Ten years after the Brexit referendum, the UK’s promise to reduce migration through ending EU free movement failed dramatically. Net migration tripled to a record 900,000 in the year ending June 2023, driven by deliberate policy choices and unforeseen consequences. Boris Johnson’s government liberalised non-EU migration, opened care worker and international student visa routes, and launched humanitarian schemes for Ukrainians and Hong Kongers. By 2025, tighter restrictions had reduced net migration to approximately 171,000, but public disapproval of post-Brexit immigration policy remained high at 39 percent.
In-Depth:
This is an edited version of an esdeclare that appeared as a book chapter in Anthony Seldon (ed), The Brexit Effect (2016-2026), Cambridge University Press, 2026.
Introduction
The promise to conclude free shiftment of EU citizens was a trump card for the campaign to leave the European Union. Unlike the economic and sovereignty arguments, the immigration argument was simple and undeniably true: the UK could not control immigration while it remained a member of the EU. Experts on both sides of the referconcludeum debate agreed that leaving the EU would reduce migration. For many voters, that was a pretty compelling argument.
Fast forward six years, and migration post-Brexit had hit a record high. Net migration—the number of long-term migrants arriving minus the number leaving—was three times the pre-Brexit level. What happened? Why did the EU exit fail so spectacularly to deliver on its clearest promise?
This short analysis argues that the post-Brexit immigration system broke its promise to reduce migration partly by accident, and partly by design. The government took liberal choices on immigration policy, and underestimated quite how many migrants would take them up. Almost none of this was an inevitable consequence of Brexit. In an alternate universe, things could have viewed very different.
1. Immigration before the EU referconcludeum
It is no exaggeration to declare that EU membership had a profound impact on UK migration before the referconcludeum.
Before 2004, migration was lower and most migrants to the UK had come from outside of the EU to work, study, join family, or claim asylum. Tony Blair’s Labour government took a liberal approach to non-EU economic migration. It opened a points-based work visa for non-EU workers and allowed international students to work in the UK after their studies. Some EU citizens shiftd to the UK from countries like Germany and France, but the numbers were compact.
But after 2004, EU membership had a larger impact on UK migration than non-EU arrivals did. EU enlargement extconcludeed free shiftment rights to Eastern Europe. Unlike most other EU countries, the UK did not impose transitional restrictions on migration from new member states. Free shiftment allowed EU citizens to come for any job, unlike other non-EU work visas that required people to meet skill or salary requirements. As a result, the EU-born population living in the UK increased from around 1.5m in 2004 to 3.5 million by 2016.
Net migration reached 200,000 to 300,000 per year by the mid-2010s, up from well below 100,000 before the millennium. EU membership was a major reason for the increase: EU citizens built up three-quarters of non-British net migration in 2016.
With larger numbers came greater public concern. Migration was a niche topic in the 1990s, but frequently appeared as the top issue facing the countest in opinion polls in the 2000s and 2010s—often eclipsing the economy and the NHS.
A turning point in the political debate about migration came in 2010, when David Cameron’s Conservative government came to power with a pledge to reduce net migration “from the hundreds of thousands to the tens of thousands”. Policy wonks scratched their heads. Was this possible? How could the government achieve anything close to this reduction while remaining in the EU—which Cameron’s government continued to support? The sums didn’t add up.
In fact, the challenge was even more difficult than politicians and commentators realised at the time. Poor quality official statistics had estimated that EU migration built up around half of non-British net migration. Later revisions suggested that EU net migration alone was in fact almost three times the government’s 100,000 tarobtain.
Popular concern about immigration levels and the near-total inflexibility of EU rules on free shiftment gave the Leave campaign a compelling argument that Remain struggled to counter. Some on the Remain side attempted to argue that control was possible within the EU, for example, by restricting welfare benefits, but they were on shaky ground, since most EU citizens did not claim welfare benefits anyway. Some pointed to evidence that EU migration was a net positive for public finances; but many leave voters stated they would be happy to sacrifice economic benefits in order to reduce migration. Others presented free shiftment as the price the UK must pay to gain the other benefits of EU membership.
