Location, Location, Location: Should Different Parts of the UK Have Different Immigration Policies?

Immigration policy—that is, the rules that determine who is able to come to the UK to live, work and potentially settle long term—is decided at the national level. But over the past few years, various proposals have emerged for devolving power over parts of immigration policy to the constituent nations and regions of the UK.

These discussions have intensified since the EU referendum. While proposals for a subnational visa system post-Brexit do not only concern EU citizens, the prospect of a large-scale overhaul of UK immigration policy has prompted proposals on broader changes to the whole immigration system for both EU and non-EU citizens.

The All Party Parliamentary Group (APPG) on Social Cohesion, for example, recently recommended a regionalised immigration system (APPG, 2017); the London Chamber of Commerce and Industry has argued for London-specific work permits post-Brexit (LCCI, 2016); the Scottish Government has proposed devolved powers over immigration policy in Scotland (The Scottish Government, 2016) (see also Boswell et al, 2017); and similar schemes have been put forward by the City of London Corporation (City of London Corporation and PwC) and by the Institute for Public Policy Research (IPPR) (see: Griffith and Murray, 2017 forthcoming).

This analysis outlines the main arguments in favour and against regionalising the immigration system. It compares a possible subnational work-permit system to a UK-wide one such as the ‘Tier 2’ system that currently applies to non-EU citizens but that may, in the future, also cover EU citizens.

Proposals for regionalised immigration policies generally focus on work-based migration rather than migration of family members or refugees. (Although the Scottish government has made the much broader proposal of retaining free movement—including for family members—in Scotland after Brexit.) This paper focuses on work-related migration.

A regional work visa system could be designed in many different ways. Two important questions are how large the regional variations would be, and who would decide them. A regionalised system could involve:

• Modest variation: for example, a work permit system with mostly uniform eligibility criteria but some regional variation—such as lower salary thresholds in regions with lower average earnings, or region-specific ‘shortage occupation lists’ which exempt employers from having to advertise positions locally before hiring a migrant worker.
• Large variation: for example, whole visa types existing in one region but not in another. This could involve post-study work visas in some areas but not others; or region-specific quotas for low-skilled workers coming from the EU after Brexit

A separate but related question is how regional variations would be determined. This could involve:

• ‘Top down’ variation. Under this design, criteria might vary by region but the variations would be decided in Westminster – for example, based on Home Office or Migration Advisory Committee analysis of labour market statistics and/or stakeholder views.
• ‘Bottom-up’ variation. Under this model, regions would play a greater role in deciding policy, either by receiving devolved powers to determine criteria or through a process of negotiation with the Home Office.

Assuming that UK-wide visas (e.g. Tier 2 work permits) continued to exist under a regionalised system, region-specific visas would be mainly relevant for admitting low or middle-skilled workers, rather than people with the highest levels of skills. This is because the most skilled migrants would continue to be eligible for UK-wide visas. As a result, regional visas are not necessarily a tool for regions to ‘compete for the best and brightest’; they are more suitable for bringing in people who would not otherwise qualify to come to the UK.

Precedents in other countries
Most other high-income countries operate relatively uniform work permit systems at the national level, although some have modest top-down variation in criteria depending on local circumstances. For example, the United States has place-specific salary thresholds and Spain has regional shortage lists.

There are two prominent examples of countries where significant immigration policy decision-making powers have been devolved: Canada and Australia. These systems have been introduced in addition to the national-level work-permit systems, with the explicit purpose of helping areas of the country with lower levels of immigration to increase population growth by relaxing eligibility criteria for new migrants (Sumption, 2014). In both cases, subnational governments nominate candidates to be given work visas using their own criteria. Subnational discretion is greater in Canada than in Australia, where all candidates must pass a national-level points test.

A commonly cited concern about regional variation in policy is that it would create enforcement problems, because a region with more liberal policies could create a ‘back door’ into the rest of the country. In other words, people might apply to migrate to one region but then live or work in another.

As far as work is concerned, this argument is relatively weak. The visa rules could make it illegal for migrants to work outside their designated region for a number of years, and for employers in other regions to hire them. Assuming that employers understood the system and did not hire visa holders from a different region by mistake, the question in the short term is therefore whether people would violate the rules and work illegally.

