The Big Brexit

• Immediately following the referendum, Brexit contributed to a rise in the cost of living and a fall in business investment. This rise in the cost of living was equivalent to a £870 increase in the cost of living per year for the average household. UK business investment fell by 0.1 per cent a quarter on average in the three years post-referendum, compared to growth of 1.7 per cent a quarter on average growth in the previous three years, and ongoing growth in other G7 countries.
• The UK has now been trading under the terms of the TCA for 18 months, and, although UK imports to the EU have fallen relative to the rest of the world, the share of goods exports to the EU remains at its pre-Brexit levels.
• Our research finds that following the implementation of the TCA, the UK has suffered a broad-based fall in both openness and competitiveness in 2021. Between 2019 and 2021, UK trade openness fell by 8 percentage points, significantly more than in countries with similar trade profiles, such as France which experienced a 2-percentage-point decline.
• We find that UK firms are using ‘coping strategies’ to help ease the adjustment to the new trading relationship with the EU. Our analysis suggests that given the increased regulatory burdens, UK firms may also be choosing to ship directly to a single EU destination and use EU-based distributers, rather than exporting directly to smaller markets. Some sectors are particularly affected, such as the agriculture and forestry where the share of trade directed to these three port countries increased by 6 percentage points.
• This report provides the most detailed assessment to date of the impact of the TCA on trade flows over the longer-term. Our modelling finds that over the next decade, as the UK adjusts to the TCA, the UK will have 1.3 per cent lower productivity, and real wages will fall by £470 per person every year than in the absence of Brexit.
• We find that trade barriers look set to increase by more in agriculture and services (and particularly in more highly-regulated professional services) than in manufacturing. This is bad news for UK exports, as 20 per cent of our services exports to the EU are in the highly-regulated category of finance and insurance, and a further 18 per cent are in other highly-regulated services sectors, including legal and accounting, architecture and engineering, and air transport services.
• Within the manufacturing sector, there is considerable variation in the performance of individual industries, reflecting the differing opportunities available to them to reorient to the domestic market. A few will gain, such as food manufacturing, which is expected to be 5 per cent larger, but others will see significant falls in output, such as the manufacture of basic metals. Indeed, manufacturing sector growth is concentrated in lower productivity sectors – the weighted average productivity of shrinking manufacturing sectors was £47 per hour but it was only £37 per hour for growing sectors.
• Our assessment finds that the North East, one of the poorest regions in the UK, will be one of the hardest hit, and that Brexit will increase its existing (and large) productivity and income gaps.

Leaving the EU represents the largest change in the UK’s relationship with the rest of the world in nearly half a century. It is a profound change in economic governance that will reorient production away from trade with the EU and towards the domestic market, impacting people, places and firms across the UK.

The public discourse, as well as the pre-referendum economic modelling, has focussed on describing the anticipated overall economic effects of Brexit, creating the impression that it will have a discrete, and relatively rapid, one-off impact. But the reality is that the impact will be different from that anticipated in a number of important ways. Crucially, it will take time to fully materialise, and will occur in three distinct phases. First, in the immediate aftermath of the referendum, and in anticipation of permanent impacts, household incomes and business investment were affected. Second, trade itself responded following the implementation of the new Trade and Cooperation Agreement and the new barriers that this introduced. And third, structural changes to the UK economy will take place over the long-term, as capital and labour adjust to the new trading arrangements.

Overall, we find that the long-run impacts will mean significant change for some sectors of our economy –for example, fishing – but the aggregate effect will be to reduce household incomes as a result of a weaker pound, and lower investment and trade. This adjustment will be substantial, but we should not expect it to fundamentally alter the nature of our economy, including the UK’s overall services focus and export specialisation. Understanding the scale and nature of this change, and the extent of progress so far, is crucial for policy makers looking to reset the country’s economic strategy. That is the focus of this report, part of the Economy 2030 Inquiry.

