The issue of migration was a primary focal point for the referendum campaign. In his renegotiation with the EU, David Cameron had attempted to find a mechanism that would reduce incentives on EU nationals to come to the UK. The mechanism that was agreed stopped short of the sort of cap on numbers or ‘emergency brake’ which had been discussed. This was important given that David Cameron’s Conservative Party had been elected to government in 2015 with its election manifesto continuing to pledge to reduce net migration to the UK to the ‘tens of thousands’ as set out in its 2010 manifesto. Yet it was clear that the government was significantly adrift from its target. Despite having control over non-EU migration, overall net migration was increasing and immigration from the EU was playing a role in the overall figures. Meanwhile, the refugee crisis had also added to anxieties about movements of people across borders, including border-free movement within the EU’s Schengen area.

Both Vote Leave and the United Kingdom Independence Party (UKIP) argued that outside of the EU, the UK would be able to exert greater control over its borders. Anxieties about migration were fuelled by sections of the print media that displayed a daily obsession with ‘migrants’. These anxieties were exploited by UKIP during the campaign with its controversial ‘Breaking Point’ poster. It was also claimed by Leave campaigners that European courts made it difficult or impossible to deport criminals, though it was unclear whether this was principally an EU law or European Convention issue. The concerns were often contradictory: were migrants in the UK to take jobs or take welfare benefits? Migrant labour was also blamed for placing pressures on public services, and yet economically active migrants contribute to the tax revenues that pay for public services as well as providing a source of labour for valued public services like the National Health Service.

The Remain campaign claimed that leaving the EU would not in fact increase control over net migration especially if the ‘points-based’ immigration system proposed by UKIP and supported by prominent Vote Leave campaigners including Boris Johnson, Michael Gove, Priti Patel and Gisela StuartFootnote ¹ was put in place. There was also an attempt to shift the focus of debate around free movement onto the benefits that arose from EU-facilitated free movement: low-cost air travel, reduced mobile phone roaming charges and access to emergency healthcare as a tourist. Nonetheless, these arguments gained little traction. They tended to emphasise episodic, even optional, benefits that were, moreover, experienced by travelling outside the UK to EU countries. They didn’t speak to the experience of life in the UK itself.

By the time of the June 2016 referendum, estimated annual net migration to the UK – the number of people moving to the UK minus those leaving the UK – stood at 335,000 people. The estimated total number of EU nationals moving annually to the UK was at its highest at 284,000 people. Net migration of EU nationals was estimated at 189,000 or just less than the net estimate of non-EU nationals.Footnote ² If we differentiate between migrants from states that were Member States prior to the 2004 EU enlargement – the EU15 – and those from the post-2004 enlargement of the EU – the EU8 – and subsequent enlargement to include Bulgaria and Romania – the EU2 – we can identify some trends. Of these, perhaps the most important is the very significant spike in migration from Bulgaria and Romania following the lifting of transitional restrictions in 2014.

Up until the early 2000s there was a broadly equitable balance between people leaving the UK and EU15 nationals moving to the UK. Migration from the EU15 began to grow from this period and accelerated following the financial and economic crisis with estimated EU15 migration to the UK reaching 138,000 or 49 per cent of the total EU migration to the UK by the end of 2015. In net terms in the year preceding the referendum, 84,000 EU15 nationals were estimated to have moved to the UK. Immigration from the EU8 to the UK peaked in 2007 at 112,000 falling to an estimated 73,000 in the year preceding the referendum (or 42,000 in net terms). However, this fall was offset by the sharp spike in immigration from the EU2 once transitional restrictions on Bulgarian and Romanian nationals ended. By the time of the referendum, estimated annual net migration from the EU2 was 61,000; some 25 per cent of the total EU migration to the UK. Taken together, long-term immigration from the EU was rising from 2012 so that by the time of the referendum, just less than half of immigrants to the UK came from within the EU.Footnote ³

Tim Shipman’s detailed and insightful analysis of the background to the EU referendum suggests that had David Cameron’s new settlement negotiations sought and obtained an ‘emergency brake’ on the numbers of EU migrants coming to the UK, this might have tipped the balance in favour of the UK remaining in the EU.Footnote ⁴ At least one EU academic lawyer suggested that perhaps more could have been done to emphasise the power of states to derogate from free movement principles on treaty-derived grounds of public policy,Footnote ⁵ although it does need to be acknowledged how limited this derogation is in the context of direct discrimination against nationals from other EU states. The decision to focus instead on restricting access to in-work benefits became regarded as an illustration of civil servants and lawyers restricting the Prime Minister’s room for manoeuvre. This narrative subsequently gained further ground with the resignation of Sir Ivan Rogers, the UK ambassador to the EU on 3 January 2017. Sir Ivan and others’ advice as to what was politically and legally feasible was important to David Cameron’s approach to the negotiations. Following the referendum, Sir Ivan’s attitude of speaking ‘truth to power’ was regarded by some Brexit supporters as mere ‘pessimism’. But it wasn’t a lack of optimism or a failure of negotiating strategy that meant that Prime Minister Cameron opted for controls on access to benefits. A push for treaty change would have significantly altered the timeframe for the referendum.

However, the appeal of the UK as a destination for EU nationals has largely been its labour market rather than its welfare system. By the time of the referendum, 57 per cent of EU migrants had a definite job. The percentage of EU15 and EU8 migrants with a definite job had, however, declined from the previous year.Footnote ⁶ Yet this migrant entry to the labour market was not at the expense of UK nationals looking for work. Looking at the data, although UK employment rates fell following the financial and economic crisis, in the period from 2012 (during which time EU migration rose), employment rates increased from 70.3 to 74.5 per cent.Footnote ⁷ Migrant labour was accessing new jobs, not least as chains of coffee and sandwich shops expanded their networks across the UK. In other words, there is no fixed reservoir of jobs but rather a constantly shifting labour market. Losses of employment are associated with changes in technology and global competition rather than a burgeoning number of EU-national baristas.

For those EU nationals without a job but seeking employment, EU law itself does not require Member States to provide financial support to job-seekers within the first three months of residence and, thereafter, the right to invoke the principle of equal treatment to access social assistance has been qualified and limited by the Court of Justice in several rulings. In the week before the EU referendum, the Court rejected a European Commission complaint that the UK’s habitual residence test – when applied to in-work child benefits and tax credits – violated EU rules. Even without changing the resources of EU law available to the UK government through the ‘new settlement’, a combination of domestic rules and the exploitation of existing capacities within EU law afforded the UK government opportunities to exercise certain types of control over access to benefits for migrant EU nationals.Footnote ⁸

However, two other assertions were often made about the effects of increased immigration to the UK: it depressed wages and put public services under strain. During the referendum campaign, former Work and Pensions Secretary and leading Leave campaigner Iain Duncan-Smith suggested that UK wages were 10 per cent lower because of rising EU immigration. But as the Centre for Economic Performance reported, this dramatic fall in wages during this period was a consequence of the post-2008 recession and not increasing immigration: coincidence is not causation.Footnote ⁹ Indeed, right across the EU, reductions in wages was a common response to the crisis as a means of averting more significant job losses.

As for public services like education, transport, health and social care, these are typically delivered at a local level. Any impact of immigration on such services is difficult to evidence let alone to break it down as between EU and non-EU migration. Under the last Labour government, a Migration Impacts Fund had been established in 2009 with £35 million available annually, funded by a £50 levy on immigration applications from non-European Economic Area (EEA) nationals. The distribution of funds between local authorities saw some areas receiving hundreds of thousands of pounds while others received substantially less. In any event, in 2010, the Coalition government withdrew the fund as a cost-saving measure.Footnote ¹⁰ Indeed, in the crucial time period leading up to the referendum, very significant pressure on public services was coming not from the effects of immigration but from government policies connected to deficit reduction and fiscal discipline. This more local dimension of immigration – explored further below – also draws attention to potential North–South and other disparities in terms of the geographic distribution of wealth, employment opportunities and pressures on public services.

In thinking about the locality of immigration, statistically it can be difficult to tell the composition of local populations. Census data is helpful but the last UK census was in 2011 and it is difficult to find sources of data post-2011 that distinguish between EU and non-EU migration flows. This time period is particularly sensitive. From 2004 to 2011, migrant workers from EU8 states had to register under the UK’s Worker Registration Scheme. For workers from Romania and Bulgaria, tougher restrictions applied for seven years up until the end of December 2013. From January 2014 onwards, nationals of Romania and Bulgaria also enjoyed unrestricted rights of free movement. So the period from 2011 onwards is particularly interesting but poses an issue in terms of data at a local level. One useful proxy indicator is National Insurance registrations (NINos) where not only is it possible to identify EU nationals, it is also possible to distinguish between nationals of the ‘old’ and ‘new’ EU states as well as the post-2004 EU8 and post-2007 EU2 states.Footnote ¹¹

Looking at individual counting areas with the strongest Leave votes – of more than 68 per cent – these were typically grouped in the West and East Midlands and the East and North-East of England. Local counting areas in the Tees Valley in the North-East all showed significant Leave votes with Middlesbrough, Hartlepool and Redcar and Cleveland each returning Leave results in excess of 65 per cent of the votes cast. In an area like Middlesbrough, while the 2011 census actually indicated a decline in the overall number of inhabitants – with the Tees Valley showing a net population outflow – its Black, Asian and Minority Ethnic (BAME) population proportion doubled and unemployment rose (with an employment rate well below the national average).