Migrants would not necessarily have more of an incentive to violate immigration rules under a regional system compared to a UK-wide one. Under the current Tier 2 system for non-EU citizens, visas are valid only for a specific job and employers must notify the Home Office if their employee stops working there. If regional visas were also tied to specific jobs in a given region, the situation would in principle be no different. Because of the limited information available on illegal working it is not possible to know whether a meaningful number of Tier 2 work permit holders currently violate the conditions of their stay, although it is generally assumed that compliance is high among this group of skilled professionals and that migrants who are working without authorisation are more likely to have come on visas that are easier to obtain, such as tourist visas or visa-free travel. (Looking specifically at numbers of visa overstayers, data from the United States support this hypothesis (DHS, 2017)).

If regional visas were not tied to specific jobs, the incentives to move elsewhere might be greater. For example, a place-specific post-study work visa or the continuation of free movement in only one part of the country (e.g. as Scotland has proposed) could mean that regional visa holders did not have work already lined up in the right region. If unable to find work locally, they would have a greater incentive to move elsewhere. However, they would still need to find an employer willing to hire them without authorisation. As noted above, migrants willing to work illegally in the UK do not necessarily need a ‘back door’ in the form of a regional visa, since it would generally be easier to enter via the ‘front door’ as a short-term visitor.

Beyond the question of where regional visa holders would be allowed to work, there is also the question whether they would be allowed to reside in a different region from the one on their visa. Unlike Canada and Australia, the UK is relatively small and commuting across regional boundaries is not difficult. Preventing people from living in a region other than the one on their visa would be more difficult. In theory, private landlords could be required to check tenants’ authorised region of residence in addition to UK immigration status, although doing so would add complexity to a task that some stakeholders have argued is already difficult for them to carry out (JCWI, 2015).

How long would regional visa holders be tied to a particular region?
Assuming that regional visa holders had a pathway to indefinite leave to remain and citizenship, they could not be tied to a particular place indefinitely. Skilled workers on Tier 2 visas are currently eligible for indefinite leave to remain after 5 years of work if they meet the criteria (including a £35,000 income threshold).

How many regional visa holders might move region if they gained the right to do so after some time is an empirical question, although the longer workers are required to remain in one place, the more likely it is that they will develop social and work networks that make them less likely to move away once they are free to do so.

The main empirical evidence on whether migrants do, in fact, move away from the regions that sponsor them is from Canada. The Canadian context is different from the UK one due to its different geography and institutions. In Canada, provincially selected migrants are not legally obliged to stay in the province at all. As a result, some provinces have found it difficult to retain the people they have sponsored. A Canadian government evaluation found that in 2008, four of the eight provinces with regional visa programmes had retained more than 80% of the sponsored migrants who arrived since 2000, but two provinces had retention rates of less than 40% (CIC, 2011). A UK system that did tie workers to specific regions would not necessarily face the same problems, although it is relevant to note that in Canada:

• Retention rates were on average lower in provinces that did not require migrants to have a job offer before they arrived (Pandey and Townsend, 2010), as well as for people who were former international students (CIC, 2011).
• Most of those who moved to another province did so within the first three years, with relatively few departing after 5 years (CIC, 2011).

This suggests that higher retention rates could be expected in a programme that required participants to have a job offer and to stay in the region for several years before being able to move elsewhere.

Arguments made in favour of and against regionalised visa systems come in various forms, both political and economic. The main arguments are summarized in Table 1 above.

Economic arguments

The economics and demographics of regions and localities across the UK vary. As a result, the impacts of migration are also likely to vary. This means that there is a strong theoretical argument that subnational variation could increase the benefits or reduce the costs of migration policy by creating a better match with the needs of particular places. In practice, however, things are more complicated. This section lays out the reasons.

It is clear that economic circumstances do vary by region. First, different regions have a different balance of industries and occupations. For example, there are relative concentrations of financial and legal services jobs in London, scientific research and development jobs in the East of England, South East and Scotland, and of manufacturing jobs in the East Midlands, West Midlands and Wales (ONS, 2017). In part as a result, earnings also vary regionally and are significantly higher in London than in the rest of the UK (Figure 1).

Figure 1

Second, some areas have stronger labour demand than others. For example, unemployment in the North East was roughly double the level in the South East in 2016 (Figure 2).

Figure 2

There are also significant regional variations in population growth. While overall UK population growth is relatively high, it varies enormously around the country. In recent years the working-age population has grown fastest in London, the South East, East and the Midlands, while it has decreased in the North East, Wales and North West (Figure 3). Population decline may have negative economic impacts, such as making it more expensive to sustain local amenities such as schools or shops for a smaller number of people. By contrast, rapid population growth may have negative impacts if public services are not able to adjust fast enough.