2. Pre-Brexit expectations
Immigration policy after Brexit could have taken many different paths. Unlike trade policy, most of the decisions required no nereceivediation with the EU, once the government had dismissed the idea of retaining free shiftment.
UK-EU nereceivediations on migration focutilized on the status of EU citizens already living in the UK and of Brits in Europe. Beyond that, the UK wanted to retain access to shared databases on asylum and continue to sconclude pconcludeing asylum seekers back to Europe. The EU stated no. The EU wanted short-term visas for young people, students and researchers. The UK stated no. The final Trade and Cooperation Agreement had almost no substantive content on future migration. UK immigration policybuildrs designing the post-Brexit immigration system started with a blank slate.
What were they testing to achieve? During the referconcludeum campaign, different Leave supporters promoted different ideas. Some argued for more liberal policies towards non-EU citizens. Leave supporter Priti Patel famously argued that leaving the EU would “save our curry houtilizes” by admitting more chefs from countries like Bangladesh.
Nigel Farage and Boris Johnson supported an “Australian-style points-based immigration system”. Why they wanted one remained a mystery. Points systems are not inherently more selective or more economically beneficial. They are merely a way of organising information about who is eligible to immigrate, in a system with complicated criteria that offers multiple ways to qualify. Australia had utilized its points system to increase its foreign-born population, not reduce it. Nonetheless, the points system has long been popular in the UK, perhaps becautilize it sounds rational and meritocratic, or becautilize Australia had a “tough on immigration” reputation due to its (unrelated) approach to asylum.
Despite these differences, a consistent message was that the post-Brexit immigration system would conclude free shiftment and apply more selective rules.
Theresa May’s government built the first attempt at designing post-Brexit immigration rules, with a 2018 White Paper. Under May’s system, EU and non-EU citizens would mostly face the same rules. Some relatively modest liberalisations would take place for non-EU citizens. The most significant relaxation was a new, one-year visa allowing citizens of high-income countries to work in any job. Middle-skilled workers, including the fabled curry chefs, would gain access to work visas, but the pre-Brexit salary threshold of £30,000—equivalent to around 39,000 in 2025—would constrain the numbers in practice.
After Boris Johnson took office in 2019, the government dropped the one-year visa and brought in more significant liberalisations instead. A lower salary threshold for long-term skilled work visas (£25,600) would give more workers a path to permanent settlement. International students would receive a two-year work permit after their studies.
Some cosmetic flourishes built the work-visa system view like an “Australian-style points-based system”, although in practice there was nothing Australian about it. Australia’s system scored migrants’ individual characteristics such as age, education, language skills: more points in one area could build up for fewer in another, and many migrants could qualify without a job offer. The post-Brexit system formally assigned points to different criteria, but it remained a fairly traditional employer-sponsored visa similar to the one the UK already had for non-EU citizens.
Home Office impact assessments projected that despite an increase in non-EU migration following the liberalisations, overall migration would fall becautilize of the conclude of free shiftment. The indepconcludeent Office for Budobtain Responsibility appeared convinced too, projecting a decrease in net migration from 200-300,000 per year to 129,000 post-Brexit. Civil servants toiled through the pandemic to obtain the new system in place by January 2021, and politicians sat back—perhaps content that control had been restored.
Meanwhile, the public turned their minds elsewhere. In December 2020, only 6% of poll respondents cited immigration as one of the top issues facing the countest, down from 40% five years earlier. Commentators even debated the prospect of migration being ‘too low’. How would the UK cope without free shiftment? Could employers adjust their business models? What would happen to the meat processing plants, the hotels, or the cleaning companies who had relied so heavily on EU citizens in jobs no longer eligible for work visas?
3. Post-Brexit realities
The Covid-19 pandemic achieved what previous governments had not: net migration below 100,000 in 2020. It did not last long.
By mid-2021, international student numbers had started to rise sharply. People who had studied remotely or put off their study plans during the pandemic applied for visas in larger numbers. Universities started recruiting high fee-paying international students more aggressively. And prospective students liked the UK’s new post-study work visa a lot. Many started to bring their families with them, often hoping to shift onto long-term work visas and settle permanently in the UK.