If we look at nationals from post-2004 EU states registering to work under the Worker Registration Scheme, the numbers in the Tees Valley were lower than elsewhere in the region and declined significantly from 2007 to the scheme’s closure in 2011. Indeed, in the post-2011 period the number of EU8 migrants resident in Middlesbrough, Hartlepool, Redcar and Cleveland and making National Insurance registrations were comparatively small and relatively stable. But following the removal of restrictions on nationals from Romania and Bulgaria from 2014, Middlesbrough showed a distinct increase in NI registrations from 25 in 2013 to 145 in 2014 and 434 in 2015.

While London voters overall voted for Remain – indeed seven of the top ten Remain constituencies were in London – the adjacent constituencies of Barking and Dagenham, and Havering returned Leave votes. Based on the 2011 census, Havering was the least diverse London borough with over 80 per cent of residents being of White British ethnicity compared to a London average of 45 per cent. Yet, as between the 2001 and 2011 census, Havering’s BAME population doubled, largely drawn from Africa, the Caribbean and Pakistan.Footnote ¹² While BAME voters might have been expected to have a preference to vote Remain, post-referendum data does suggest that in some London wards with high numbers of residents of Asian origin, a distinctive pattern of voting Leave can be detected.Footnote ¹³

Looking again at the 2011 census data, in Havering, only 2.3 per cent of residents reported holding an EU passport compared to 82.1 per cent holding UK passports. The highest percentage of worker registrations from the EU8 between 2004 and 2011 was 0.19 per cent of the working age population compared to a range of 0.39–0.54 per cent for London and 0.3–0.57 per cent for England as a whole.Footnote ¹⁴ From 2011 onwards, the number of migrants from EU8 states making National Insurance registrations in Havering continued to be relatively stable each year and the total of less than 2,000 registrations across that period was significantly lower than in a ‘Remain’ area like Ealing with almost 19,000 registrations. But in 2014 – following the end of restrictions on their nationals – the number of registrations from Romanian and Bulgarian nationals shot up to over 1,000 per year, with the total number of registrations for these nationals being almost double that of the EU8 across the same period. What also appeared to change in Havering between 2001 and 2014 was unemployment. Beginning towards the end of 2008 and lasting until early 2014, unemployment – and particularly male unemployment – doubled. But, by the time of the referendum campaign in spring 2016 unemployment rates had fallen back to their 2008 levels.

These examples are simply a snapshot. We find a mix of experiences of increasing or decreasing local populations; increased non-EU immigration; fluctuations in local employment opportunities; and a spike in immigration from Bulgaria and Romania. This tends to confirm that, rather than there being an ‘immigration paradox’ – in which cities like London with high levels of immigration were more likely to vote for the UK to stay in the EU – it is a more localised experience of population change combined with other factors, including levels of educational attainmentFootnote ¹⁵ – that may go some way towards explaining voter behaviour.Footnote ¹⁶

The issue of immigration is difficult for mainstream political parties. The political saliency of the issue has increased over time, cutting across both Right and Left and exposing tensions not just between political parties but within them. It clearly divided those Conservatives for whom a cap on numbers was a political red line and those perhaps more inclined to the idea that increased immigration signalled a growing economy with numbers adjusting as labour market forces played out. Meanwhile, the Labour Party had its own troubles with immigration. While out on the General Election campaign trail in Rochdale in April 2010, the then Labour Prime Minister Gordon Brown encountered Gillian Duffy. Following a conversation in which the Labour-supporting woman expressed her concerns about levels of immigration, Brown was recorded as referring to Duffy as a ‘bigoted woman’. A political firestorm ensued. Labour lost the General Election. Some six years later, and interviewed by a British tabloid newspaper, Duffy stated she was voting for the UK to leave the EU.

If the control of immigration was important to referendum politics, then post-referendum, UKIP expected to be a major beneficiary of Labour’s apparent difficulties in connecting with its traditional voter base, including over its position on immigration. However, as two by-elections held on 23 February 2017 demonstrated – one in Stoke-on-Trent (which voted overwhelmingly for Brexit) and one in Copeland – it may be that it is the Conservatives that have squeezed Labour more than UKIP.Footnote ¹⁷

Brexit will have its own effect on immigration but as with the wider economy, it is difficult to predict precisely what the effects will be and when they will manifest themselves. Figures reported in February 2017 showed a fall of 49,000 people in the overall estimated net migration figures, with the numbers of migrants equally split between EU and non-EU nationals. Migration from the EU8 decreased but, consistent with the trend since 2014, the number of migrants from the EU2 was at its highest levels.Footnote ¹⁸ As Portes has noted, employment growth in the UK prior to the referendum was showing signs of easing and is likely to reduce economic migration. Negotiations on the rights of EU nationals currently legally resident in the UK, the legal uncertainty, mixed with emotional disquiet may also lead European nationals to decide to quit the UK.Footnote ¹⁹ In the end, the effects of the referendum on immigration may lie less in the power of governments to exert direct control over migration and more in economic and wider social factors, including the control which migrant EU nationals and their families will take into their own hands in deciding where they wish to live.

Emblazoned across posters, leaflets and big red buses, the Vote Leave campaign stated in stark terms that membership of the EU cost UK taxpayers £350 million per week. It was a figure repeatedly challenged by the Remain campaign and by fact-checking organisations who pointed out that it ignored the automatic deduction of a rebate before any contributions are made to the EU budget. This was, after all, the rebate that Margaret Thatcher had secured in 1984 and which helped create the political conditions for a new impetus to complete the European Single Market.Footnote ¹ It wasn’t even the case that money was ‘sent’ and then a proportion ‘returned’: it was never sent in the first place.

Actual contributions vary between years depending on changes in Gross National Income, but with the application of the rebate, in 2014 the weekly contribution was more like £282 million or £14.7 billion for the year. The £350 million figure also failed to account for the money transferred back to the UK public sector under a range of EU programmes. Once these are taken into account the Office for National Statistics calculated the UK’s net contribution in 2014 at £190 million a week or £9.9 billion for the year.Footnote ²

But whether the figure was £350 million or some other figure probably wasn’t relevant. The point was that voters either couldn’t see or didn’t value what they got for their money, underlying what has sometimes been seen as a ‘transactionalist’ approach to EU membership rather than a more value-driven concept of membership. Voters were, however, clearer that the UK had experienced a post-financial crisis period of austerity and budget cuts. EU contributions were being made as domestic spending was being squeezed. If we are to understand what taking back control over money might mean, answers lie in the different features of EU and domestic public spending and the relationship between the two.

For the EU 2016 budget, annual spending commitments on all EU policies amounted to €155 billion.Footnote ³ By comparison, UK government spending in 2016–17 was expected to be £772 billion, with spending on health alone (£145 billion) amounting to more than the total annual EU budget.Footnote ⁴ These figures tell us something important about the differences between EU and domestic public spending. The main items of domestic expenditure reflect the primary role of the nation state in redistributive policies and the provision of public goods. It is precisely in the areas of welfare, health and education that the EU plays a limited role, principally by ensuring non-discriminatory access to these services for certain beneficiaries under EU law and in creating a framework for co-ordination to ensure continuity of access when people move across borders. The EU itself does not provide these public goods and services.

Instead, EU spending focuses on a set of policy initiatives that reflect the specific tasks and competences of the EU as well as priorities set by EU leaders. To give consistency over time and to set spending ceilings, each annual EU budget is subject to the revenue and expenditure provisions set out in the relevant Multiannual Financial Framework (MFF). Under Article 312(2) of the Treaty on the Functioning of the European Union (TFEU), acting unanimously on a proposal from the Commission, the Council adopts the regulation setting out the MFF, with the consent of the European Parliament. However, national leaders acting through the European Council took overall political control over the 2014–20 MFF,Footnote ⁵ with the then UK Prime Minister David Cameron leading the call for an overall cut in EU spending. The deal finally agreed in February 2013 saw a €34 billion cut compared with the previous MFF period. This was, after all, a time when the EU itself was demanding that Member States – and particularly eurozone states – exercise fiscal discipline and budgetary restraint.