Immediately after the referendum on 23^rd June 2016, sterling depreciated. This brought forward the impact on household incomes of what would otherwise be a slow burn change for the UK economy. A year after the referendum, sterling settled at more than 12 per cent below its previous level, and the higher price of imports led to higher inflation, with the resulting increase in the cost of living equivalent to £870 per year.

Meanwhile, firms faced increased uncertainty about the terms on which they might be able to access the world’s largest integrated marketplace, providing a major headwind to investment. UK business investment fell by 0.1 per cent a quarter on average in the three years post-referendum, compared to growth of 1.7 per cent a quarter on average growth in the previous three years, and ongoing growth in other G7 countries. Firms expect this decline in investment to taper off now that Brexit-related uncertainty has fallen, but the UK’s relative investment position remains weak as we emerge from the pandemic.

The driver of investment underperformance has not been the widely expected fall in foreign direct investment (FDI), an important driver of productivity and innovation. UK FDI inflows as a percentage of EU-27 inflows has remained above pre-Brexit levels since the referendum and, despite a decline in the UK’s share of global FDI, it remains in line with that in Germany and higher than France.

These impacts materialised before any new trade (or migration) barriers were introduced. Indeed, the Office for Budget Responsibility (OBR) estimates that two fifths of the overall expected impact on productivity materialised before the new trading arrangement with the EU, in the form of the Trade and Cooperation Agreement (TCA), was implemented in January 2021.

result, but trade flows themselves have proved more resilient than expected, showing little evidence of responding to Brexit before the TCA was implemented. Some argued that sterling’s depreciation would make UK exports to the EU more competitive, while others believed that the uncertainty following the referendum would depress trade flows. However, the share of UK exports to and imports from the EU changed little in the period immediately following the referendum. But significant change has taken place since the implementation of the TCA.

However, this major change to trading arrangements occurred during the Covid-19 pandemic, meaning that care is needed to disentangle the impact of these two shocks. Covid-19 affected global trade flows through supply chains and global travel disruptions, causing inflated freight costs and transport delays. In 2020, world trade fell by 8.9 per cent, the steepest fall since the financial crisis, and only the fifth time world trade has fallen on an annual basis since World War Two. Global services trade was disproportionately impacted, falling by more than 20 per cent. That was the context when the TCA entered into force a year into the pandemic. For this reason, our focus is largely on the relative performance of UK trade with the EU relative to that with the rest of the world, or on the UK’s trade performance relative to similar economies during this exceptional period, rather than changes in the level of UK trade with the EU.

On the face of it, UK trade data do not show the response to Brexit that we would have expected. There was a near consensus that leaving the EU would disproportionately hit UK-EU trade, causing an enduring decline in the share of imports from and exports to the EU. The UK has now been trading under the terms of the TCA for 18 months and, although UK imports from the EU have fallen relative to the rest of the world, there is no clear evidence of this occurring for UK exports to the EU. Between December 2020 and January 2021 (when the TCA was implemented), the share of UK goods exports to and imports from the EU fell by 15.7 and 8.1 percentage points respectively. Although the share of UK goods imports from the EU has remained somewhat below its pre-Brexit level, the share of goods exports to the EU recovered in the following month, and has remained at the same level since. Services trade shows similar patterns to goods trade: the share of UK services exports to and imports from the EU declined by 0.4 and 2.6 percentage points respectively between 2020 and 2021, but this is in line with the longer-term decline in the share for exports, meaning that, here again, the expected relative decline in EU exports is not clearly observable in the UK data.

On the other hand, although not observable in the UK data, our trading partners’ data does suggest that UK goods exports to the EU have fallen by more than those to the rest of the world. EU trade data shows that the UK’s share of total EU goods imports fell by more than a quarter between 2020 and 2021, more than double the fall of some large non-EU partners, such as the US. However, EU and UK trade data diverged in 2021 due to measurement changes on both sides; as a result, UK exports to the EU are recorded as 10 per cent higher in the UK’s data than EU-recorded imports from the UK. If the share of EU imports coming from the UK is estimated using the UK exports data, this fall would be just 13 per cent, which aligns much closer to the falls seen across non-EU partners.