Figure 3

In other words, the data show that there are clearly economic and demographic differences among UK regions. Some of the variations are relatively small (for example, earnings levels across UK regions other than London and the South East). Others are relatively large (such as variation in unemployment rates and population growth).

The question for policy is whether there are potential economic gains from varying policy to take account of these differences. In practice, there are three main obstacles to doing this:

• In practice it would be exceptionally difficult to determine what actual needs are at the regional level.
• Regions are not homogeneous and indicators such as population growth can vary enormously by local authority within a given region.
• Regional variation makes the immigration system more complex, increasing costs for those who use it.

First, there is no optimal amount or type of migration and deciding what is ‘needed’ is a challenge. Some modest, top-down variation in policy by region would be relatively straightforward to design, such as a lower salary threshold for skilled work visas outside of London and the South East. For more significant variations in policy, however, statistical indicators do not automatically produce obvious policy recommendations. As the figures above show, for example, the North East has experienced low or negative population growth (which could potentially be addressed with higher migration) but also has high unemployment (suggesting that local jobseekers should be available to fill available jobs).

The Migration Advisory Committee (MAC), which is tasked with identifying ‘shortage occupations’ for the UK and Scotland, emphasises that there is no single way to identify economic needs and that the results will depend on subjective judgments and methodological choices (MAC, 2017). This subjectivity is particularly important when it comes to deciding not just whether a shortage appears to exist, but whether it is sensible to address it using migration, rather than rely on other solutions such as automation and increased training or incentives for UK job candidates. If similar analysis were conducted at the regional level, these challenges would not disappear; in fact, they would become even greater because smaller statistical samples mean less reliable results.

As a result, subnational policy decisions on migration would in practice have to be quite subjective. This could increase the influence of interest groups in shaping policy outcomes—something that could variously be described, depending on one’s perspective, as either ‘driven by special interests’ or ‘responsive to local stakeholders.’

Second, regions are not homogenous. Population growth, for example, varies substantially within regions. In Scotland, major cities like Glasgow, Edinburgh and Aberdeen have growing populations while Inverclyde, Dumfries and Galloway, and North Ayrshire have declining populations (ONS, 2015). Areas of high population growth are sometimes located very close to areas of much lower growth. For example, projected population growth over the next 20 years is 14.9% in Coventry but only 4.8% in nearby Warwick; in the East of England, South Norfolk has projected growth of 12.5% but nearby Great Yarmouth only 4% (ONS, 2016). If the goal is to match policy to the demographic context, therefore, regional variation may not be ‘enough’.

Finally, regionalising the visa system would increase its complexity, increasing the administrative burden on users. For example, employers with offices in more than one region may find that employees cannot easily move between offices, and would have to become familiar with multiple sets of immigration rules. In areas near the border between two regions, nearby employers would face different requirements – potentially creating winners and losers.

The MAC has periodically considered the question whether it would make economic sense to introduce a modest variation in the work-permit system, introducing regional salary thresholds to take account of the fact that both salaries and the cost of living are lower outside of London and the South East. In consultations on this topic, the MAC has received conflicting responses from employers and other stakeholders such as unions (MAC, 2012). Some welcomed the possibility to ‘level the playing field’ for employers outside of London, while others were concerned that doing so would make it ‘complex, bureaucratic, impracticable or unworkable.’ The scale of the additional administrative costs that would arise, however, is rather difficult to quantify.

Political arguments

Political arguments in favour of a subnational work-permit system come in three main forms:

• That it would allow migration to better reflect differences in public attitudes within the UK;
• That it would enable local/regional policymakers to make migration policy decisions consistent with decisions in related policy areas such as migrant integration and service provision;
• That it would give people a feeling of greater control over migration.

Addressing the first argument, the recent report from the APPG on Social Integration quotes the Australian High Commissioner arguing that a regional visa system could be ‘responsive to the threshold of community comfort.’ Attitudes to migration do differ across the UK with the public in London and Scotland generally expressing slightly more favourable/less unfavourable attitudes to migration than those in the rest of the country (Migration Observatory, 2011).

The variations are relatively modest, however. The 2011 and 2013 British Social Attitudes survey found that majorities in all UK nations and regions except Inner London supported reducing migration (Figure 4). A more recent poll for British Future found only slightly more liberal attitudes towards migration in London and Scotland compared to the UK as a whole (Katwala et al, 2016).