For 2016, the EU annual budget set out principal headings for expenditure: €19 billion allocated to supporting growth, jobs and competitiveness, including €9.5 billion for Horizon 2020, the EU’s strategy to support research and innovation; €50 billion for territorial cohesion (tackling economic disparities between different parts of the EU); and €42 billion supporting farmers under the CAP. The EU budget is, of course, partly spent in the UK.Footnote ⁶ From the EU budget, the UK is allocated funding through its ‘European Structural and Investment Funds’ (ESIF) and the ‘European Agricultural Guarantee Fund’ (EAGF). ESIF is a key instrument of EU cohesion policy and is composed of a number of separate funds. The most significant funds for UK purposes are the ‘European Regional Development Fund’ (ERDF) which supports disadvantaged regions and the ‘European Social Fund’ (ESF) which promotes skills and training to support labour market participation. In terms of the MFF for 2014–20, the UK is allocated €17.2 billion in funding under the ESIF.

Particularly in its first decade of membership, the UK was a major beneficiary of EU funding under the EU’s cohesion policy, with numerous regions and localities qualifying for support. The level of support gradually declined, and with the post-2004 enlargement of the EU – which saw a large number of less economically developed states join the EU – funding to the UK reduced rapidly. Bachtler and Begg’s analysis suggests that the UK’s share of funds declined from 25 per cent in the 1970s to 2.5 per cent by the time of the referendum.Footnote ⁷ The distribution of funds within the territories of the UK is also different with Wales (particularly West Wales and the Valleys), the South-West of England (particularly Cornwall) and the North-West of England being the main recipients of ERDF and ESF funding.Footnote ⁸ Yet despite this asymmetric EU expenditure within the UK, the effect was not that constituencies in those regions receiving funding were more likely to vote for the UK to stay in the EU. Indeed, the opposite was clearly the case with 52.5 per cent of voters in Wales and 56.5 per cent of voters in Cornwall voting for the UK to leave the EU.

The EAGF is the principal source of direct payment to farmers under the Common Agricultural Policy (CAP). The MFF allocation to the UK for 2014–20 is €22.5 billion. The position of the National Farmers Union was that it supported the UK’s continuing membership of the EU, while recognising that there were diverse views among its members. More broadly, the CAP has always been a source of controversy given the direct subsidies that are given to farmers. The Vote Leave campaign claimed that the CAP kept food prices artificially high (it remains to be seen whether the depreciation of sterling following the referendum will itself increase the price of food imported from the EU). Moreover, with large farming conglomerates receiving particularly significant slices of EU funding, the accusation could also be made that EU membership was simply working for the interests of big business.

Once it is recognised that sums of money flow back to the UK from the EU budget, it becomes even more apparent that there never could have been a £350 million a week Brexit bonus for the UK to spend on the NHS. The only way of getting close even to the figure of £282 million that took the rebate into account would be to cut to zero all the expenditure on things like research, regional policy and support for farmers that currently comes back out of the EU budget. Certainly in the short term, that doesn’t seem a likely prospect with the UK Chancellor committing to maintain current levels of direct payments to farmers under the CAP until 2020. Nonetheless, what does seem likely is that reforms to direct payments to farmers will occur, driven not just by Treasury demands but also the devolution of control over agricultural spending to the devolved administrations (and all subject to compliance with World Trade Organization (WTO) requirements on agricultural subsidies). Having control over money also entails having responsibility for the political choice to spend money on the NHS or on farmers. While that control may be important, the trade-offs it creates may not have been apparent to voters.

Moreover, depending on how the UK defined its future relationship with the EU, the UK could continue to make contributions to the EU budget post-Brexit: Norway contributes around £724 million annually.Footnote ⁹ Continuing contributions would reduce any potential Brexit dividend still further. None of which takes into account any of the potential costs should Brexit have a significant effect on the UK economy. There will simply be much less money to control than voters had in mind when they thought about how much the UK paid for its EU membership and what sort of Brexit bonus might be reaped to spend on valued public services.

For the Remain side, one of the principal privileges of EU membership is the UK’s participation in the Single Market and the benefits that it brings in terms of tariff- and barrier-free trade. Free trade is not free as in ‘costless’. But the costs are principally those involved in supporting the negotiation, adoption, amendment, implementation and enforcement of EU rules. The EU is sometimes described as a form of ‘regulatory state’ in recognition of the primary role it plays in market liberalisation and market regulation rather that the provision of core state activities like public spending on health, education and welfare.Footnote ¹⁰ The EU’s overall administrative costs are relatively low: 5.7 per cent of the budget or €8.9 billion was allocated for all its administrative costs.Footnote ¹¹ Assuming a UK contribution of 8.5 per cent of EU budget commitments, the UK would contribute around €14.5 million a week as its share of the administrative costs of the EU: this compares with around £16 million a week for the administrative budget of the UK’s Department for Work and Pensions.Footnote ¹²

In terms, however, of who benefits from the Single Market, the suggestion could be made that the main beneficiaries of this EU regulatory activity are large corporations engaged in cross-border trade and not the thousands of small businesses for whom domestic markets remain important. Of course, economic growth and supply chains mean that small businesses can be beneficiaries too, but perhaps in ways that are less obviously attributable to EU membership. However, when it came to compliance costs (or ‘red tape’), the burden of compliance with EU legislative requirements may have felt greater for small and medium-sized enterprises than larger multinational corporations.

The same sorts of arguments about costs and benefits of EU membership have been made at various points in the UK’s history of EU membership. What sharpened this concern about control over money at this particular moment in time was its linkage to the policies of austerity and anxieties about where the EU was heading in terms of the deepening of integration within the eurozone.

Following the financial and economic crisis, fiscal discipline and cuts in public spending became the key tenets of government policy in the UK and elsewhere in Europe. In the eurozone, the picture was worse with Greece, Cyprus, Portugal, Ireland and Spain receiving financial support through loans and loan guarantees from variously the EU budget, other EU states, the European Stability Mechanism or the International Monetary Fund. Attached to those loans were an extensive range of conditions from a variety of legal and para-legal sources, entailing significant cuts in public expenditure and increases in taxes for the citizens of indebted statesFootnote ¹³ The EU also responded by revising and strengthening existing rules and sanctions on public deficits and debt, and put in place new rules to permit EU institutions to evaluate the national budgets of eurozone state and to check for emerging ‘macroeconomic imbalances’. With the agreement of a ‘fiscal compact’, eurozone states also agreed to enshrine in domestic law balanced budget rules, including making domestic constitutional adjustments.Footnote ¹⁴

Many of the EU’s responses to the crisis simply did not apply to the UK because of the historic stance it has taken on not joining the single currency. Retention of the national currency has been a focal point for different degrees of Euroscepticism and not just the United Kingdom Independence Party (UKIP) variety, although not for nothing does UKIP have a large ‘£’ symbol as the backdrop to its logo. William Hague’s spell as leader of the Conservative Party saw him launch a ‘Keep the Pound’ campaign. At the time, while the then Prime Minister Tony Blair was open to the idea of the UK joining the euro, his then Chancellor Gordon Brown had made entry into the single currency conditional on five economic tests being met, making it difficult for the Prime Minister to make a purely political decision to take the UK into the euro.Footnote ¹⁵ In the context of David Cameron’s renegotiation of the UK’s relationship with the EU, the ‘new settlement’ deal underscored that financial support to secure the stability of the euro could not entail financial contributions from non-eurozone states like the UK.

Nonetheless, the difficulties experienced by eurozone countries contributed to a heightened voter anxiety about the UK’s EU membership. The problems in the eurozone were frequently used to suggest that the euro and the European Union were failing projects and were a manifestation of what happens when nations lose control over their currency. That the EU might respond by deepening integration within the eurozone could constitute evidence of inherent federalist and state-building ambitions while exacerbating a problematic division in the status of EU Member States between eurozone insiders and outsiders. Moreover, the claim that economic ‘experts’ hadn’t predicted the crisis and/or had supported the UK joining the euro became part of a narrative to undermine the forecasts of ‘experts’ that leaving the EU would itself entail significant economic costs.

But beyond the narrow narratives of the referendum campaign, the wider context for concerns about control over money were to be found in the post-crisis politics of austerity. For voters who had felt the effects of a shrinking public sector both in terms of employment and the provision of public services, rejection of EU membership was an opportunity to challenge ideas that the economy was either simply a product of market forces or the object of a type of post-political technocratic managerialism that seemed to characterise the European response to the crisis. That a Conservative government was backing EU membership while pursuing policies of austerity that were, in any event, consistent with the prevailing European preference for fiscal discipline created a focal point for political resistance, not least for traditional Labour voters. Viewed in this way, we can also see more clearly why the idea of diverting resources from the EU to public spending on the NHS might have resonated strongly with voters.

