Stonewall in a national park

Home / publications / primers /

Sub-National Immigration Policy: Can it Work in the UK?

18 Sep 2013


Introduction: the referendum context

Scotland’s independence referendum will take place on 18 September 2014. If the majority of the population aged 16 and older vote “yes”, this will be interpreted by the Scottish Government as a mandate to leave the political union of the United Kingdom and become an additional European Union (EU) member state. If Scotland does achieve independence it will need to construct political, social and economic institutions that it currently does not have. This goes well beyond deciding what currency to use or how to pay out welfare benefits and state pensions. At present, immigration is a “reserved power”. Immigration policy is decided by the UK Government and the immigration system is managed by the Home Office (formerly by the UK Border Agency). The Scottish Government plays no significant role in immigration matters.  Therefore, an independent Scotland will need to put in place an immigration system and formulate an immigration policy—presumably one that would focus on Scotland-specific needs and interests.

An independent Scotland will still be constrained with respect to immigration policy if it becomes a member of the EU. The EU has clear expectations on what is required of “new” member states with to respect immigration. These expectations are embodied in the “Lisbon Treaty”, which are the rules and regulations reinforced by Directives and Regulations. These are the most important forms of binding EU law issued by the EU Council and European Parliament. Others will debate about whether Scotland will make a smooth transition or have to reapply to be the 29th EU member state. This paper assumes that either outcome will require an independent Scotland to “toe the line” with respect to EU requirements. However, as has been argued elsewhere, adopting EU immigration requirements will not result in an immigration system that will hamper Scotland’s ability to compete in the international labour market for highly-skilled workers (Angrisani and Wright, 2013).

Although EU law makes it more difficult to envision this outcome, an independent Scotland might be allowed an “opt-out” from the Schengen membership and in that case might retain its present arrangements for free movement among Scotland, the rest of the UK, and Ireland. Membership in Schengen implies no internal borders with individuals being able to travel between countries without a passport and border controls. The UK is not part of Schengen and has stated that it has no intention of joining. Instead, the UK along with Ireland form the “Common Travel Area” (CTA), which  works in a similar manner to Schengen, allowing unrestricted movement between these two countries. An independent Scotland might try to become the third member country of the CTA.

But, what if the majority of the Scottish population votes “no”, as the opinion polls suggest?  If this is the case, the present policy dilemma regarding immigration remains: the UK Government and the Scottish Government currently have very different views on immigration. The UK Government is committed to reducing net-migration in the UK from “the hundreds of thousands to the tens of thousands” by the end of their first parliament (2015). The Scottish Government, however, wants to maintain the currently historically high levels of net-migration driven mainly by historically high levels of immigration in part to achieve its population growth target and ensure longer-term growth in labour force. As the former First Minister Jack McConnell was fond of putting it: “We need to grow the population to grow the labour force to grow the economy.”

Back to top

Policy dilemma: Can immigration policy accommodate regional differences?

Hence a policy dilemma arises, if Scotland remains either a part of the UK or a Common Travel Area with the UK and Ireland. Presuming that the Scottish and UK Governments continue to have different aims for immigration policy, the practical policy question becomes: How, if at all, is it possible to increase immigration to Scotland as the Scottish Government wants (particularly for highly-skilled immigrants) and at the same time reduce immigration to the United Kingdom (as the UK Government wants)? At first these policy objectives may appear to be totally incompatible. Immigration policy is set for the UK as a whole by the UK Government and any policy that reduces immigration to the UK as a whole will also reduce immigration to Scotland, as long as Scotland is a part of the UK.

This sense is reinforced by the lack of attention to differences in demographic and other conditions in different part of the UK in current immigration policy. There is little in the current points-based system that address differences among the regions and countries of the UK. The only notable differentiation of Scotland in current immigration policy is the Scotland-specific “shortage occupation list”. If job is on the shortage occupation list then an employer can offer the job to non-EU nation without having to advertise the job nationally. There is a list for the UK as a whole but in addition there is a list for Scotland, to enable Scottish employers facing Scotland-specific labour shortages to more easily fill these jobs with non-EU nationals.

