out of business.
These responses will vary depending on the industry and the circumstances of the business, making blanket predictions about how employers overall will respond unreliable.
…and in some domains, the evidence or data are too limited to draw strong conclusions.
In some cases, the potential impacts of lower EU migration cannot be predicted because current data or methods are insufficient.
For example, lower EU migration into the public sector workforce could, in theory, affect the cost or quality of public services. By the year ending Q3 2015, for example, about 12% of nurses and midwives who had started their job within the past two years were born in EU countries. For public sector employers like the NHS, the impacts would be different than those discussed above for private employers. The NHS, of course, will not go out of business or charge higher prices to ‘consumers’. If the cost of recruiting NHS nurses increased as a result of lower EU migration, this would increase costs to the taxpayer rather than NHS users directly. The quality of services – such as waiting times or access to specialists – could also be affected. But as the Migration Advisory Committee has pointed out, the impacts of foreign-born public sector workers on the cost or quality of public services is too complex to be quantified using currently available data and methodologies.
Evidence on the impacts of migration on the cost of housing at the national level is extremely limited. Migration contributes strongly to population growth and thus to demand for housing in a country where the supply of new housing has lagged behind demand. This is expected to push up prices. However, there are methodological difficulties measuring the impacts of migration on house prices, which also depend on other factors such as the incomes and demographics of the UK-born population. Increases in house prices or rents also have different implications for different groups of UK residents (such as home owners vs. renters).
More generally, the fact that migration affects the UK economy in different ways that all interact with each other makes it hard to model the knock-on impacts of migration throughout the economy. For example, if lower EU migration did increase opportunities for UK-born workers in low-wage occupations, this could lead to savings in the welfare budget; but if lower EU migration reduced the viability of UK businesses or the productivity of high-skilled workers, tax revenues would be expected to fall. The potential for numerous, nuanced impacts of this kind is quite significant, making the overall effects of reducing EU migration difficult to model.
Second, some types of movement between the UK and the EU, such as people who split their time between countries or who commute across borders, are not accurately captured in migration statistics. Statistically, the line between long-term ‘migrants’ and short-term ‘visitors’ can be blurred. This makes it hard to understand their impacts and what the UK and its citizens might lose or gain if new barriers to movement were introduced.
For example, national censuses and labour force surveys provide the most reliable data on the numbers of UK citizens abroad and EU citizens in the UK, but they undercount people who are not full-time, long-term residents. At the time of the 2011 Censuses across the EU, 1.1 million British citizens were living in 25 other EU member states (Belgium and the Netherlands were not included because of a lack of data). This estimate is based on high-quality data but will not count some UK citizens who live abroad for part of the year or who have at some point taken advantage of free movement but have since returned home.
Uncertainty about the policies that would follow Brexit, together with gaps in the data and the fact that many of the effects of migration are inherently difficult to measure, make the migration-related impacts of Brexit very difficult to predict.
Of course, there is also uncertainty associated with migration after a vote to remain in the EU. Even if policies towards EU citizens remain broadly the same, EU migration itself could still change considerably over time. The UK’s economic performance and that of other EU member states, for example, are key factors in shaping migration flows and their trajectories are difficult to predict in the medium and long term.
Confident predictions about large economic effects, whether positive or negative, as a result of Brexit’s impact on migration should therefore be treated with caution.
Thanks to Martin Ruhs for comments on an earlier draft of this commentary. This post first appeared at MigrationObservatory. It represents the views of the authors and not those of BrexitVote, nor the LSE.