Of course, voters in Scotland had, in 2015, delivered to Westminster, fifty-six Scottish National Party MPs (out of fifty-nine seats in Scotland) on a policy platform to resist and reform national austerity policies. Some 62 per cent of voters in Scotland had also voted for the UK to remain in the EU. It was apparently possible to be anti-austerity and yet pro-EU membership. But this reflected the different character of Scottish politics not least following the independence referendum where continuing membership of the EU was largely accepted by those seeking independence and those wishing to remain part of the UK. Compared with voters in England, when it came to discussing control over money, Scottish voters had different concerns and alternative resources for political expression outside of the EU referendum. The same may not have been true for voters in Wales for whom the benefits of EU funding may simply not have compensated for the effects of economic change and for whom the institutions of devolution (and the EU preferences of the parties in Wales) were less relevant compared to the effects of UK government policies.

If the legacy of austerity gives us clues as to the reasons why the UK voted to leave the EU, then it also holds important lessons for the rest of the EU. The ‘mission legitimacy’ of the EU has traditionally rested upon a post-war project of preserving peace and promoting stable, democratic government and an enlarging union of European states.Footnote ¹⁶ Nonetheless, as de Búrca observes, the raison d’être of the EU has been increasingly questioned, not least following the economic crisis.Footnote ¹⁷ As she puts it: why should EU citizens ‘suffer unemployment, economic hardship and austerity in order to “save” the euro and the European Union?’ The disillusionment felt by voters in the UK about its experience of EU membership is not a uniquely British phenomenon. As one observer has described it, there is now a ‘European union of disenchantment’.Footnote ¹⁸

What will remain to be seen following Brexit, is how this disenchantment plays out in elections in the founding EU states of France, Germany and the Netherlands. But Brexit may also produce its own sense of disillusion within the UK once the relative costs and benefits of non-EU membership become apparent, not least if it turns out that there is little if any new money over which to exercise control.

Making democracy work within nation states is never easy. Making it work in the context of a Union of twenty-eight states is a fundamental challenge. Perhaps one broad area of agreement between Leave and Remain voters is a feeling that many of the decisions which affect daily lives are taken in ways that voters often don’t comprehend. Decision-making in the EU is one context – but one of many – where the exercise of power by institutions appears to be beyond the control of familiar national institutions and mechanisms of democratic accountability.

Arguments about the nature and extent of the EU’s ‘democratic deficit’ have long featured in European studies.Footnote ¹ At an abstract level, it is said that as there is no single European people – or demos – democracy in the EU is simply an impossibility at least in the way it is practiced within nation states.Footnote ² For others, the EU doesn’t actually need much by way of democracy inasmuch as it is conceptualised as a technocratic body fulfilling the mandate laid down by Member States in the treaties.Footnote ³ But if we accept that the exercise of power by the EU is a problem for democracy, potential responses have tended to be cast in institutional terms.

At EU level, the question of democratic control is related to the particular EU inter-institutional style of decision-making and the relationships between key political institutions: the European Council, the European Commission, the Council of Ministers and the European Parliament (EP). This parade of institutions is baffling to most voters and even causes confusion among those with greater understanding of how the EU works. The idea that to understand these institutions is to require a type of specialised knowledge exacerbates the sense of the remoteness of these institutions – and the people who study them – from the everyday preoccupations of citizens. It is hardly surprising that perhaps voters conclude that the better strategy is to seek to restore democratic control to more familiar domestic institutions.

Certainly the referendum debate attempted to increase the level of information about the EU’s decision-making processes. But time and again the focal point of discussion returned to what seemed like the nub of the matter: the complaint that decision were being taken by unelected elites. Concerns about democratic control found a specific target in the form of the European Commission.

In thinking about the role of the European Commission and concerns about democratic control, we can begin to unravel the issues across four dimensions. The first dimension highlights how the European Commission’s powers have been controlled within the EU’s own institutional apparatus. Attention is paid to the influences of the directly elected EP and the Council of Ministers (representing national governments). The second dimension evaluates the plausibility of national parliaments controlling the Commission. The third dimension suggests that perhaps the problem is not so much that the Commission is doing things that the national governments don’t like but rather that national governments have often worked with the Commission in extending the range of executive power and dominance not just beyond the nation state but also beyond the normal cycles of domestic electoral politics. The fourth dimension highlights the implausibility of extending electoral politics to the selection of the European Commission’s President. The willingness of Member States to cede control over this process to the EP demonstrated the kind of gaps in the system of political responsibility which could be exploited to urge voters to take back control over democracy.

The European Commission exercises a range of powers. It is the body that initiates the EU’s legislative process by drafting the proposals which the EP and the Council of Ministers negotiate and adopt. The Commission is also responsible for drawing up more detailed rules that supplement, amend or help in the implementation of the legislation that the EP and Council adopts. The Commission acts on behalf of the EU in conducting international trade negotiations. It monitors the economic and fiscal policies of Member States to ensure their compliance with EU rules that impose budgetary and fiscal discipline, especially on the states that are members of the eurozone. The Commission enforces competition and state aid rules that seek to prevent concentrations of economic power – whether in the hands of big business or in the hands of the state – from distorting competition in the Single Market. More generally, the Commission is tasked with ensuring that Member States abide by their obligations under EU law.

In exercising these powers, the fact that the European Commission is unelected is in many ways unremarkable. The European Commission is an executive, composed of civil servants acting under the direction of a College of Commissioners and the European Commission president. Bodies exercising executive authority are often unelected. But what matters is that executive power is under control.Footnote ⁴ Those controls can be democratic or they can take other forms including judicial oversight. The focus here is more specifically on those exercises of the Commission’s powers that require political oversight.

Democratic control is expressed institutionally in the EU directly through the EP and indirectly through the representative of national governments (domestically elected) in the Council of Ministers and, at leader level, through the European Council. In terms of the exercise of democratic control through the EP, while it may be true that the EP has been directly elected since 1979, these periodic elections have struggled to assume significance in domestic political life. All too often, EP elections are ‘second order’ affairs with voters using the elections not to pass judgment on EU matters but on the state of domestic politics.Footnote ⁵ Rather, it is through the representatives of national governments that we may need to look for expressions of democratic control.

Historically, it was the individual and collective control of national governments that could keep the European Commission in check. In the legislative process, during the 1960s and 1970s, it was the Council of Ministers that had to agree to the Commission’s draft proposals. Voting in the Council was by unanimous consent of all the Member States. In effect, any state had a veto. Even when the treaties anticipated a transition to a system of qualified majority voting, the so-called ‘Luxembourg Compromise’ allowed a national government to claim that an issue was of national interest that could not be overridden by a majority vote. Decisions would be made by consensus. In this way, national governments – democratically elected at national level – had collective and also individual control over the adoption of European laws proposed by the European Commission.

The institutional balance began to change from the 1980s and not just as a result of the growing power of the EP. With successive treaty revisions, the reach of qualified majority voting in the Council was extended meaning that, in the end, a state could be outvoted. The claim was made during the referendum campaign that the UK was the ‘most ignored country in the EU’ when it came to voting in the Council.Footnote ⁶ The source of the claim was research by VoteWatch Europe that showed that while the UK’s voting position was aligned with the consensus 97 per cent of the time, in the situations where the UK opposed a measure – a phenomenon that increased in the years immediately leading up to the referendum – it was outvoted by a majority.Footnote ⁷

Looking beyond the adoption of legislation, the reach of the European Commission’s rule-making powers is perhaps felt more directly by European firms and businesses when the Commission exercises its powers to adopt delegated rules and implementing measures. It is at this level that the banning or authorising of particular goods, foods, supplements and chemicals acquires greater visibility and increased significance. The original system of decision-making required the European Commission to consult committees composed of representatives of national administrations (and also committees of experts). This is known as ‘comitology’. This involvement of the national administrations recognised that for the most part the application and implementation of EU rules on the ground is a matter for the Member States. But it also gave the Member States a capacity to exercise control over the European Commission.Footnote ⁸

The system of comitology expanded from the 1960s and went through a variety of procedural and institutional changes. The creation of new EU agencies took on some of the heavy-lifting in carrying out technical and scientific risk assessments. And in certain areas, the power to grant or restrict EU-wide authorisations was given to agencies – particularly in areas like aviation safety, new medicines, certain types of chemicals – on a mandate direct from the EU’s legislative institutions. But much of this type of administration still rests with the European Commission albeit in a revised system since the 2009 Treaty of Lisbon.Footnote ⁹

The capacity for this type of Commission-initiated rule-making to spark wider social interest can be illustrated in a range of areas, from applications to put genetically modified food onto the European market through to banning certain pesticides because of the risk they pose to bees. It is in the areas where there are diverse national opinions on risks, with strongly held, but competing views, that the Commission has often ended up having to take decisions, precisely because the national governments cannot form a clear view one way or another to either ban or permit something. This is a function of the rule-making system which leaves decisions to the Commission where there is no clear qualified majority vote one way or another. Yet the irony of this is that this is a situation which the European Commission itself laments. It is an area where the national governments are neither exercising control nor taking responsibility. It is no accident that, following the referendum decision in the UK, this is an area where the European Commission wants to see reforms aimed at making national governments take greater responsibility. As the Commission President Jean-Claude Juncker stated in his ‘State of the Union’ speech to the EP on 14 September 2016:Footnote ¹⁰

Under the presidency of Jean-Claude Juncker, the European Commission has also sought to tackle complaints about the exercises of its law-initiating powers by simply legislating less. In its annual work programme for 2015 only fourteen new legislative proposals were identified (this compares with twenty pieces of legislation identified in the 2015 Queen’s Speech setting out the legislative agenda of the UK government). Nonetheless, much of the recent activity to control the Commission’s power has been focused on empowering national parliaments to raise their concerns.