Public opinion is also not homogeneous within regions or local authorities. The June 2016 share of people voting to leave the EU, which is associated with negative attitudes towards migration, tended to be lower in big cities than in rural areas or smaller urban areas, and larger regional areas will contain both (House of Commons, 2016). Regardless of the level of geography chosen, there will still be significant opposition to the majority view.

Figure 4

The second political argument is that after migrants arrive, policies to facilitate their integration are largely the responsibility of local government—and thus that it makes sense for these communities to have more control over the levels and types of migration they receive. The idea is that they would be able to take coherent decisions about immigration and integration at the same time, and be accountable for them. The APPG on Social Integration argued that devolution of work-permit policy should be accompanied by devolution of skills policy too, so as to allow coherent regional skills plans.

The APPG argues that for these two reasons, more regional control over immigration would improve public confidence in migration policymaking, since policy would no longer be imposed by a distant elite in Westminster and subnational policymakers would be more accountable for policy decisions. IPPR, similarly, argues that local debates on migration could be more meaningful and constructive if matched with local powers to shape it (Griffith and Morris, 2017 forthcoming).

There are two main arguments against regional visas from a political perspective. The first is that regional political institutions do not exist everywhere in the UK to manage devolved migration policies. While Scotland, Wales and Northern Ireland have devolved administrations that could provide democratic input into migration policy, the English regions do not. Some city regions such as London and Manchester now have devolved powers and elected majors, but these only cover some parts of the country. English regional political institutions could in theory be created to deal with migration and other policy areas, but this would be a longer-term and significant project that could take some time to implement and that would represent a significant shift in the way the UK is governed.

There are some non-political institutions with more comprehensive geographical coverage. For example, Strategic Migration Partnerships are networks of local authorities that share information on migration issues; they exist in all areas of the UK although vary in the level of resourcing, scale and ambition. Local Enterprise Partnerships (LEPs) are business-led partnerships with local authorities that help develop local economic development plans and thus may have views on the relationship between migration policy and regional industrial strategy; they also cover all areas of England but are numerous (38 in total) and not mutually exclusive as some local authorities participate in more than one LEP. However, these institutions are not democratically accountable and thus may not suitable vehicles for realising the political goals described above.

Notably, the two international examples of significant devolution of migration policy—Australia and Canada—are both federal countries where states and provinces have well-developed political institutions with their own elected legislatures, executives and judiciaries.

Finally, regional visa proposals generally focus on work and do not cover other important reasons for migration such as family unification. From 2010-2016, 31% of the foreign-born population interviewed in the Labour Force Survey reported that they came for work-related reasons, with the rest reporting family, study, asylum and ‘other’ reasons (although many long-term residents who initially came to study will have held a work visa at some point) (Giuntella et al, 2017). Assuming that policy towards non-work migrants and many work migrants continued to be decided at the national level, there is a limit to the extent to which regions would genuinely be in control of migration policy and thus how much regionalisation can be expected to shift the politics of migration overall.

Subnational visas are feasible from an operational perspective. Especially if designed to require a job offer and prevent visa holders from working legally in the ‘wrong’ part of the country, the task of enforcing the rules should in principle be similar to a nationally uniform work-permit system.

The question whether they should be introduced is considerably more complicated. The main economic argument in favour of regional variation in the immigration system—that different policies should be responsive to different regions’ true economic needs—sounds intuitive but encounters the strong counterargument that it is simply too difficult to define what migration is ‘needed’ in an objective way at the subnational level. From an economic perspective, it is therefore not clear that significant regional variation would lead to a better match between economic needs and policy design. At the same time, regionalisation has a potential drawback in the form of higher costs of a more complex immigration system, although these costs are hard to quantify.

Many of the arguments for subnational visas are political rather than economic. Here, the key arguments are that devolving at least some of the responsibility for migration policymaking would enable ‘joined-up’ decision-making (e.g. for migration, integration and service provision) and foster a sense of control and accountability closer to home. Whether a subnational visa system would improve public confidence in policy is a complex empirical question that would have to be tested in practice. There would also be challenges, such as the lack of existing political institutions in many parts of England and the fact that under most proposals only a small share of migration would be affected by regional migration policies.

