Edited by Jonathan Portes

The Economics of Brexit:

What Have We Learned?
The Economics of Brexit:
What Have We Learned?
CEPR PRESS

Centre for Economic Policy Research

33 Great Sutton Street
London, EC1V 0DX, UK

187 boulevard Saint-Germain

75007, Paris, France

Tel: +44 (0)20 7183 8801

Email: cepr@cepr.org
Web: www.cepr.org

ISBN: 978-1-912179-61-9

Copyright © CEPR Press, 2022.

The Economics of Brexit:
What Have We Learned?

Edited by Jonathan Portes

CENTRE FOR ECONOMIC POLICY RESEARCH (CEPR)

The Centre for Economic Policy Research (CEPR) is a network of over 1,500 research
economists based mostly in European universities. The Centre’s goal is twofold:
to promote world-class research, and to get the policy-relevant results into the hands of
key decision-makers.

CEPR’s guiding principle is ‘Research excellence with policy relevance’.

A registered charity since it was founded in 1983, CEPR is independent of all public and
private interest groups. It takes no institutional stand on economic policy matters and
its core funding comes from its Institutional Members and sales of publications. Because
it draws on such a large network of researchers, its output reflects a broad spectrum of
individual viewpoints as well as perspectives drawn from civil society.

CEPR research may include views on policy, but the Trustees of the Centre do not give
prior review to its publications. The opinions expressed in this report are those of the
authors and not those of CEPR.

Chair of the Board Sir Charlie Bean

Founder and Honorary President Richard Portes
President Beatrice Weder di Mauro
Vice Presidents Maristella Botticini
Ugo Panizza
Philippe Martin
Hélène Rey
Chief Executive Officer Tessa Ogden

UK IN A CHANGING EUROPE (UKICE)

The Economic and Social Research Council funds the UK in a Changing Europe initiative,
which since 2015, has analysed the UK-EU relationship and the Brexit process and its
implications. Based within the Policy Institute at King’s College London, the initiative’s
work is tailored to be timely and easily accessible to non-specialist audiences.
Contents

Foreword vii
Introduction 1
Jonathan Portes
1 A short history of the political economy of Brexit 9
Thiemo Fetzer
2 Brexit Britain in a changing global economy 19
Adam S. Posen and Lucas Rengifo-Keller
3 The impact of Brexit on UK–EU trade 35
Rebecca Freeman, Kalina Manova, Thomas Prayer and Thomas Sampson
4 Post-Brexit imports, supply chains, and the effect on consumer prices 41
Jan David Bakker, Nikhil Datta, Josh De Lyon, Luisa Opitz, and Dilan Yang
5 The price impacts of trade agreements 47
Meredith A. Crowley, Lu Han, and Thomas Prayer
6 Brexit and UK services trade 55
Jun Du and Oleksandr Shepotylo
7 The economics of UK financial services post Brexit 63
Sarah Hall
8 The economics of the UK’s post-Brexit immigration system 71
Jonathan Portes
9 How is the end of free movement affecting the low-wage labour force
in the UK? 81
Madeleine Sumption
Foreword

Some eighteen months since the UK exited the EU’s Single Market and Customs Union,
we are finally beginning to see data indicating what the impact of Brexit has been. This
collection brings together a number of acknowledged experts in the field to consider
what we have learned to date. In keeping with the missions of UK in a Changing Europe,
CEPR and VoxEU, the various contributions that follow present the findings of the latest
research in a clear, concise and accessible manner. They deserve a wide audience.

This eBook was put together on a very tight timescale; the conference at which the
underlying papers were presented took place in London on 27 April 2022. We are grateful
to the authors for preparing their papers so quickly after the conference; to others in the
CEPR and UKICE teams; and especially to Anil Shamdasani for his quick and efficient
work in editing and proofing.

CEPR, which takes no institutional positions on economic policy matters, is delighted to

provide a platform for an exchange of views on this extremely important topic.

Tessa Ogden Anand Menon

Chief Executive Officer, CEPR Director, UK in a Changing Europe

May 2022
Introduction
Jonathan Portes
King’s College London 1

What has Brexit meant for the UK economy? For those of us who live in the UK and

INTRODUCTION | PORTES
follow the political convulsions unleashed by Brexit, it may seem like an eternity
since the referendum. However, in the sense of being able to examine hard data on
economic developments since the implementation of the EU–UK Trade and Cooperation
Agreement (TCA), Brexit has just begun. But while it is of course far too soon to draw
any firm conclusions on the longer-run economic impacts, now seems like an opportune
moment to move from models and forecasts to empirical analysis. So, six years on from
the referendum, and over a year since the implementation of the TCA, this volume brings
together leading academic researchers on trade, immigration and the political economy
of Brexit to present new research and evidence on different aspects of this question.