One of David Cameron’s ambitions for his ‘new settlement’ deal was to enhance the power of national parliaments to control draft legislation proposed by the European Commission. Since the Treaty of Lisbon, a ‘subsidiarity’ early warning system has operated to allow the chambers of national parliaments to highlight draft proposals that they consider do not require legislative action at EU level.Footnote ¹¹ It was this system which David Cameron sought to improve by extending the amount of time national parliaments would have to review draft EU proposals and to flag their concerns. From its inception, the number of ‘reasoned opinions’ from national parliaments raising objections to EU measures steadily increased but in most cases in insufficient numbers on any individual proposal to trigger the requirement on the European Commission to reconsider its proposal. Certainly this is an area where some have argued for reforms designed to enhance direct democratic parliamentary control over the European Commission.Footnote ¹²

However, this is also an area where reform is difficult. If national parliaments were to be permitted an effective veto over Commission proposals it would risk returning the EU to the days of decision-making paralysis before qualified majority voting in the Council was introduced. National parliamentary vetoes could delay or derail EU decision-making; a risk that can already be seen when national parliamentary approval is required for the ratification of EU trade agreements. A single national parliamentary veto could undermine the will of a majority of the EU governments and their national parliaments. Simply escalating the power to veto does not manage democracy in a multinational union.

Trying to understand the referendum debate about democratic control by reference to the European Commission as a British political bogeyman may be to miss what is actually important about the exercise of executive power in the EU. A better way to begin to grasp what’s at stake is to suggest that the fundamental problem is not one of conflict but, paradoxically, the success of co-operation. From specifying maximum working weeks, to the proper use of metric weights and measures; from determining applicable rates of VAT to the co-ordination of social security, EU rules impact on a myriad of aspects of national life. If these rules are a source of voter discontent it is not because they were imposed by an unelected European Commission against the wishes of an elected national government, but if anything, because national governments have been all too eager to exercise control collectively through, and in partnership with, the EU. The EU and its executive arms have facilitated national governments in their capacity to exercise executive power when confronted with transnational issues that might have eroded the capacity of any single national government to exert its preferences. Put simply, EU membership has been a means of preserving and expanding executive power across borders.Footnote ¹³

National government’s play two-level games in pushing for more initiatives to be pursued at EU level to escape domestic political constraints and to shift responsibility away from national governments to the European institutions. So the issue of democratic control isn’t simply one of how to control the Commission but perhaps more significantly in how to control what national governments do in, and through, their co-operation with other national governments through international organisations. The appropriate course of action would seem to be for national parliaments to take control over their own national governments and what they do or do not commit to at EU level. In the UK, a significant amount of scrutiny is routinely undertaken by the select committees of both House of Parliament, as well as the devolved parliaments. But here, there are limits to what the EU can or should do. It is not for the EU to tell national parliaments how to scrutinise or control their governments: that would transgress constitutional boundaries. The German Constitutional Court has done much to seek to protect parliamentary oversight by the Bundestag.Footnote ¹⁴ The UK’s European Union Act 2011 also sought to enhance parliamentary controls, but primarily in respect of future EU treaty revisions.

But, the connection back to electoral politics is missing. European politics – as opposed to domestic politics about EU membership – has been almost entirely absent from British electoral politics. Political parties have not contested elections or discussed their policy preferences in terms of what they would, or would not like to see pursued in co-operation with other EU governments. Indeed, it is striking that David Cameron focused all his energy on getting a ‘new settlement’ to safeguard UK interests rather than setting out a vision of how he and his government wanted to pursue British interests through the EU, including under its ‘presidency of the EU’ which was due to take place in 2017 but which was cancelled following the referendum.

Rather than mainstreaming European politics into domestic political debates and electoral politics, ‘Europe’ has been a policy silo focused almost exclusively either on the question of membership of the EU itself, or joining the Single Currency. In this way, the referendum on EU membership was a proxy for all the debates about EU policies and politics that simply never took place, or only took place between an elite of policymakers and academics who were in the know. The British debate over Europe came down to a simple binary choice of whether to stay or go.

For some, the solution to this problem of democratic control lies not just with domestic election cycles but also with the need to inject a degree of contestation into the selection of the European Commission President. The political power of the Commission has varied over time and under different Presidents. Former Commission President Jacques Delors was regarded as a particularly significant leader of the Commission and was famously rubbished by elements of the British press. Yet the experience of the nomination of Jean Claude-Juncker as President did more to show how national governments lost control over the process and how Brussels elites carved up jobs.

The office of Commission President is one traditionally occupied by former prime ministers of EU states, historically, nominated by consensus by European leaders. Together with the national nominees for the positions of Commissioners, the entire College requires the approval of the EP. Following the entry into force of the Lisbon Treaty, a new provision was added to the treaties. It provided that ‘taking into account the elections to the European Parliament’ and after appropriate consultations, the European Council, by qualified majority vote, shall propose a candidate for the Commission presidency to the EP. The new idea was that the Commission President should be politically responsive to voters, albeit indirectly via the medium of the elections to the EP.

It was possible to read the new provisions as only making one fundamental change: the move from a nomination process that gave each state a veto to one where – contrary to the normal style of decision-making in the European Council – a majority vote could be held. Institutionally, there was no reason to view the European Council’s collective control over the nomination as being diminished. Of course, the EP could – as in the past – refuse to accept a nomination and any sensible institution would seek to ensure that a candidate could be accepted (hence the appropriate consultations). But that could not turn a downstream right of rejection unto an upstream power to demand that the European Council nominate a particular candidate. Yet this was precisely what happened when the major political groupings in the Parliament conspired to invent a nomination process known as the Spitzenkandidat or ‘lead candidate’.

Members of the EP sit in different political groupings depending on their national political affiliations. The Group of the European People’s Party (EPP) represents centre-Right parties while the Group of the Progressive Alliance of Socialists and Democrats (S&D) represents centre-Left parties. During the 2014 EP elections, the EPP and S&D each chose a candidate who they expected to see take over the Commission presidency if the result of the election gave their grouping the largest share of the seats in the Parliament. Former Luxembourg Prime Minister Jean-Claude Juncker was the candidate for the EPP and the German politician Martin Schulz was the candidate for S&D. In the 2014 election, parties affiliated to the EPP secured 29.4 per cent of the vote (with a turnout of 42.6 per cent of the electorate), while parties affiliated to the S&D attained 25.4 per cent of the vote. The EPP share of the vote had fallen from 35.7 per cent since 2009, with the share of votes of more Eurosceptic groupings rising.

The EPP insisted that Juncker be the nominee for the presidency. Yet it was perfectly plausible to argue that the one message that came from the election was that voters were increasingly unhappy with the state of EU politics and wanted to see change. Nominating yet another Brussels insider to a top post looked just like business-as-usual. That the ‘defeated’ Martin Schulz would become the President of the European Parliament smacked of a political stitch up between the pro-EU centre Right and centre Left. David Cameron sought to block Juncker’s nomination and counted on Angela Merkel to halt his nomination. Her support proved elusive (her own CDU party is affiliated to the EPP) and despite knowing he was going to lose, Cameron insisted that the matter be put to a vote. He lost. It was a spectacle that reinforced rather than challenged a view of the institutions and its leaders as elite and remote. It certainly did not help David Cameron in his bid to persuade British voters that the UK had influence in the EU.

Although the European Commission has been the focal point for democratic discontents, the desire for control over democracy runs deeper than this. As Curtin makes clear, it is a discontent with the phenomenon of the growth and dispersion of executive power.Footnote ¹⁵ National parliaments have struggled to cope with this phenomenon. The spirit of a new nationalism is meant to be the solution that will ensure that decisions will be taken by governments responsible to parliaments at both UK and devolved levels. Yet it is not clear how easy it will be to put the executive genie back into the parliamentary bottle. Nor is it obvious how the new internationalism and the striking of deals outside of the structures of EU decision-making will be rendered responsive to domestic parliamentary processes.