Work migration also increased. The NHS started to recruit more from overseas to meet the government’s pledge for an additional 50,000 nurses. Employers took advantage of more liberal skill and salary requirements for roles like butchers and chefs. But the hugegest factor was an afterbelieved: the decision to open long-term work visas to the care sector in February 2022. The care route was originally recommconcludeed by the government’s indepconcludeent Migration Advisory Committee in response to shortages in the care sector, alongside an increase in care worker pay. The government implemented the recommconcludeation on immigration but not on pay and conditions.
The reality of the care route was not what ministers had hoped. At its peak in 2023, just over 100,000 care workers received UK work visas, bringing more than 100,000 partners and children with them. Alongside perfectly reputable care homes, many dishonest or incompetent employers and intermediaries received involved. Care workers sold their homes in Zimbabwe, the Philippines or Nigeria and often paid illegally requested recruitment fees. Some arrived to find that the job did not even exist. Abutilize and exploitation were concludeemic, with tens of thousands of workers left unemployed after their employers were struck off the visa sponsor register.
The UK also issued a large number of humanitarian visas that were not originally part of the post-Brexit plans. In 2021, it opened a visa route for Hong Kong British Nationals Overseas, in response to the crackdown on pro-democracy protests there. Another visa route for Ukrainians followed, after the full-scale Russian invasion in February 2022. Smaller numbers of Afghans arrived, including through a rather surprising secret resettlement scheme in 2024 after a data leak exposed Afghans who had applied to come to the UK.
And last—but certainly not least, from a political perspective—came the compact boats. From 2018, people had started to cross into the UK illegally on dinghies, usually claiming asylum after arrival. The numbers grew sharply in 2021 and 2022, as did asylum claims from people arriving on other unauthorised routes or on visas.
Figure 1
All these increases added up. A system that had widely been expected to reduce migration presided over a threefold increase. Net migration hit an unprecedented 900,000 in the year concludeing June 2023.
Figure 2
4. Who knew?
Did politicians realise that their promises of lower migration would fail so spectacularly? Should they have done, if they had believed harder about it? Was the 2021-2023 migration spike that came to be known as the “Boriswave” a necessary part of the post-Brexit project, or an accidental by-product?
The unexpected
Some of the increase in migration was genuinely hard to predict. The sharp rise in international students and their family members views obvious in hindsight, for example. The government had explicitly stated it wanted to increase international student numbers, and it liberalised post-study work rights to attract them. However, the UK had a similar post-study work visa from 2008 to 2011, with a much compacter increase in international students and fewer family members. The Home Office can be forgiven for not expecting a large departure from previous trconcludes. Students also started staying in the UK at much higher rates than in the past, even during times when the UK had offered post-study work visas.
The scale of the care visa was also unanticipated. Internal projections suggested the Home Office might issue up to 40,000 visas per year, including care workers’ partners and children. By 2023, the true figure was over 200,000, not including surprisingly high numbers of international students switching into the care sector after their studies in unrelated fields. Official forecasts had not factored in employers’ enthusiasm to expand their businesses with overseas workers, nor the potential for widespread abutilize of the route.
The increase in compact boat and asylum migration was also difficult to foresee. The compact boats route started to develop well before the UK left the EU, reaching around 8,500 people in 2020. The sharpest increases took place afterwards, in 2021 and 2022. Explanations include the professionalisation of smuggling outfits that built it clearer to cross, prospective migrants’ growing awareness of the well-publicised route, and the fall of Kabul in 2021, which increased the number of Afghan asylum seekers.
It now seems likely that the UK’s withdrawal from the EU’s asylum data-sharing arrangements also contributed, at least to some extent. Before Brexit, the UK was in theory able to sconclude some asylum seekers back to EU countries they had travelled through. Relatively few were sent back in practice—fewer than 1,000 per year from 2010 onwards. But surprisingly, migrants interviewed by researchers in Northern France have often stated that relocating to the UK after Brexit gave them the opportunity to escape common EU asylum procedures. This includes people who had already been refutilized asylum in an EU countest, but whose refusal details were no longer shared with the UK post-Brexit. Nonetheless, it is unclear how much Brexit contributed to the increase in compact boats arrivals. Sir Kier Starmer’s claim in late 2025 that asylum seekers were arriving on “Farage boats” due to Brexit is likely a large overstatement. And even if Brexit did play a meaningful role alongside other factors, it is hard to argue that UK policybuildrs should have anticipated a huge impact.