Back to top

Adding regionality to a points-based immigration policy: The Canadian example

In February 2005, the then Home Secretary Charles Clarke outlined a “five year plan” aimed at changing fundamentally the way immigration to the United Kingdom is managed. Central to this plan was the adoption of a “points-based system” (PBS), where applicants are allotted points or “scored” for possessing human capital characteristics that make them more employable, such as education, technical skills, and work experience. If some threshold level of points is achieved (which can be varied), then the individual is entered into a pool of individuals who will eventually be allowed to immigrate to the UK (usually conditional on satisfactory security and medical checks). With such a system the policy shifts away from matching “jobs to people” to matching “people to jobs”. Such a shift is desirable since immigration can be used to strategically fill job vacancies and help plug skills gaps caused by population ageing and labour force decline. The UK’s system is a variant of the system introduced in Canada in 1967 and copied by Australia in 1973. However, recent changes (such as the elimination of the Tier 1 General route for highly educated and/or high earning workers without job offers) have reduced the human capital component of the PBS (particularly around labour migration) while leaving in place more of the employer-driven elements of the system.

One way of assessing the possibilities for a Scotland-specific immigration policy within the UK or British CTA is to examine Canadian policy, with its comparable Points-Based System. Unlike the UK, regional differences are a key feature of the Canadian immigration system. All the ten provinces of Canada (and one of its three territories) have agreements with the federal (Ottawa) government relating to immigration which takes into consideration specific provincial (territorial) requirements. Beginning in the late 1990s, “Provincial Nominee Programmes” (PNPs) have been established. PNPs are negotiated agreements that essentially mean that responsibility for immigration is shared between the provincial and federal governments. Similar agreements exist between the territorial and federal governments in Australia, although regionalisation is less central in Australian immigration policy.

In practise these programmes mean that applicants with certain skills face a lower immigration threshold if they agree to live, work and stay in a particular province/territory for a minimum period of time. This minimum period of time is often 1,095 days of residence, which is also what is needed to be eligible for Canadian citizenship. Once citizenship is obtained (or the minimum period expires), the individual can reside anywhere in Canada. One of the main reasons PNPs were introduced was to counter the historical tendency of immigrants to concentrate in the three main cities of Toronto, Montreal and Vancouver, much as immigrants to the UK tend to concentrate in London. They are based on the empirical regularity that once an immigrant arrives in one province, after two years of residence, the probability of moving to another province drops off considerably. In other words, if migrants move to a particular region in the first place, with a period of required residence, there is an elevated probability that they will stay for the long-term.

The Canada-Quebec Accord (CQA) goes one step further and essentially devolves responsibility for immigration to the province of Quebec. In this arrangement, potential immigrants apply directly to the Province of Quebec and not the Dominion of Canada. The CQA is also a points-based system. However, the weighting is different, with more emphasis on language and less on human capital and employability. Essentially the CQA system awards fewer points for education/qualifications/employability and more points for knowledge of the French language. Quebec “picks” the immigrants and the federal government issues the visas and work permits, and administers the medical and criminal background checks.

In principle, the UK Points-Based System could easily be modified along these lines to address Scotland’s distinct situation relative to the rest of the UK. The simplest modification would allot more points to applicants who agree to live, work and stay in Scotland. Immigrants who choose this option could be issued with a visa that states that they are only allowed to work in Scotland. To emulate the Canadian policy, the period of this permit could be set to match same amount of time needed to applying for citizenship.

Of course, this simple modification will only work if the government is willing and able to enforce the terms of the residence requirement. Enforcement here might entail that those who fail to do so would have their work permit revoked, or lose the right to reside in Scotland or the UK and be subject to deportation if necessary. Even given the political will to do so, enforcement can raise significant practical difficulties.

Back to top

Does regionality work?

The meaning of the Canadian system for Scotland and the UK depends, of course, on whether Canadian regional-level immigration policy has achieved its goals. If the Provincial Nominee Programmes and the Canada-Quebec Accord “don’t work”—in the sense that people do not stay in the province or territory they agreed to immigrate to—then it would have little value as a precedent for Scotland. At face value, this seems unlikely, since the sanctions are considerable for breaking the immigration contract, but there is no substitute for empirical evidence. If the system did not work as planned, then one would expect to see high rates of internal migration between the provinces of Canada among international migration in the years immediately after they arrive.

