57.The European Free Trade Association is an intergovernmental organisation set up for the promotion of free trade and economic integration on the part of its four Member States–Iceland, Liechtenstein, Norway and Switzerland. The European Economic Area (EEA) unites the EU Member States and three of the EFTA States (Iceland, Liechtenstein and Norway) into an internal market governed by the same basic rules. One EFTA member, Switzerland, has not joined the EEA, but instead has a series of bilateral agreements with the EU which allow it to participate in the internal market (although with less access than EEA members). There have been suggestions that the UK should seek to join the EEA or negotiate a bespoke EFTA style agreement with the EU in order to retain single market access.
58.Zsolt Darvas told us that EEA membership was the only model currently open to the UK, in which it could be outside of the EU but retain single market access. However, membership of the EEA requires the full transposition of EU laws, regulations and product standards without having a vote on how these are decided and implemented, involves the ultimate interpretation of the European Court of Justice and requires ongoing payments into the EU’s budget. Zsolt Darvas explained:
As an economist, I think an EEA arrangement would be good for the UK and good for the rest of the EU27, but I understand that for political reasons this is not the option that the United Kingdom will likely choose.
59.Sir Ivan Rogers suggested to us that EEA membership, whilst “implausible as the medium, longer-term destination”, could offer a short-term, transition route out of the EU. The drawback would be “rule-taking” of “significant chunks” of the single market acquis, but the potential advantages for the UK would be “not in a customs union, clear sovereignty and autonomy on trade policy, as the EFTA states have and they have their own free trade agreements”. EEA countries are subject to free movement rules, but witnesses suggested that there might be more potential for the UK to control free movement as an EEA member or as part of a bespoke EFTA-style agreement than as an EU member.
60.The Prime Minister has stated that the UK will not seek to join the EEA or EFTA but will instead pursue a bespoke partnership with the EU. Nonetheless, until the negotiations are complete, an EFTA-style option remains, either as a transitional step or as a longer-term option and, as we set out below, it also contains the possibility of safeguard provisions that may merit consideration as part of future policy for managing EEA migration.
61.The Citizens’ Rights Directive grants to citizens of the EU and their family members the right to move and reside freely within the territory of the EU Member States, where previously such rights were limited to people who were economically active, and circumscribes the possibility of stronger controls on EU migration. While EEA members participate in the single market including free movement, those not in the EU are not bound by the same terms of the EU Citizens’ Rights Directive. The Citizens’ Rights Directive has been part of the EEA Agreement since March 2009 but its incorporation was accompanied by a Joint Declaration setting out reservations in the areas of Union Citizenship and immigration policy. Karin Fløistad, of the European University Institute and Norwegian law firm Simonsen Vogt Wiig, draws attention to these reservations but cautions “Whether they are sufficiently different to be of interest to the UK and the EU for future association remains to be seen”. In Professor Meardi’s view the key principle is that the Citizens’ Rights Directive is not fundamental to the Single Market. He told us:
In the case of the EEA, or any single market deal outside the European Union, there will be some important changes in what can be done and what cannot be done. In particular, the Citizens’ Rights Directive would not apply. It does not apply to the single market; it is within the whole citizen pillar of the European Union. Therefore, any movement for reasons that are not for work or business could be entirely regulated unilaterally by any country.
62.Switzerland is not in the EEA; its relationship with the EU is governed by a series of bilateral agreements within which it was able to negotiate the right to initially apply unilateral safeguards to the movement of people. Sir Ivan Rogers told us that EFTA countries are not bound to “the full panoply of citizens’ rights that you have within the EU”.
63.The EEA Agreement contains a safeguard clause embedded in Articles 112 and 113, which provides that in the circumstances of “serious economic, societal or environmental difficulties of a sectoral or regional nature”, a contracting party may unilaterally take appropriate “safeguard measures” to restrict the rights in the Agreement. This means that an EEA member can apply an emergency brake on free movement from other EEA countries. The EEA contracting party must notify and consult with the rest of the EEA. The agreement makes it clear that if measures taken create an “imbalance between the rights and obligations under this Agreement” then, under Article 114, the other EEA members are entitled to take rebalancing measures.
64.In practice, the procedural requirements for initiating a safeguard clause are stringent and subject to consultation. Evidence must be produced that immigration was causing “serious economic, societal or environmental difficulties”, and reciprocal restrictions can be instituted if the measures taken by an EEA member are not well-justified. However, Guy Verhofstadt suggested that the EEA arrangement was significantly different to that for the EU:
in the single market the European Commission can give the green light to use the emergency brake. In the EEA that can be done by the member state itself. I think that is a huge difference.