But perhaps analysing the pros and cons of subnational visa systems is missing the point. A poll of Scottish public opinion conducted for the Migration Observatory before the 2014 independence referendum hints at this possibility (Migration Observatory, 2014). A majority of Scottish respondents (60%) said that they would prefer to see the most important decisions about immigration made by the Scottish government, not the UK government—even though a majority (58%) also wanted to see reduced immigration in Scotland and relatively few (22%) believed that an independent Scotland would actually deliver less open migration policies. This implies that respondents did not necessarily favour devolved control over migration purely because they thought it would lead to policy that better matched their preferences.

The level of governance at which policy decisions should be taken is not just a question of assessing the social or economic effects, but depends on broader question such as how much people identify with a particular level of government and where they intuitively believe that power should lie. In other words, whether different parts of the UK should have different immigration policies may well be a matter of principle rather than of costs and benefits.

Funding for this report comes from the UK in a Changing Europe’s Brexit research project which is funded by the Economic and Social Research Council. The Migration Observatory’s work on post-Brexit immigration policy is also made possible with the support of Trust for London and the Paul Hamlyn Foundation.

Thanks to Rachel Marangozov, Jonathan Portes and Jacqueline Broadhead for comments on a previous draft of this report, and to Yvonni Markaki for sharing her analysis of the British Social Attitudes survey.

• APPG. 2017. Integration not Demonisation. All Party Parliamentary Group, London.Available Online
• Boswell, C., Kyami, S., and Smellie, S. 2017. Scottish Immigration Policy After Brexit: Evaluating Options for a Differentiated Approach. University of Edinburgh, Edinburgh. Available Online
• City of London Corporation and PwC. 2016. Regional Visas A unique immigration solution?. City of London Corporation and Price Waterhouse Cooper, London. Available Online
• CIC. 2011. Evaluation of the Provincial Nominee Program. Citizenship and Immigration Canada, Ottawa. Available Online
• DHS. 2017. Fiscal Year 2016 Entry/Exit Overstay Report, US Department of Homeland Security Washington DC. Available Online
• Griffith, P. and Morris, M. 2017 Forthcoming. An immigration strategy for the UK: Six proposals to manage migration for economic success. Institute for Public Policy Research (IPPR), London.
• Giuntella, O., Kone, Z., Ruiz, I., and Vargas-Silva, C. 2017. Reasons for Immigration and Immigratnt’s Health. SSRN Working Paper. Available Online
• House of Commons. 2016. European Union Referendum 2016. House of Commons Library Briefing Paper Number CBP 7639. London. Available Online
• JCWI. 2015. “No Passport Equals No Home“ An independent evaluation of the ‘Right to Rent’ scheme. Joint Council for the Welfare of Immigrants, London. Available Online
• Katwala, S., Rutter, J., and Ballinger, S. 2016. Britain’s immigration offer to Europe How could a new preferential system work?. British Future. London. Available Online
• LCCI. 2016. Permits, Points and Visas: securing pracitcal immigration for post-Brexit London. London Chamber of Commerce and Industry, London. Available Online
• MAC. 2012. Limits on Migration Limit on Tier 2 (General) for 2012/13 and associated policies. Migration Advisory Committee, London. Available Online
• MAC. 2017. Assessing labour market shortages A methodology update. Migration Advisory Committee, London. Available Online
• Migration Observatory. 2011. UK Public Opinion towards Migration: Determinants of Attitudes. Migration Observatory briefing, COMPAS. University of Oxford. Available Online
• Migration Observatory. 2014. Scottish Public Opinion. Migration Observatory briefing, COMPAS. University of Oxford. Available Online
• Pandey, M and Townsend, J. 2010. Provincial Nominee Programs: An Evaluation of the Earnings and Retention Rates of Nominees. University of Winnipeg Department of Economics Working Paper Number: 2010-01. Available Online
• ONS, 2015. Annual Mid-year Population Estimates: 2014. Statistical Bulletin. Office for National Statistics. London. Available Online
• ONS. 2016. Subnational population projections for England: 2014-based projections. Statistical Bulletin. Office for National Statistics. London. Available Online
• ONS. 2017. The spatial distribution of industries in Great Britain: 2015. Article. Office for National Statistics. London. Available Online
• Sumption, M. 2014. Giving Cities and Regions a Voice in Immigration Policy: Can National Policies Meet Local Demands. Migration Policy Institute. Washington DC. Available Online
• The Scottish Government. 2016. Scotland’s Place in Europe. Scottish Government, Edinburgh. Available Online