This is a legitimate question that could be answered through empirical research. Unfortunately, a search of the academic literature revealed only one study that specifically addressed this issue. A number of studies, however, demonstrate the related point that the inter-provincial migration behaviour of native-born and foreign-born Canadians is surprisingly similar (e.g. Edmonston, 2002; Finnie, 2000; Lin, 1998; Newfold, 1996; Nogle, 1994; Robinson and Tomes, 1982).

The single directly relevant study was done by Mosca and Wright (2009, 2013). Using micro-data from various Canadian censuses, they employed a “difference in differences” methodology in order to examine the rates of interprovincial migration “before” and “after” the introduction of Provincial Nominee Programmes and Canada-Quebec Accord. The main hypothesis is that if the systems “work” then the rates interprovincial migration rates for immigrants should be lower after the introduction of the these programmes after other factors that influence mobility (such as age and education) are held constant. The analysis suggests that there appear to be no major differences in the “before and after migration rates” for immigrants. However, they do note that since the introduction of Provincial Nominee Programmes and Canada-Quebec Accord there has been a more equal distribution of immigrants across the provinces and territories. That through time there has been a reduction in the gap between provinces with respect to their share of the total population and their share of total number of immigrants entering Canada.

Back to top

Possible pitfalls: Enforcing and defending residency requirements

Canadian precedents show that it is possible to make immigration policy at the regional level within a national points-based system. This could be attempted in at least two ways. The responsibility for immigration could be devolved to the Scottish Government in the same way that it has been devolved to the Quebec Government. Likewise, some form of power-sharing arrangement could be agreed similar in principle to Provincial Nomination Programmes. Both possibilities, however, would mean a loss of control of what is a “reserved power” by the UK Government.

It can be argued, of course, that devolving the responsibility for immigration to Scotland will create a “back door” way of immigrating to England, and thus undermine UK immigration policy. The concern is that immigrants will agree to reside in Scotland but on arrival or shortly afterwards move to England or Wales. This will be true if the policy’s residency requirement is not rigorously enforced. Evidence from Canada suggests that there are few violations of this requirement and only a handful of deportations caused by such violations. Canada’s enforcement mechanism is that if the residency requirement is violated, the visa allowing employment is revoked and the individual loses all rights to social and welfare benefits. This is stated formally in a legally-binding contract. However, it is unclear given UK, Scottish and EU law, whether such a contract would be legal. And of course, perhaps even more importantly, enforcement poses challenges as a practical matter as well.

Another obstacle comes from EU law. It can be argued that tying employment to geographical residence is illegal under European Union laws. The “freedom of movements of workers” is a key feature of the acquis communautaire – the body of EU legislation that EU member states must adopt. As such any form of residence requirement would be illegal since it violates a “right” central to EU membership. But, while this may be true for citizens of the EU, it is not the case for immigrants from outside the EU or EEA. Non-EU immigration policy remains the responsibility of member states, with significant freedom to determine their own policies. In the majority of cases, immigrants only have the right to live and work in other EU member states after they become national citizens and hold EU passports. In fact, the EU has a system, known as the EU “Blue Card”, which aims specifically to increase the mobility for immigrants between EU member states. It attempts to attract highly qualified immigrants by supporting member states and EU companies’ efforts to fill gaps in their labour markets that cannot be filled by their own citizens, other EU nationals or legally resident non-EU nationals. Once a member state grants a Blue Card to an immigrant, after two years that person can move to a job in another member state in an unrestricted manner—regardless of the passport they hold.

Another argument against regional immigration policy is that devolving the responsibility for immigration will result in Scotland attracting “lower quality” immigrants since the immigration hurdle will be lower relative to the rest of the UK. This would of course be true if the potential supply of immigrants was small and Scotland and the rest of the UK were somehow competing for the most qualified among them. However, given the number of potential immigrants, this claim becomes more difficult to sustain. Scotland’s share of the global population is around 0.08 per cent (5.3 million out of a total of 7 billion). Since the number of immigrants to Scotland relative to the potential supply of immigrants to the UK is so small, there would seem to be little chance of a watering down effect caused by the transfer of responsibility.