Anxieties about a loss of control over democracy blend with concerns about a loss of control over national laws. But three specific strands of disquiet can be detected operating at instrumental, constitutional and institutional levels. The instrumental level points to the density and reach of EU legal instruments within national law. The constitutional level refers to the accommodation of the primacy of EU law within a system in which the sovereignty of Parliament is the cardinal constitutional principle. The institutional level draws the European Court of Justice directly to the fore of critical analysis. Indeed, following the 2016 referendum, the political litmus test of taking control over laws increasingly seems to rest on removing the UK from the jurisdiction of the Court of Justice.

During the referendum campaign, one of the striking claims made by Boris Johnson – Vote Leave campaigner and subsequently Foreign Secretary – was that 60 per cent of British laws emanated from the European Union. It was a claim said to be based on research conducted and reported in a House of Commons Library research paper. How much of UK law derived from EU sources depends on the question you ask and as the original 2010 research paper made clear, ‘there is no totally accurate, rational or useful way of calculating the percentage of national laws based on or influenced by the EU’.Footnote ¹ But in a campaign where impressions mattered more than hard realities, Johnson’s assertion amounted to a claim that the UK had lost control over its laws.

In fact much of EU law does not need to be incorporated into domestic law at all: it is ‘directly applicable’. Key provisions of the EU treaties from the rules on free movement, to equal pay for men and women, through to the rules requiring businesses not to engage in anti-competitive activities can all be enforced in national courts without national legislation. When it comes to EU legislation – the more specific rules created by the EU’s legislative institution – only directives require the Member States to transpose the requirements laid down in EU legislation into domestic legal rules. By contrast, ‘regulations’ are also directly applicable.

The consequence of this distinction between those sources of EU law that are directly applicable and those that are not is that while both apply and influence UK law, only directives need to be put into domestic legal form. Which explains why – using figures updated from the original 2010 research – on average only 9.4 per cent of UK statutory instruments implemented EU directives.Footnote ² As chapter 14 demonstrates, one of the ironies of Brexit is that it is likely to produce a legislative explosion as the UK seeks to bring directly applicable law into UK law in order to keep UK law compliant with EU law until it is decided otherwise. The consequence will be a taming of control over laws rather than an immediate taking of control.

Yet, for all the complaints about British businesses getting bound up in Brussels ‘red-tape’, concerns about control over law probably have very little to do with the density of law as such. The effects of a single instrument unrelated to EU law – the European Convention on Human Rights, incorporated into UK law through the Human Rights Act 1998 – are enough to illustrate how anxieties and fears about European interference in domestic legal affairs can be generated. It is the idea of recognising an external source of law as a constraint on what nation states can do that seems to be at the heart of the problem. And in the context of the European Union, it is the ‘primacy’ of that law over domestic law that heightens the existential angst about sovereignty.Footnote ³

The principle of primacy was established by the Court of Justice in the 1960s.Footnote ⁴ Having established that provisions of EU treaties could be directly applicable and so enforced in national law, it was obvious that courts would be faced with conflicts between domestic and EU law. The principle of primacy determines which rule applies when there is a conflict –a rule of national law or one of EU law. The Court of Justice reasoned that it flowed from the nature of the EU legal order that if national rules could take priority over EU rules when they conflicted, then this would prevent the uniform application of EU rules across all Member States and, in the process, make EU rule-making less credible. Therefore, EU law would take primacy in the event of any conflicts and national measures would have to be disapplied to the extent that they conflicted. It is important to note that this primacy principle neither rendered national law invalid – it could be applied and enforced to the extent that there was no conflict with EU law – nor did it empower the Court of Justice to ‘strike down’ national law.

While the principle of primacy made sense from the perspective of the Court of Justice, nonetheless, it would be the national courts that would have to be persuaded to accept the practical implications of the primacy doctrine. This was not without its difficulties, not least in those states with written constitutions, domestic guarantees of fundamental rights and specialised courts with jurisdictions to protect their constitutions or otherwise to control the exercise of domestic public authority. Perhaps most famously, the German Constitutional Court laid out the domestic constitutional conditions under which it will accept the primacy of EU law. Less visibly, in 2016 it was the Danish Supreme Court that found itself unable to disapply a provision of national law alleged to conflict with unwritten general principles of EU law.Footnote ⁵

Despite the principle of the sovereignty of Parliament, it proved relatively easy for EU law to be given primacy and effect in the UK. In its ruling in Factortame, Lord Bridge identified that the principle of the primacy of EU law was well-established in the case law of the Court of Justice by the time that the UK joined the European Economic Community (EEC) in 1973.Footnote ⁶ Any limitation on the sovereignty of Parliament was, in his view, voluntarily undertaken by Parliament in enacting the European Communities Act 1972. If this implied that English courts simply went along with what the Court of Justice had said, for Lord Justice Laws, British courts were simply upholding the will of Parliament.Footnote ⁷ In the ‘Metric Martyrs’ case – concerning the prosecution of a trader in Sunderland for selling goods weighed in Imperial measure only – Lord Justice Laws concluded that Parliament, through the 1972 Act, had established that UK courts were to give priority to EU law over inconsistent domestic enactments. If Parliament used express language to legislate contrary to its EU law obligations, then UK courts would feel obliged to uphold the express will of Parliament. In more recent times in the HS2 case, the Supreme Court suggested that the primacy principle may also reach its limit if it conflicts with other important domestic constitutional principles.Footnote ⁸ If there was doubt about the conceptual foundation upon which the primacy principle was accepted in UK law, then the European Union Act 2011 was meant to put it beyond doubt. Yet the ‘sovereignty’ clause in section 18 does no more than restate that EU law has effect in the UK by virtue of section 2(1) of the 1972 Act: it deals with the direct applicability side of EU law and not the primacy side. The ‘new settlement’ deal between the UK and EU agreed by David Cameron contained a ‘sovereignty’ heading but it did nothing to elaborate on the conditions under which primacy of EU law was either accepted or limited. Rumours of domestic legislation to make clear the continuing sovereignty of Parliament came to nothing. Intriguingly, the UK government’s post-referendum Brexit White Paper stated that the principle of the sovereignty of Parliament had never actually gone away:Footnote ⁹

If correct, the taking of control over laws is largely about changing voter sentiment about how it feels to be governed. That said, the primacy issue will be resolved once and for all by the UK’s withdrawal from the European Union and the repeal of the European Communities Act 1972.

Control over law is also about having control over who interprets the law. The role of the Court of Justice as the authoritative interpreter of EU law was at the fore of criticisms about a loss of control over law. The Court’s interpretation of the primary law of the treaties is crucial because the treaties contains fundamental rules on inter alia the reach of EU free movement rules, citizenship, aspects of equality law and competition law, together with the fundamental rights protected by the EU Charter of Fundamental Rights. Given that primary law can only be changed by national governments through a treaty revision process requiring the unanimous consent of national governments as well as ratification by all Member States, it is clearly extremely difficult for an individual state to try and reverse a Court of Justice ruling with which it disagrees.

The Court’s extension of gender equality protection in respect of pay to include occupational pensions schemes can be cited as an example where the preferences of the Court and national governments diverged. It is also a rare example of a situation where Member States sought to regain control by agreeing the Barber Protocol to the treaties in order to limit the temporal effects of the Court’s judgment in the occupational pension sphere and thereby minimise its financial repercussions.Footnote ¹⁰

The Court of Justice’s interpretation of EU legislation is also significant. As with the interpretation of the primary law of the treaties, there are examples where the Court gives binding interpretations that are out of step with the preferences of governments, employers and sometimes also employees. The example of the Working Time Directive is again a case in point. Clarifications given by the Court on what constitutes working time – to include the time that a person is required to be available for work (even if the person may rest while ‘on call’) – have been criticised for their impact on the organisation of emergency and health services. Legislative acts are not burdened by the same constitutional impediments to their alteration but it may, nonetheless, be difficult to find majority support for legislative change. Indeed, following controversial interpretations of the Posted Workers Directive by the Court of Justice,Footnote ¹¹ the Commission found it extremely difficult to come up with substantive legislative proposals that would meet the very different interests of workers from ‘old’ and ‘new’ Member States. The Member States are also not free to amend EU legislation in a way that would conflict with the primary law of the treaty as interpreted by the Court of Justice.Footnote ¹²

In an earlier time when the sorts of cases coming before the Court concerned prosaic issues such as the imposition of customs duties on the import of urea formaldehyde or the requirement that imports of Scotch whisky be accompanied by a certificate of origin, it was possible for the Court to cloak its institutional role in the guise of terse, abstract legalism. With national courts also willing to supply the Court of Justice with a steady stream of new cases, the domain of legal integration could continue to expand.Footnote ¹³ But in more modern times when the issues before the Court are more controversial, the Court, as an institution, is more socially and politically exposed, and its apparent ‘activism’ scrutinised.Footnote ¹⁴

Yet for all this, it is difficult to pinpoint clear examples of judgments of the Court of Justice which generated the sort of level of political or social controversy that might lead voters to consider that the UK should leave the EU. Indeed, the complaints were more typically about the judgments of the European Court of Human Rights – an organ of the Council of Europe and not of the European Union.