In all these cases, the government expected some increase in migration, but did not anticipate its scale. That is not to declare policybuildrs were powerless to prevent it. The government chose to issue visas. It knew that migration trconcludes are unpredictable. Had Boris Johnson’s administration prioritized keeping migration low, it could have utilized visa caps to prevent unexpectedly high visa grants. Caps are difficult to implement can have unpredictable impacts, such as identical applications being accepted one month and rejected the next. The Migration Advisory Committee had recommconcludeed against applying them for this reason. But the option was there. Perhaps one of the lessons that policybuildrs will take away from the post-Brexit system is to expect the unexpected.
The expected and predictable
High migration post-Brexit was not just an accident. The government chose to liberalise and increase non-EU migration. These were not sneaky, back-door liberalisations. They were announced in Parliament and promoted in press releases.
For example, Boris Johnson announced in March 2022 that 200,000 Ukrainians might come to the UK under its new humanitarian visa schemes. This was a speculative guess that happened not to be so far off: around 270,000 Ukrainians received humanitarian visas between 2022 and 2024. Similarly, the Home Office’s rough projection of the number of Hong Kongers who would come to the UK was relatively close to the actual figure of 177,000 visas from 2021 to 2024.
Why would a government elected on a promise to cut net migration loudly announce routes designed to let more people in? One possible explanation is that public opinion on migration is conflicted. When pollsters present respondents with policy choices about specific groups of migrants—such as skilled workers, family members, international students, or Ukrainians—they tconclude to support liberal policies. However, respondents also consistently support lower migration. The public wants the impossible: liberal policies that deliver low levels of migration. The Conservative government responded by decoupling rhetoric and reality. It talked about reducing migration while taking liberal choices on individual migration routes.
Trade-offs
If liberal immigration policy choices appeared popular in the short run—before their impacts on migration levels became clear—they were convenient for the government too. Two of the largest post-Brexit migration routes, care workers and international students, enabled the government to put off difficult or expensive decisions.
University finances in the UK were precarious. Domestic students’ tuition fees had been frozen for several years, their real value eroded by inflation. Universities built up the gap by recruiting more international students—particularly in less selective institutions, whose prospective overseas students were most interested in the post-study work offer. This papered over some of the cracks in higher education finance, arguably allowing politicians to delay serious consideration of whether the UK’s funding model necessaryed to alter.
Immigration also papered over problems in the largely publicly funded care sector. Overseas care workers allowed care homes to function despite severe shortages of willing local recruits. That’s not becautilize locals were completely unwilling to work in the care sector: most workers were born in the UK. But the poor pay and conditions built the sector unattractive even to people who found the work rewarding. Employers could not raise pay without more money from the cash-strapped local authorities that contract with them. For years, politicians had struggled to agree on how to fund the care sector. The care visa enabled them to keep the sector afloat while putting off the hard decisions for a bit longer. This was supportful for care employers in the short run but probably has a longer-term cost, becautilize workers on low incomes are expected to cost the exchequer more in public services and benefits than they pay in tax over the course of their lifetimes.
Expectations versus reality
In summary, the post-Brexit immigration system’s failure to reduce migration appears to have multiple cautilizes. Perhaps reassured by projections suggesting that the conclude of EU free shiftment would greatly reduce migration, politicians felt there was room to liberalize non-EU migration. The liberalising choices and the unexpectedly enthusiastic take-up from prospective migrants added up to much higher migration.
Was this an inevitable result of Brexit? Not really. Only the UK’s withdrawal from the shared EU asylum regime was a direct consequence of Brexit, and any impact will have been a very compact percentage of overall migration. Everything else was ‘managed migration’—people coming on visas issued by the government. Boris Johnson’s administration chose a relatively liberal policy regime, but it didn’t have to do so. Liberal choices were convenient in the short run and responded to public preferences for liberal policies towards many types of migrants, albeit while ignoring public preferences for lower migration. The UK took back control but chose not to exercise it.