Back to top

Conclusion: Precedent exists for regional differentiation of immigration policy

As it stands at the moment, the UK points-based system does not include any provisions that would help the Scottish Government to deliver on its promise of increasing Scotland’s population and economic growth through immigration. The UK Government’s electoral promise to reduce immigration to the UK can also be expected to reduce immigration to Scotland.  Yet Scottish specificity could be built in to the Points-Based System through bonus points or lower thresholds for those who agree to live, work and stay in Scotland for a minimum period of time. Alternatively, the responsibility for immigration could be transferred to the Scottish Government along the lines of the Canada-Quebec Accord.

It is fair to state that there is a considerably amount of demographic diversity in most member states of the EU, and the UK is not an exception to this generalisation. Scotland and the North of England are facing population decline or slower growth, while the populations of London and the South East are growing rapidly. One might expect an immigration system that takes into consideration these demographic differences would work more effectively than a country-wide system. As was recognised by Canada, a “one-size-fits-all” immigration system attracts immigrants to areas with high immigrant concentrations. This is expected since chain migration is a feature of unrestricted or unmanaged migration flows. In order to break this link immigration policy needs to be more prescriptive in terms of where immigrants settle. Both Canada and Australia have systems that do this so. Therefore, if Scotland remains a part of the UK or part of a common travel area including the UK, it is possible to modify the current UK immigration system to seek to accommodate Scotland. Of course, policies that have worked in one nation may not work the same way in a different context, so while regionally-differentiated immigration policy has been implemented elsewhere, it remains to be seen if it will function the same way in the UK.


  • Angrisani, M. and R.E. Wright. “Scotland Post-independence Immigration System: How Might it be Shaped by European Union Requirements?” Economic Commentary, forthcoming 2013.
  • Coldwell, D., K. Lisenkova and R.E. Wright. “The UK “Immigration Cap”: Implications for Scotland.” Economic Commentary vol. 35, no 2 (2011): 54-61.
  • Edmonston, B. “Interprovincial Migration of Canadian Immigrants.” Working Paper No. 02-10, Vancouver Centre of Excellence for Research on Immigration and Integration in the Metropolis, British Columbia, 2002.
  • Finnie, R. “Who Moves? A Panel Logit Model Analysis of Inter-provincial Migration in Canada.” Working Paper No. 142, Business and Labour Market Analysis Division, Statistics Canada, Ottawa-Gatineau, 2000.
  • Lin, Z. “Foreign-Born vs Native-Born Canadians: A Comparison of Their Inter-Provincial Labour Mobility.” Working Paper No. 114, Business and Labour Market Analysis Division, Statistics Canada, Ottawa-Gatineau, 1998
  • Mosca, I. and R.E. Wright. ”Building Regionality into Immigration Policy: Does it Work? Evidence from Canada.” Department of Economics, University of Stathclyde, Glasgow, 2013.
  • Mosca, I. and R.E. Wright, “Devolved Immigration Policy: Will It Work in Scotland?” Economic Commentary vol. 33, no. 2  (2009): 55-60.
  • Newfold, K. B. “Internal Migration of the Foreign-born in Canada” International Migration Review vol. 30, no. 3 (1996): 728-747.
  • Nogle, J. M. “Internal Migration for Recent Immigrants to Canada” International Migration Review vol. 28, no. 1  (1994): 31-48.
  • Robinson, C. and N. Tomes. “Self-selection and Interprovincial Migration in Canada” The Canadian Journal of Economics vol. 15, no. 3 (1982): 474-502.



Sub-National Immigration Policy: Can it Work in the UK?

Press Contact

If you would like to make a press enquiry, please contact:

Rob McNeil

+ 44 (0)7500 970081

 Contact Us 


This Migration Observatory is kindly supported by the following organisations.

  • University of Oxford logo
  • COMPAS logo
  • Esmee Fairbairn logo
  • Barrow Cadbury Trust logo
  • Paul Hamlyn Foundation logo