The EU courts were also useful to the UK in two ways. First, a strong compliance machinery increases the functional capacity of the EU to achieve its aims by ensuring that Member States comply with their obligations. Secondly, governments are also litigators before the EU courts. Member States may litigate before the EU courts for a number of reasons but one of those reasons concerns the adoption of EU rules by qualified majority votes. It is one thing for a Member State to accept the primacy of EU rules when those rules require all national governments to agree to them: the power of veto gives each state control over what it will or will not accept. Where primacy doesn’t just bite, but hurts, is when EU rules are adopted by a qualified majority of EU states. With successive treaty revisions – and following successive enlargements of the EU –majority voting has become the norm. To be sure, institutional practice is such that where possible, a consensus is sought without the need for a vote.Footnote ¹⁵ But, there are instances where EU rules are adopted in the face of opposition by a state and this is where a state may wish to elicit the help of the Court of Justice in striking down the offending rule.

An example of the UK government challenging an EU legislative measure before the Court of Justice concerns the controversial Working Time Directive that set down limits on the number of hours that a worker could legally work. In fact, the UK had not voted against the measure but merely abstained. The Commission had also agreed to water down the directive by allowing derogations to allow individual workers to contract out of the time limits. The Court of Justice rejected the UK’s claims that the legislation had an incorrect legal basis in the treaties and was an employment policy measure, rather than one correctly relating to health and safety.Footnote ¹⁶ But it became a totemic example of claimed EU interference in a sphere of social and employment relations that British Conservative governments sought to insulate from EU control.

In the years leading up to the referendum, the UK government became a more frequent litigator before the Court of Justice in its challenge to EU rules on ‘short-selling’, the proposal for a Financial Transaction Tax and caps on bankers’ bonuses.Footnote ¹⁷ The UK lost in the first two cases and abandoned the third. Outside the legislative context, the UK did win an important case before the EU’s General Court challenging a policy of the European Central Bank that would require institutions clearing euro-denominated securities to be based in a Eurozone Member State.Footnote ¹⁸ But these examples illustrate not just how much post-crisis regulatory responses have heightened tensions between the UK and the EU but also how even sceptical governments in practice rely on EU courts to try and regain what they have lost in the political arena.

Nonetheless, it is misleading to evaluate the impact of EU law on the UK legal order simply through examples of situations of conflict or controversy. Day-to-day, national courts have simply got on with the task of interpreting and applying EU law. If the political task of doing EU business is a routine aspect of national administrative life, then enforcing EU law in national courts may also be regarded as unexceptional. Indeed, discontent with EU law may not be a function of conflict but, on the contrary, the willingness of the judiciary to carry out its responsibilities to ensure the effective enforcement of EU law in the UK. Much like complaints about the application of the European Convention on Human Rights by UK courts, it is a sense of the empowerment of unelected judges that seems to drive antipathy towards ‘European’ law and which, in part, motivates Brexit.

British workers and citizens have, of course, been the beneficiaries of a capacity to apply and enforce EU law in national courts and tribunals. But, it is not obvious that this capacity has engendered a wider social affinity between the EU and citizens and, indeed, litigating EU rights may seem like the preserve of the few not the many.

With a future Great Repeal Bill intending to maintain much of EU law within UK law after Brexit, UK courts may well find themselves continuing to apply and interpret rules which have their origins in EU law but without the direct interpretative assistance of the Court of Justice. Yet there will be nothing to prevent UK judges from looking at the published rulings of the Court of Justice as a guide to interpretation. It is one thing to seek to take control over laws by Parliament picking and choosing what laws to keep the same and what to change after ‘Brexit Day’, but another for politicians to seek to control the behaviour of the judiciary.

When it comes to thinking about what control over trade means in the context of Brexit, there is an obvious ambiguity and tension. Put simply, to what extent is control over trade actually about equipping the UK with greater freedom to conduct its own trade policies – with the outcome of Brexit judged in terms of how much new trade it produces – and how much is ‘free trade’ a source of political and social contestFootnote ¹ – and a cause of Brexit – with control oriented towards limiting the UK’s exposure to the economic forces of globalisation?

Ambiguities about trade policy defined the UK’s post-war relationship with Europe. The UK backed the creation of the European Free Trade Association (EFTA) as a body to promote intra-European trade while allowing members to strike their own free trade deals with other states. EFTA was the intergovernmental alternative to the more supranationalist European Economic Community (EEC) when the latter came into being in the later 1950s. As discussed previously, the UK’s departure from EFTA and entry into the EEC followed a realignment of British politics to view closer regional trade integration as being in the UK’s best interests. Yet for Conservatives, EEC membership was something of a second-best compared to the bigger prize of global trade, while for Labour, membership of the EEC risked its capacity to undertake domestic industrial policy objectives as was emphasised by the Labour Party’s 1983 election manifesto when it committed to taking the UK out of the EEC. Both parties, however, shared a commitment to trade relations with Commonwealth countries.

But through the 1980s and into the new millennium, the British political mainstream accepted the EU as a powerful promoter of free trade. The drive to complete the Single Market was not only backed by a Conservative government under Margaret Thatcher, its architects included the British European Commissioner Lord Cockfield,Footnote ² as well as many of the senior Commission officials working in what was the Internal Market Directorate of the European Commission. British European Commissioners – Leon Brittan (Conservative) and Peter Mandelson (Labour) – were instrumental in promoting the EU as a global actor pursuing rules-based international trade both through its support of the World Trade Organization (WTO) and through the EU’s own capacity to enter into trade deals with non-EU states. The centre ground of British politics aligned itself with trade in a Single Market and globally through membership of the EU.

The ‘Single Market’ is more than a free trade area in a number of important ways. First, while free trade areas facilitate trade in goods, the EU Single Market also seeks to remove obstacles to cross-border economic activity by facilitating the free movement of workers, the freedom to provide cross-border services, the freedom to establish businesses and branches across the EU territory and the free movement of capital. Secondly, it tackles not just the obstacles that arise from state policies but through its competition rules it also challenges the trade-distorting activities of companies. Thirdly, if free trade areas are primarily focused on the elimination of tariff (duties) and non-tariff (regulatory requirements) barriers to trade between its members, a customs union goes further in establishing a common external commercial policy, with common tariffs applied to trade between the EU and non-EU states.

The customs union when combined with a free trade area has significant benefits in terms of the ease with which businesses can conduct cross-border economic activity and goods have free circulation. But it also involves constraints. The deeper the level of co-operation between members in removing non-tariff barriers through EU-level harmonisation of regulatory norms and standards, the less autonomy each state has over its capacity to lay down its own rules to regulate the market. And a customs union means that externally, there is also a harmonised trade approach to non-Member States that prevents each Member State from pursuing its own trade policy agenda whether on a bilateral basis or multilaterally through organisations like the WTO.

If the political mainstream was willing to accept the relative balance of benefits and constraints of EU trade, the tensions hardly went away.

For the Left, acceptance of EU membership was tied to the hope that it might buffer the effects of globalisation and help preserve the ‘European Social Model’.Footnote ³ The Lisbon Agenda of economic, employment and social reform of the 2000s was intentionally constructed by social democratic and centre-Left political leaders as a balanced agenda in which growth and competitiveness would not be at the expense of social inclusion.Footnote ⁴ Yet the EU – and especially the Court of Justice – was open to the critique that EU law prioritised economic objectives over social ones, with certain cases decided by the Court becoming high-profile and stylised illustrations of an apparently asymmetric constitution.Footnote ⁵ Even for those on the Left supporting continuing EU membership, it was hard to reconcile decades of criticism of the EU for the paucity of its social dimension with the claims that the EU, nonetheless, was a strong protector of workers’ rights. For the Right, part of the problem was its objection to what the Left wanted, as illustrated by ongoing tensions over the Working Time Directive. But more than these skirmishes, the issue remained whether the costs and limitations of EU membership were worth being in the Single Market and participating in its common external trade policy.

Concerns about the effects of free trade were not simply manifestations of abstract or ideological debates about what EU membership meant. They took specific form in the lead up to the 2016 EU referendum in the decision of Tata Steel to restructure its European operations, with the threatened closure or sale of the Port Talbot steel works in Wales. The situation highlighted two key issues with a direct linkage to the EU. In terms of the operation of the Single Market, Tata Steel’s output supplied the European automotive and construction sectors through plants not just in the UK but also Germany, Netherlands and France. Companies within the Single Market have the freedom to choose where to locate their production, safe in the knowledge that their goods have unrestricted access to the European market. That also means that firms may seek to relocate or consolidate their activities in particular sites. Such a decision has huge economic and social implications for local economies. Indeed, it is easy to forget that a European market has distinctly local qualities.