Sir Ivan Rogers told us:
Can you argue, on the basis of Article 112 of the EEA Agreement, there is something you could do? I think you can argue that it must in principle be different because obviously it is a provision that does not exist in the [Lisbon] Treaty. [ … ] The very fact that the safeguard clause exists suggests that you must be able to use it.
65.Since 1998, Liechtenstein has exercised a brake on free movement on the basis that its geography makes it particularly vulnerable to high population flows. While ostensibly derived from the emergency brake provisions of the EEA Agreement, the Liechtenstein arrangement has effectively become permanent. To many the Liechtenstein example is too small to be relevant, but the Leave Alliance believes that it sets an important precedent. In its paper Single Market participation and free movement of persons: The use of EEA Safeguard Measures the Leave Alliance states that whilst the numbers involved are very small, “what matters is that a precedent has been set within the framework of the EEA Agreement for suspending freedom of movement in respect of a single country, and replacing with a quota system for what amounts to an indefinite period”.
66.Switzerland’s bilateral agreements with the EU deliver partial access to the single market and a closer economic relationship than any state outside the EEA. The Agreement on the Free Movement of Persons (AFMP), signed in 1999, lifted the restrictions on EU citizens wishing to live or work in Switzerland. As part of the agreement, Switzerland could unilaterally invoke a safeguard clause allowing it to reintroduce quotas on EU immigration for a specified period if inflows exceed a certain threshold.
67.In 2012, Switzerland unilaterally capped long-term residence permits for nationals from A8 Member States at 2,180. In announcing the reintroduction of quotas, the Federal Council said:
In invoking the safeguard clause, the Federal Council is seeking to apply one of the means at its disposal to control the immigration flow into Switzerland.
A year later, Switzerland imposed a cap of 53,700 on permits for citizens of the A15 Member States. Throughout this period Switzerland’s bilateral arrangements meant that it effectively participated in the single market for goods at the same time as operating caps on immigration. The right to impose immigration caps expired in 2014, after which point the AFMP stipulated that freedom of movement would apply in full to all citizens of the EU and EFTA states. In February 2014, 50.3% of Swiss citizens approved a referendum “against mass immigration”, requiring Swiss authorities to reintroduce annual immigration quotas in contravention of the AFMP. The result prompted a three-year renegotiation of the agreement which led to amended rules regarding EU migrants’ access to its labour market.
68.The new law, implemented from July 2018, stipulates that, in sectors where the unemployment rate exceeds a specified rate, Swiss residents (both Swiss and non-Swiss nationals) registered at a job centre be given five days advance notice of any job vacancy and that the employer who advertises a job will automatically receive the records of the unemployed workers in the region. During this period, the employer is banned from publishing a job advertisement elsewhere and subject to a fine if they do. If a suitable job applicant is not found within this period, the employer can continue with its usual recruiting process and may publicly advertise the job. Employers will remain free to hire a person residing in the EU, without having to justify the choice not to hire a local resident. The EU did not object to the Swiss law and did not see it as a challenge to the applicability of free movement of people.
69.The existing safeguard measures available to EFTA states as part of their trade relationships with the European Union demonstrate that they can—in principle—exercise more controls on immigration while participating in the single market than are available to EU Member States. Were the Government to change its red lines, such arrangements might provide a basis for drawing up means of controlling EEA migration from within the single market.
70.‘Transitional controls’ have been available to EU Member States in recent expansions of the European Union. For example, under the Treaty of Accession 2011 between Croatia and the EU, Member States were able to restrict the access that Croatian citizens had to their labour markets for up to seven years. The restrictions have meant that, unless an exemption applied, Croatian citizens needed permission from the Home Office to work in the UK. The restrictions can be applied initially for five years, plus an additional two years if required to protect the Member State’s labour market from serious disturbance. The Home Office recently took the decision not to extend the controls on Croatian citizens.
71.Ahead of the EU referendum, the former Prime Minister David Cameron sought reforms to the UK’s relationship with the EU including the extent to which the UK was subject to the full scope of Freedom of Movement rules. The final deal, agreed in February 2016, would have allowed the UK to:
72.Sir Ivan Rogers, then the UK’s ambassador to the EU, reports that the then Prime Minister had initially wanted an agreement that curbed the number of EU and EEA nationals moving to the UK, but that it ‘quickly became clear’ that quotas or an ‘emergency brake’ would not get agreement from the other Member States.
