The 2016 Vote Leave campaign promised four things voters could measure. £350 million a week for the NHS, on a bus. Take back control of our borders. Take back control of our laws. Strike independent trade deals around the world. Six years after the United Kingdom left the European Union, every one of those promises is in measurable reversal. The data is now settled enough to read.

This is not the same argument as ‘Brexit was a mistake’. That is a tribal question that depends on what one thinks of parliamentary sovereignty, EU institutional legitimacy, or regulatory independence as a principle. The argument here is narrower and more boring. The product as sold and the product as delivered do not match. Whether one welcomes or laments any individual outcome is a separate question. The illusion is the gap.

The £350 million a week

The bus claim was that leaving the EU would free up £350 million a week for the NHS. The figure was challenged at the time by the UK Statistics Authority and several economists, but it travelled. Voters in 2016 understood ‘Brexit means more money for our public services’ as the deal on offer.

The Office for Budget Responsibility, the independent body that scores UK fiscal forecasts, has maintained since March 2020 that Brexit will reduce UK long-run productivity by 4 per cent compared to remaining in the EU. The OBR also estimates that exports and imports will be around 15 per cent lower in the long run. The most recent assessment is dated July 2025 and did not revise these figures. The Trade and Cooperation Agreement signed in late 2020 did not reduce the productivity hit. The OBR judged its content broadly in line with the assumption of a typical free trade agreement. Approximately two-fifths of the productivity impact had already occurred before the TCA came into force, driven by investment uncertainty during the negotiation period.

A 4 per cent permanent productivity hit on UK GDP costs the Treasury revenue every year, in perpetuity. Tariff revenue collected at the UK border has been on average £1.9 billion a year higher since 2021 than the OBR’s March 2020 forecast. That is a cost paid by importers and consumers, not a freed-up surplus. The fiscal direction is not more money for the NHS. It is a smaller economy producing less tax revenue.

Take back control of our borders

The campaign argument on immigration was straightforward. Freedom of movement under EU rules meant the UK could not control its borders. Leaving would let the UK choose who came, and reduce overall numbers. The Conservative government of the day had a stated target of net migration below 100,000 per year.

Net migration in the year ending March 2023 peaked at 944,000. That is roughly nine times the political target and about three times the pre-referendum baseline. The composition of who came also flipped. EU net migration turned negative; in the year ending June 2025, EU citizens left the UK on net by 70,000. Non-EU net migration filled the gap and went well beyond it, at positive 383,000 for the same year. The top sending countries were India, Pakistan, China, and Nigeria, principally through three post-Brexit visa routes the UK government created itself: the Skilled Worker visa, the Health and Care Worker visa, and student visas with associated dependant rights.

Net migration came down sharply in the year to June 2025, falling to 204,000. The cause was a sequence of policy reforms taken from late 2024 onwards. Students were banned from bringing family members. Care workers were banned from bringing family members. The Skilled Worker salary threshold was raised. Those measures were taken under post-Brexit visa rules the UK now controls. They worked, in the narrow sense that they reduced the headline figure to roughly the pre-referendum baseline.

The honest reading: the promise was a reduction. The delivered system was a tripling, followed by a return to baseline via tightened versions of the same post-Brexit routes that produced the tripling in the first place. Whether one welcomes or laments the demographic shift is a separate question. The voters who chose Leave on the immigration argument were sold a smaller system. They got a larger one through different doors, and the recent reduction comes from policy choices that were always available.

Take back control of our laws

Regulatory independence was the clearest theoretical Brexit win. Outside the EU, the UK could in principle write its own rules on data protection, financial services, food standards, employment, environment, and AI. Six years on, the actual divergence is small.

UK GDPR is functionally identical to EU GDPR. The reason is the European Commission’s adequacy decision, which permits personal data to flow freely from the EU to the UK only as long as UK protections are judged equivalent. The decision is renewable, not permanent. Material UK divergence away from EU GDPR risks the adequacy lapsing, which would impose costs on every UK firm processing EU citizen data. The cost is large enough that successive governments have declined to make material changes.