The second and related issue was the effect on the European steel market of a glut of steel products principally from China, where over-capacity was estimated at 350 million tonnes.Footnote ⁶ The European Union has, at its disposal so-called ‘Trade Defence Instruments’ (TDIs) that can be deployed to protect the European market from cheap imports. These include the application of duties to the importation of goods from outside the EU. Yet, whereas average EU duties on steel bars and cold rolled flat steel products are in the region of 21–25 per cent, they are between 133–265 per cent in the United States.Footnote ⁷ In determining what duties to impose on dumped products, the EU has applied what is known as the ‘lesser duty’ rule. This involves a calculation of two figures – the ‘dumping margin’ and the ‘injury margin’ – and applying the lesser figure. This had the effect of significantly capping the duties which the EU had applied to steel imports. It is an approach not taken by the EU’s trading partners.

Back in 2013, the European Commission proposed modernising its approach to TDIs not least in light of the threats to the EU market from cheap Chinese steel.Footnote ⁸ But the proposal stalled in the Council as EU countries took different stances on the balance between protectionism and ‘free’ trade. In February 2016, the UK was one of seven EU states to send a letter to the European Commission urging support for the EU steel industry. But in the same month, a ministerial statement to Parliament made clear that the UK did not back the Commission’s proposal to remove the ‘lesser duty’ rule, stating that it did not strike the right balance between user and producer interests.Footnote ⁹ In March 2016 and as the crisis in the steel sector intensified, the European Commission again called on the Member States to reach an agreement on its proposal to enhance its capacity to adopt stronger TDIs.Footnote ¹⁰ But the May meeting of ministers in the Trade Council deferred making any decisions, citing the need for further work. It focused instead on preparing the EU–Canada Free Trade Agreement for signature. In these crucial months leading up to the June referendum, the EU could be seen to be further pushing trade liberalisation while failing to protect fair trade, albeit largely because of the blocking position taken by countries like the UK.

Some 56.8 per cent of voters in the Neath Port Talbot constituency voted for the UK to leave the EU.

But the issue of trade was not only being played out in the domestic politics of the referendum campaign. It became a key aspect of the United States presidential campaigns.

The EU and the United States have dominated international trade negotiations and through organisations like the WTO, both parties have exerted their global trade power. Despite this, a trade deal between the two trading blocs has been absent. Negotiations on a EU–US ‘Transatlantic Trade and Investment Partnership’ (TTIP) were officially launched at a G8 summit under the UK presidency in 2013. By the time of the referendum, thirteen rounds of talks had taken place. But not only were negotiations difficult on issues such as access to services, government procurement and diverging approaches to genetically modified food, they were also a focal point for social movements in the United States and the EU protesting against the effects of trade on standards of living, protection of the environment and labour standards.

The TTIP talks were frequently depicted as secret negotiations, but as Cremona notes, TTIP marked a ‘new approach’ to providing greater transparency in trade negotiations.Footnote ¹¹ The dispute-resolution mechanism came under fire as arguments were made that they undermined the powers of government to regulate in the public interest, cost governments money in compensation to investors and lined the pockets of an elite group of highly paid private arbitrators. In the UK, the claim that TTIP would open up the UK’s health service to competition from US health providers featured in the Scottish independence referendum campaign (the suggestion being that the UK government was all too ready to pursue a trade agenda through TTIP without protecting national and Scottish interests).

In the United States, the negotiation of TTIP and the agreement in 2015 of a US-led multilateral trade deal – the ‘Trans Pacific Partnership’ (TPP) – became focal points for diverging views on trade not just between Democratic and Republican parties, but within both parties as they selected their candidates to contest the 2016 presidential election. Historically, it was possible to keep both pro free-trade Republicans and sceptical Democrats on the same page by linking efforts to expand trade with the adoption of measures to manage the domestic consequences of international trade competition. President Kennedy’s Trade Expansion Act of 1962 authorised a new federal programme known as ‘Trade Adjustment Assistance’ (TAA). Over the years, the pursuit of new trade deals by different US administrations has been accompanied by efforts to enhance the TAA programme. Yet its effectiveness as a means of supporting workers has long been a matter of controversy with the Wall Street Journal describing it as ‘everyone’s second-favourite option’. For the Right it is an example of unnecessary federal funding, while for the Left its focus on a subset of workers said to be particularly affected by trade deals limited the scope of this social safety net, while doing little to help communities where industry relocation had led to mass lay-offs.Footnote ¹²

While the US President has constitutional authority to negotiate international treaties, to the extent that these treaties impact on the authority of Congress to regulate commerce – by requiring changes in US law to implement lower tariffs or remove non-tariff barriers – then Congressional approval is required.Footnote ¹³ Through what is known as ‘Trade Promotion Authority’ (TPA), Congress gives – for a fixed period of time – authority to allow a President to negotiate trade deals in the knowledge that implementing legislation will be subject to a straight ‘up-or-down’ vote without amendment.Footnote ¹⁴

TPA and TAA have always come as a package, with TPA finding support from trade-supporting Republicans and Democrats accepting the ‘consolation prize’ of compensatory TAA.Footnote ¹⁵ However, the legislative package that would facilitate fast-track adoption of TPP became a focal point for organised labour and environmental movements in the United States, and also for presidential-nominee hopeful, Senator Bernie Sanders. While Hillary Clinton had expressed her support for trade agreements like the North American Free Trade Agreement (NAFTA) and for TPP, in May 2015, Senator Sanders publicly and vigorously opposed Congress giving Trade Promotion Authority, citing the number of jobs he believed had been lost to free trade. But the Senate backed TPA and its connected TAA. However, when the Trade Bill moved to the House of Representatives – and in a move by Democrats designed to frustrate the adoption of the TPP – they voted down the TAA part of the legislative package. Only when the TPA and TAA parts were separated did both the House and Senate allow for fast-track authority for TPP.

While Senator Sanders’ tactic may have made his Democratic opponent Hillary Clinton shift her stance on trade, the more dramatic change in political positioning was occurring on the Republican side with Donald Trump taking on his own party’s pro-trade policy. Like Sanders, Trump viewed global trade as one of the causes of the relocation of jobs outside of the United States, including to lower-wage countries like Mexico. He pledged that if he became president, he would withdraw the United States from the TPP.

The UK’s EU referendum campaign and the US’s presidential campaigns were, of course, parallel events and one should hesitate before overstating the connection between the two. But in their own ways, each highlighted a collapse of the political centre-ground consensus. In the UK, the consensus on the importance of trade in a Single Market and through EU membership gave way, while in the United States voters elected a Republican President openly challenging political orthodoxy on trade policies not just of the Obama administration but his own party.

In one way, the trajectories of Brexit and the Trump presidency both assert the need to have greater control over trade. In the UK, that control means having freedom to expand the scope and partners of future trade deals. In her speech to the World Economics Forum in Davos in January 2017 – just days after she outlined her principal negotiating objectives for Brexit – the Prime Minister Theresa May spoke of her ambitions for a post-Brexit ‘Global Britain’.Footnote ¹⁶ It was a speech that avowedly claimed the virtues of ‘liberalism, free trade and globalisation’. Yet, the speech also recognised that the ‘taken-for-granted’ benefits of trade were being called into question by both Left and Right. For Prime Minister May, the task of making post-Brexit Britain the ‘most forceful advocate for business, free markets and free trade’ is also about showing that ‘the politics of the mainstream can deliver the change people need’.

In the United States, the Trump trajectory is towards having greater control over, and limiting the sort of trade deals the United States is willing to have and with whom. The inaugural address of President Trump could not have been clearer. The trade policy ambitions of the new administration will conform to the overriding directive of ‘America first’, and the imperative to ‘buy American and hire American’. It is an internationalism controlled, and tamed, by nationalism. Within hours of taking office, the White House website stated that the United States would withdraw from the Trans-Pacific Partnership and seek a renegotiation of the North American Free Trade Agreement, with the threat of withdrawal from the agreement. References to the progress made by the Obama administration on TTIP were simply excised from the website.

In the 1960s, the loss of the option of an encompassing trade deal with the United States pushed the UK closer to its European neighbours and, ultimately to EEC membership. Despite the lure this time of an agreement with the United States to make the process of leaving the EU seem more attractive, it may again prove difficult for the UK to strike a trade deal with a United States administration. This time of Brexit may not be a time of – or for – more trade, and the control which UK voters want is perhaps more of President Trump’s style of limits on trade than Prime Minister May’s new internationalist aspirations for freer markets and freer trade.