73.Zsolt Darvas described the Cameron deal as “probably very close to the limit of what was possible under the current EU Treaty”. Mats Persson told us that he thought it was a good agreement. He explained that at the outset of the negotiations the then Prime Minister was told it would be impossible to achieve changes to secondary legislation that would differentiate between EU and UK citizens for the purpose of accessing benefits, “but we got that and the Commission, for example, showed a lot of flexibility and creativity in trying to achieve a political compromise”.
74.Free from free movement obligations, the Government could potentially go further than the pre-referendum negotiations conducted by David Cameron, for example, by restricting access to all benefits for the first few years of an EU national’s life in the UK. Professor Catherine Barnard argues that a close reading of the text of the deal suggests that other EU leaders acknowledged space for restricting migration. She suggested that widening the definition of the scope of limitations placed on the free movement of workers from “public policy, public security or public health” to include the reduction of unemployment seems to have been a fairly major concession to the UK and that “It is surprising that David Cameron did not trumpet his achievement here more loudly”.
75.Professor Meardi explained to us that Cameron had failed to obtain an emergency brake in the pre-referendum negotiations because it was incompatible with the EU, but that an emergency brake for the UK “was not incompatible with the EEA, or the single market outside the EU”. He suggested that if the UK chose to remain in the single market, either in the EEA or a new arrangement, safeguard measures would be possible:
It is possible to use all possible regulatory measures that the single market allows and have a relatively controlled migration regime, which is probably not so different in terms of effect from no single market and no free movement regime, where you will still probably need some fast reaction work permit system in order to allow the needs of business or economies with a different population.
76.Overall there are a range of existing models and precedents which involve participation in the single market, but with greater immigration controls than under full EU free movement. In the course of our inquiry, we heard evidence from witnesses who have drawn on these examples and made proposals for specific safeguard measures or immigration controls that may be applicable to a future UK-EU trading relationship within the single market.
77.In our inquiry, we heard proposals that would permit the application quantitative restrictions on the free movement of people during periods of high inflows while allowing the UK to continue participating in the single market.
78.Professor Michael Ambühl, a former Swiss State Secretary who was chief negotiator of the Bilateral II Agreements between Switzerland and the EU, gave evidence to us on his proposal for an emergency brake which would allow for “regulatory measures if statistically exceptionally high net migration numbers are encountered”. Based on the existing safeguard clause found in article 14(2) of the Swiss-EU free movement agreement, Professor Ambühl’s proposed model would allow the activation of an emergency brake if inward migration from the EEA was exceptionally high relative to other EU member states. The threshold would be based on concrete and clear criteria. He explained:
What I would propose is to have a clear definition of what the trigger should/could be, what would enable the country, if the conditions are fulfilled, to go unilaterally—without retaliatory measures from the other side—to take appropriate measures. The appropriate measures would also have to be predefined so that it is useful. [ … ] I think there would be slightly an advantage to negotiate—if one wants to negotiate and if the EU would also be willing to go into such a negotiation—and to do it within a new framework, a bespoke framework of a bespoke agreement and not within an existing given framework, such as the EEA.
79.As Professor Ambühl’s paper sets out, were relative inward flows to exceed the threshold, the UK would be permitted to adopt measures to:
Calculations accompanying his model indicated that the comparatively high levels of EEA migration to the UK during the Eurozone crisis in 2013 to 2015 would have justified measures to reduce net migration in 2016 by tens of thousands.
80.We heard from Professor Catherine Barnard that “With some imagination, it could be possible to design an emergency brake which took account of regional criteria, looking at county level indicators”. She told us that ‘buried in the text’ of the ‘New Settlement Agreement’ negotiated by David Cameron in February 2016 was language which might permit “some sort of not just emergency brake on benefits, which is very clear on the face of the text, but even an emergency brake on actual migration”. We heard that:
There is language in the text that talks about where there is a need to reduce unemployment or protect vulnerable workers or avert the risk of seriously undermining the sustainability of social security systems, it might be possible to impose discriminatory restrictions. This is not fleshed out in a lot of detail but there is language there that does create a bit of wriggle room.