At first glance, the lack of clear evidence of the expected relative decline in UK exports to the EU seems like good news. However, developments should instead lead us to take seriously the signs that Brexit is impacting UK trade openness and competitiveness more broadly. In particular, the fall in UK trade openness since the introduction of the TCA, measured as trade as a share of GDP, has been larger than that experienced by our peers. Between 2019 and 2021, UK trade openness fell by 8 percentage points, significantly more than in countries with similar trade profiles, such as France, which experienced a 2 percentage point decline. Furthermore, the UK is the only large European country to experience a decline in openness since 2020, with openness falling 1 percentage point; France, for example, saw openness rise by 4 percentage points. These relative declines in UK openness are driven by goods trade:  UK goods exports as a share of GDP fell between 2020 and 2021, despite rising for all EU countries except Ireland. Research using the so-called ‘synthetic control’ method to estimate a counterfactual for UK trade flows corroborates these findings, indicating that goods exports were 15.7 per cent lower in December 2021 than they would have been in the absence of Brexit. UK openness also fell across services trade between 2020 and 2021, but, whereas the UK was almost unique in experiencing a decline in goods exports as a share of GDP, the Netherlands, Belgium and Canada saw similar declines in services trade (around 1.0 percentage point of GDP). This is unsurprising, given the lasting impact of the pandemic on international travel, which is key for trade in services.

Data on trade in international goods points to a broader loss in UK competitiveness across several of its most important markets. The UK lost market share in 2021 across three of its largest non-EU goods import markets: the US, Canada and Japan. Some have attributed this weak performance to unfavourable changes in the composition of global demand. But, although the fall in global demand for transport manufacturing imports has contributed to the fall in the UK share, our analysis shows that the change in the nature of global demand has contributed less than the decline in the UK’s competitiveness: changes to sectoral demand explain £6.7 billion of lost annual exports to these markets, but the remaining £11.7 billion comes from the decline in the UK’s share across products. Evidence of disruption to UK supply chains is also clear – imports have fallen, and the prices of imports have increased – and this could explain our weak export performance across trading partners, and the limited extent to which UK firms were able to exploit opportunities of the global trade recovery in 2021. It is unclear exactly how persistent these changes in non-EU trade will prove, but these are worrying signs that Brexit may have had a broader impact on the UK’s openness and competitiveness than expected.

Trade relationships are long-term and sticky, suggesting that changes to trade will take time: more than 60 per cent of UK exports are via trade relationships that have existed since the 1960s. There are signs that UK firms are using ‘coping strategies’ to help ease the short-term adjustment to the new trading relationship with the EU. For example, the number of UK export relationships (measured as the number of country-product export ‘varieties’) to the EU has fallen by 30 per cent relative to the rest of world following the implementation of the TCA.  Some have attributed this to a squeeze on small and medium-sized companies’ ability to export, but our new analysis suggests that, given the increased regulatory burdens, UK firms may also be choosing to ship directly to a single EU destination and use EU-based distributers, rather than exporting directly to smaller markets. Consistent with this, exporters are increasingly sending trade to countries linked to the largest UK ports (specifically: France, the Netherlands and Belgium, whose share of EU imports from the UK increased by 7.5 percentage points between 2020 and 2021). Some sectors are particularly affected, such as the agriculture and forestry. Here, the number of export relationships from the UK to the EU almost halved relative to non-EU relationships, and the share of trade directed to these three port countries increased by 6 percentage points.

Overall, although we haven’t seen the sort of large relative decline in UK-EU exports that was expected, Brexit appears to already be weighing on both UK openness and competitiveness. But it will take many years for the economy to fully adjust, as firms gradually wind down capital invested in EU exports, the labour market adjusts, and changes to the rate of exit and entry of firms across sectors affects the structure of our economy. Modelling the expected long-run impacts of trading with the EU under the TCA can provide evidence on the scale and distributional impacts of these longer-run structural changes.