Financial services have followed the same pattern. The promised post-Brexit deregulation has been incremental rather than structural. Equivalence with EU rules governs UK access to EU markets in most regulated activities. The EU AI Act, which began phased enforcement in February 2025, applies extraterritorially. Any UK firm placing AI products on the EU market is bound by it, regardless of what UK domestic AI law eventually says, and the UK has not yet passed its own AI bill.

The May 2025 UK-EU summit, the first formal summit since Brexit, concluded fresh agreements on defence cooperation and on easing trade flows. That direction is regulatory re-convergence by political choice. The current Labour government has framed it as a pragmatic reset. The trade math made it inevitable. Six years of regulatory independence sit on a small handful of consequential divergences and a much longer list of areas where the UK has chosen to track Brussels because the cost of not doing so is higher.

Independent trade deals

The post-Brexit trade pitch had two parts. The UK would secure deals with the major economies the EU had failed to land, particularly the United States and India, and the UK would sign agreements faster and on better terms than the EU could.

The headline number is real. The UK has 40 trade agreements covering 74 countries plus the EU. Most are rollover agreements replicating EU deals the UK already had as a member, repackaged for continuity after exit. The genuinely new agreements of consequence are three: the UK-Australia FTA, the UK-New Zealand FTA, and the UK-Japan Comprehensive Economic Partnership Agreement, which closely mirrored the EU-Japan deal the UK was already covered by. UK accession to the Comprehensive and Progressive Agreement for Trans-Pacific Partnership entered into force on 25 February 2025. The UK-India Comprehensive Economic and Trade Agreement was signed on 24 July 2025 and is still in ratification, not yet in force.

The biggest prize was always a UK-US Free Trade Agreement. It has not been signed. Trump’s first administration declined to deliver it. So did the Biden administration. So has Trump’s second administration so far. Goods exports to the EU have not recovered to their 2019 baseline. The OBR’s March 2024 assessment found that UK trade intensity post-Brexit was broadly on track to be 15 per cent lower than EU membership would have produced, with goods trade underperforming peer G7 economies.

The trade-deal scoreboard, six years on: rollovers replicating what the UK already had, three new deals of moderate scale, one prize unsigned, and a 15 per cent permanent reduction in total trade. The independent-trade-deal argument has not been disproved as a possibility. It has not been delivered as a result.

What the current government is doing about it

The most striking confirmation that the four promises have not landed is in the policy of the government currently in office. The Labour administration elected in July 2024 has pursued a UK-EU reset, producing the May 2025 London summit and its agreements on defence and trade flows. It has tightened the post-Brexit visa routes that produced the migration tripling. It has not announced material divergence from UK GDPR. It has not signalled that it will seek to disapply the EU AI Act for UK firms in EU markets.

None of those moves are rejoining the EU. The current government has been explicit that it is not pursuing single-market or customs-union membership. The reset is narrower than that. The direction is convergence on regulation, alignment on defence, and easing trade frictions where they cost the UK measurable revenue. Whatever the politics of that direction, it amounts to a sitting Labour government partially rolling back the regulatory and migration consequences of the Brexit settlement, on the grounds that the alternative costs too much.

What the illusion actually is

The 2016 vote was a real democratic decision. There were arguments for Brexit that did not reduce to economics: parliamentary sovereignty, control of immigration as a matter of self-determination rather than absolute number, regulatory autonomy as a principle independent of whether it is exercised. None of those are illegible.

The illusion is in the version of Brexit that was actually marketed to voters: more money for the NHS, fewer immigrants, freedom from Brussels, better trade deals than the EU could secure. On the only axis those four promises can be measured by outcome data from the OBR, the ONS, and the UK government’s own publications, none have landed as sold. The fiscal cost is permanent and large. The migration system has produced more migration through different doors. The regulatory autonomy is mostly theoretical and is now actively being unwound. The trade deals are mostly rollovers, with the prize unsigned.

Six years on, the trade math is forcing reversal. That is the cost. Not a verdict on the original vote. A reading of the receipts.