81.Professor Barnard proposes that an emergency brake mechanism could be devised “at the level of devolved administrations or other regional groupings, to take account of the substantial variation in the needs of the regions”. She explains:
Relying on both economic data (such as labour market criteria e.g. relative levels of unemployment, demands for unemployment benefits, wage levels), demand for public services (e.g. population growth, population churn, waiting lists) and political experience (e.g. what constituents are saying through the ballot box and in person at surgeries), these regions could make a request to national government to impose restrictions on migration for a time limited period. These restrictions might be sectoral, based on skill levels, or more general and for a defined period of time.
82.In its strategy for negotiations with the European Union, the Government has not considered the range of possible safeguard provisions that could be applied to a trade agreement that allowed the UK to participate in a single market after Brexit, which would combine new immigration controls and maintain economic benefits. It should immediately do so.
83.Professor Jonathan Portes submitted that “quantitative targets and caps, either for net migration overall (as with the Government’s target) or for highly skilled workers (as with the cap on Tier 2 visas) have no place in a sensible immigration policy”. Instead, he argues in favour of a “Swiss-style system of temporary and targeted regional and/or occupation specific controls” that would “enable a targeted, temporary and proportionate response to migration pressures”.
84.Professor Portes suggests that the Swiss system of prioritising domestic recruitment could provide a potential model for the UK. He explains that:
In the UK context, it would be, for example, possible to state that in sectors and/or regions where unemployment was high, or significantly above the national average, or where wages were low and/or falling, employers would be required to recruit via Jobcentres, and to give initial preference to those claiming unemployment-related or disability benefits. This would, by definition exclude, recent arrivals from elsewhere in the EU (although not longer-term non-UK residents).
85.Professor Barnard and Sarah Fraser Butlin propose a scheme, broadly based on accession arrangements for Croatian nationals coming to the UK, in which employers issue a job offer certifying that the position pays above a minimum threshold with a minimum number of hours. The prospective employee is then able to apply for a residence permit at negligible cost. Their proposals leave employers with the power to determine who they want and in what field, subject to a salary threshold, rather than being subject to quotas and criteria drawn up by the Home Office, and avoid the need for a bureaucratic visa process. They further propose that an individual should have to work for a period of three months before they can bring family members with them (parents, spouse, dependent children). The Barnard and Butlin paper makes reference to a leaked draft Home Office proposal of a salary threshold of £20,500.
86.The IPPR has suggested that jobseekers from the EU should no longer have the right to reside in the UK unless they already have a job offer. There would be little to prevent an EU national from entering the UK as a visitor, searching and applying for jobs, returning home, and then if successful coming back to the UK with the right to reside as a worker. Professor Portes argues that a ‘prior job offer’ system, even if it were less restrictive than the current Tier 2 model for non-EEA migrants, would be likely to impose a significant bureaucratic hurdle for employers, would present significant enforcement issues and that there would be a risk of intermediaries (employment agencies) being set up purely to offer jobs to EEA nationals and potentially facilitating abuse of the system. The Government has proposed a “mobility framework so that UK and EU citizens can continue to travel to each other’s territories”.
87.While we are not recommending any particular model for future migration from the EU, we do note that—based on the evidence we have received—there are options for controlling migration within the single market which go much further than the previous Prime Minister’s negotiation with the European Union. We recognise that these options have not been the subject of negotiations between the UK and the EU, and that negotiations would be complex, but we believe these options should be explored.
71 Financial Times, , 13 May 2015
75 Simonsen Vogt Wiig, , 4 January 2016
76 Karin Fløistad, European University Institute
81 Migration Watch, , 31 August 2016; Q225
82 Leave alliance, , 27 July 2016
83 Swiss State Secretariat for Migration (SME),
84 Swiss State Secretariat for Migration (SME), , 18 April 2012
85 On 8 December 2017, the Swiss Federal Council confirmed that the new law’s implementation would take effect on 1 July 2018.
86 KPMG, , 30 November 2017
87 HM Government, , 19 March
91 Centre for European Legal Studies, 14 September 2016
93 Professor Michael Ambühl and Daniela S. Scherer,
95 Professor Michael Ambühl and Daniela S. Scherer,
96 Q16 and see paper Professor Catherine Barnard and Sarah Fraser Butlin, , 25 June 2018
99 Professor Catherine Barnard and Sarah Fraser Butlin, , 25 June 2018
100 Professor Jonathan Portes,
101 Professor Jonathan Portes,
102 Professor Catherine Barnard and Sarah Fraser Butlin, 19 June 2018
103 IPPR, (2017) P3
Published: 31 July 2018