This report provides the most detailed assessment to date of the impact of the TCA on trade flows. Many papers assessed the potential impacts of Brexit before its final form was known, but far fewer have assessed the final deal, and even fewer have modelled the granular impacts across regions and sectors of the economy. Our approach to assessing the impacts also innovates on existing modelling approaches, enabling an assessment of regional outcomes and of the adjustment paths to the Brexit shock. Based on a detailed reading of the TCA and the associated literature, the new relationship with the EU implies an increase in trade costs of 10.8 per cent for exports to the EU and 11.0 per cent for imports from the EU, and these rise to 16.2 per cent and 16.6 per cent when we account for the fact that the EU is likely to integrate further in future years.

Trade barriers look set to increase by more in agriculture and services (and particularly in more highly-regulated professional services) than in manufacturing. This is bad news for UK exports, as 20 per cent of our services exports to the EU are in the highly-regulated category of finance and insurance, and a further 18 per cent are in other highly-regulated services sectors, including legal and accounting, architecture and engineering, and air transport services. The combined share is almost twice these sectors’ share of global services exports. So, although the provisions in the TCA are comparable with other agreements that are considered to be deep on services, they secure only a very small share of the services market access that the UK enjoyed as part of the single market, and this is particularly important for the UK’s oversized regulated services exports. As with other comparable agreements, the services provisions in the TCA, tend to simply lock in the liberalisation that already applied to non-EU trading partners, rather than securing equivalence agreements that would replicate the pre-existing market access (an example would have been agreeing equivalence to replace EU passporting rights for financial services). These new barriers to trade with the EU are set to lead to substantial changes in output, with regulated and professional services sectors expected to be hit harder than most other sectors. For example, the ‘other professional, scientific and technical activities’ sector is set to shrink by 13 per cent compared to the counterfactual of remaining in the EU.

Within the manufacturing sector, there is considerable variation in the performance of individual industries, reflecting the differing opportunities available to them to reorient to the domestic market. A few will gain, such as food manufacturing, which is expected to be 5 per cent larger, but others will see significant falls in output, such as the manufacture of basic metals. In the primary sectors, the new trade barriers (considered in isolation) are expected to deliver gains for British agriculture, but fishing is expected to be one of the hardest hit sectors, with output set to be 30 per cent lower. This is because British fishers are reliant on exporting to the EU for their revenues, and now face new barriers to sell to EU consumers. On the other hand, British farmers are set to benefit from less fierce import competition from EU producers, who had been successful in supplying produce to the UK market, and these greater domestic opportunities are expected to outweigh any lost market share overseas. However, the extent to which British farmers can take advantage of these opportunities will also depend on policy choices beyond trade, including on migration.

These shifts will mean significant labour market adjustment for the relatively small numbers of workers in the worst-hit sectors. For example, the 5,000 workers employed in the fishing sector in 2019 and the 75,000 employed in the manufacture of basic metals may face a painful adjustment, with increased job uncertainty and potentially big hits to their livelihoods. Those that do experience involuntary job loss not only face an immediate income hit but are also expected to return to jobs that pay less than the one they left. For example, our previous research showed that, in the period from 1995 to 2020, median real hourly pay growth was 1.1 per cent lower among those who had experienced an involuntary period out of work within the previous year, compared to an average of 2.1 per cent growth among all workers.

Despite these significant impact on some sectors, the new trading relationship with the EU will not drive a large or swift labour market adjustment when we consider the UK economy as a whole. Our modelling suggests less than 0.5 per cent of the workforce, equivalent to an extra 132,000 people, will move from their current region-sector as a result of Brexit. This reflects that the UK’s comparative advantage and the structure of the economy will not be fundamentally transformed by Brexit in the way that some hoped, and others feared. Tradeable professional services, including finance and business services, are set to lose out the most, but our analysis suggests its share of gross output will shrink by just 0.3 percentage points, to 20.2 per cent. Some hoped that Brexit would lead to a revival of manufacturing: the modelling suggests that this will not happen, and that it will decline as share of the economy, but this decline is marginal, at just 0.1 percentage point. This will leave the broad industrial structure of the UK relatively unchanged: the UK will remain a highly services-dominated economy with a smaller manufacturing sector than France.