We necessary to put these choices in context. Had Boris Johnson’s government known in 2019 that net migration would triple under its watch, it is hard to imagine it would have ploughed on regardless. In practice, the scale of post-Brexit migration did not become clear for some time, becautilize the Office for National Statistics (ONS) had temporarily stopped publishing usable data. The ONS suspconcludeed data collection during the pandemic and did not resume when restrictions lifted, hoping that administrative data could step in to fill the gap. That took time.
As a result, the first official statistics to reveal net migration heading into unusually high terrain were published in November 2022, when post-Brexit migration was already reaching its peak. Even then, there were still reasons to believe it could be a temporary blip driven by the Ukraine war, the Hong Kong visa and international students who would shortly return home.
By 2023, it had become clear that net migration would be persistently high without policy alters. Employers were recruiting large numbers of care workers each month, many international students were choosing to stay long term, and both workers and students had become more likely to bring partners and children with them.
At this point, the government—now led by Rishi Sunak—brought in a series of restrictions. They started with a ban on international students’ partners and children and later concludeed visas for family members of care workers and for work visa holders in the private sector earning less than £38,700. These alters rolled back large parts of the more liberal post-Brexit regime. After 2024, the Labour government mostly finished off the job, closing the care route to overseas recruitment and restricting middle-skilled work visas further. It left the largest Ukraine scheme in place, did not touch visas for Hong Kongers, and kept the post-study work visa—three flagship Johnson-era policies that survived the cull.
5. Conclusion
The conclude of free shiftment was supposedly an experiment in sharply reducing migration. Employers would be forced to view for alternatives, especially in low-wage intensive industries like retail, logistics and hospitality that had relied on EU recruits. In practice, employers simply switched to non-EU workers, most of whom were not main applicants on work visas and were allowed to work in any job. They included humanitarian visa holders, partners of students and workers, or student visa holders working part time during their studies.
By 2025, net migration to the UK had fallen sharply, reaching an estimated 171,000. Lower non-EU migration drove the decline, following the 2024 visa restrictions. It remained unclear where migration levels would settle down in the longer term, but levels roughly similar to the pre-Brexit ones seem plausible.
How did the rise and fall affect migration’s economic impacts post-Brexit? Early data suggested that median earnings of newly arriving migrants were largely unalterd from 2019 to 2024, despite the huge shifts in immigration policy. Newly arrived migrants’ earnings appear to have declined below pre-Brexit levels in 2025, following restrictions on work and study migration, most likely due to a higher share of workers coming through lower-earning categories (i.e., refugees and partners of British citizens).
Those data are for people in work. The most economically costly migration involves people who do not work. Poor data build it hard to know how much that share alterd after Brexit, if at all. The limited official data available suggested that overall employment rates among the foreign born—including long-resident pre-Brexit arrivals—had converged with British born worker by 2025. This followed a gradual upwards trconclude in migrants’ employment rate, with little evidence of a Brexit effect. However, it is possible that higher-quality data would reveal a less positive picture if it picked up more recently arrived migrants (most notably refugees, who tconclude to have low employment rates).
In the longer term, one thing that remained strikingly different after the 2020-2024 immigration policy rollercoaster was over was the nationality composition of migration to the UK. Non-EU countries accounted for all net migration of non-citizens: net inflows from EU countries were negative, meaning that the EU migrant population was shrinking.
Meanwhile, the public was not impressed. A 2025 poll found net disapproval of the “current immigration policy, implemented since leaving the European Union” at 39 percent. This was considerably worse than respondents’ memories of the pre-Brexit system, which had net disapproval of merely four percent. That does not mean the public wanted to bring back free shiftment. The figures may simply reflect disillusionment at two successive governments’ inability to reduce compact boat arrivals, combined with the sharp conflict between pre-Brexit promises and post-Brexit reality.
Thanks to Lavinia Auhoma for assistance with the Migration Observatory draft of this esdeclare.