It is more useful to think of Brexit as driving a fall in openness, rather than a big picture sectoral restructuring. Trade openness, as measured by total trade as a share of GDP, is expected to decline significantly, by 3.6 percentage points as a result of the new barriers erected under the TCA, and a further 3 percentage points relative to a situation in which the UK remained in the EU which itself is continuing to integrate. Although the volume of trade that the UK undertakes will be significantly affected, a radical shift in the export specialisation of the UK is not expected. Goods trade is expected to specialise further in line with existing comparative advantages. We estimate that exports from two of the services sectors in which the UK is most specialised– financial services and other business services – will fall by more than average, but these changes in the specialisation of exports will be small relative to even the slow pace of change in UK comparative advantage over the past decade.

In addition to the overall economic impact, much attention has been given to how Brexit could affect different regions and the extent to which it will help poorer areas to ‘level up’. Our assessment finds that the North East, one of the poorest regions in the UK, will be one of the hardest hit, and that Brexit will increase its existing (and large) productivity and income gaps. Given its importance in driving measures of regional inequality, the focus has often been on how London will fare relative to the rest of the economy. This report finds the outcome for London is uncertain. Although it has large exposure to sectors that are among those most negatively affected by Brexit, what also matters is the degree to which its exports are focused on the EU market. There is significant volatility over time in this, but there is early evidence that London is, in fact, adapting to Brexit faster than other regions: the overall volume of London’s goods exports fell by less than average, but London experienced the largest fall in its share of exports going to the EU, which fell by 10 percentage points between 2019 and 2021. This could indicate that London-based firms, which are, on average, more productive, finding it easier than firms elsewhere to respond to the new trade barriers with the EU by exploiting export opportunities outside the EU. Overall, though, the differences between regional outcomes are relatively small, and are unlikely to drive large reductions in regional inequalities.

Brexit is not, therefore, expected to transform the nature of the UK economy, with only minor impacts likely on its industrial structure, export specialisation and pattern of regional inequalities. Instead, the impact of Brexit is better thought of as a broad-based reduction in workers’ pay and productivity.

A less-open UK will mean a poorer and less productive one by the end of the decade, with real wages expected to fall by 1.8 per cent, a loss of £470 per worker a year, and labour productivity by 1.3 per cent, as a result of the long-run changes to trade under the TCA. This would be equivalent to losing more than a quarter of the last decade’s productivity growth. And it should be noted that this analysis assesses only the direct impacts of the new trading arrangement, and does not account for wider impacts on investment levels or changes to migration policy.

Changes in production to focus on the UK market are a significant part of these productivity falls, and help explain why those suggesting that expanding manufacturing sectors can drive productivity gains are likely to be disappointed. Although some manufacturing sectors are expected to expand, these are dominated by lower-productivity sectors (such as food and wood manufacturing): the average productivity of shrinking manufacturing sectors was £47 per hour, compared with just £37 per hour for growing sectors.

Our analysis finds that, although there is uncertainty over the Brexit impacts that have occurred to date, not least because of the entangled Covid-19 impacts, we should expect the lasting impact of reduced openness to be substantial, and to lead to widespread productivity and real income shocks, much of which has already taken place. Although some sectors will be very significantly affected, changes to the broad sectoral structure of the economy and the UK’s export specialisation are set to be relatively small. And, although these won’t be evenly distributed across the country, Brexit is not expected to materially impact the challenge of ‘levelling up’ less productive parts of the UK.

Assessing the impacts of the changes to the UK-EU trading arrangement is important, not to relitigate the merits of leaving the EU decided by a public referendum six years ago, but to ensure policy makers understand how to think about the impacts that have materialised and are still to come.