PRE-REFERENDUM FORECASTS

Prior to the referendum of June 2016, there was a clear consensus among economists about
the likely impacts of leaving the EU. By facilitating trade, capital flows, and migration
within Europe, EU membership had increased UK economic growth and per capita
GDP over the four decades since the UK joined the Common Market in 1973; reversing
this would, correspondingly, reduce future growth. The heads of the UK’s three leading
independent economic research institutes – the Institute for Fiscal Studies, the National
Institute of Economic and Social Research, and the Center for Economic Performance –
summarised the impacts as follows (Chadha et al. 2016):

prolonged uncertainty, reduced access to the Single Market, and reduced

investment from overseas... resulting in lower real wages, a lower value of the
pound, higher borrowing, lower public spending or higher taxes; and in the short
run, higher unemployment

A slightly more nuanced, but complementary, perspective noted that the impacts would
depend crucially on the form of Brexit (Armstrong and Portes 2016):

Economists agree that trade, migration and access to large markets are good
for economies. EU membership has led to a relatively liberal approach to both
and provides full access to the largest single market in the world. If this could be
maintained outside the EU and better trade deals negotiated, then the economic
impact might indeed be neutral or even a slight positive
 One Leave voter dismissed such forecasts with the now famous putdown, “That’s your
bloody GDP, not ours” (Menon 2016). This anecdote is often interpreted as showing that
Leave voters did not see the economy as a priority, focusing more on cultural issues or
more abstract concerns like ‘sovereignty’.
2
But Thiemo Fetzer argues in his chapter that this is an oversimplification. Rather, those
who swung to Leave were those who had not seen the benefits of rising national-level GDP
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

and had suffered most from austerity – in particular, cuts to benefits and public services –
in the years running up to the referendum. Instead of mitigating the economic impact of
structural changes – globalisation, immigration, trade, and wider demographic trends –
government policy, especially after 2010, exacerbated them. The level of GDP does matter
– but, arguably, its distribution matters at least as much.

Developments immediately following the vote did not improve the credibility of economists.
Forecasters – research institutions like NIESR and IFS, as well as the Treasury and the
IMF, and City economists – who predicted economic gloom as a direct result not of Brexit
itself but of the referendum result did so on the basis of the expected impact on financial
markets, business and consumer confidence (HM Treasury 2016). But they were (mostly)
wrong. While the pound did indeed fall much as expected, market interest rates did not
rise, and neither the equity market nor house prices fell. More importantly, after an initial
shock to confidence, businesses and consumers largely shrugged off the result. The labour
market remained strong, and unemployment actually fell slightly.

However, as I noted then (Portes 2017), this reputational damage should arguably have
been confined to those who predicted an immediate negative shock. The failure of
economic forecasters to predict the short-run macroeconomic impacts of the referendum
result – where the key transmission mechanisms were market, business and consumer
confidence and uncertainty – did not necessarily invalidate the long-run predictions of
the damaging effects of restrictions on trade and migration, any more than the uncertain
record of short-term weather forecasts invalidates models of the impact of greenhouse
gas emissions on the climate.

BREXIT MEANS BREXIT

Between the referendum and the election of a majority Conservative government

committed to a ‘hard’ Brexit in late 2019, the economic debate, like the political one,
focused on the nature of the future trading relationship between the UK and the EU,
as well as the related issue of the post-Brexit migration system. Here, the economic
consensus was again clear: the UK’s economic interests (and the best observers were clear
that they were referring simply to economic interests) would be best served by seeking to
remain in the Single Market. As Richard Baldwin put it in an earlier CEPR eBook, “[t]he
alternative that seems most sensible from an economic perspective is the Norway option”,
while noting that this would require the UK government to accept the continuation of
free movement (Baldwin 2016).
Indeed, even the government’s own analysis broadly supported this position; the most
comprehensive modelling exercise, undertaken in late 2018 by the (then new but now
defunct) Department for Exiting the European Union, analysed four potential Brexit
options, concluding that an “EEA-type arrangement” would reduce long-term per capita
GDP by about 1.5%, less than a third of the impact of exiting the Single Market while 3
concluding a comprehensive free trade agreement (DExEU 2018).

INTRODUCTION | PORTES
Fetzer argues that while the former option may have represented the political centre of
gravity, both of Parliament and the country as a whole, the fragmentation of the political
system, combined with the UK’s ‘winner-takes-all’ political system, and the first-past-
the-post electoral system, meant that it was the latter that eventually materialised. The
UK government, first under Theresa May and then Boris Johnson, rejected the EEA
model, making clear “we are not leaving the European Union only to give up control of
immigration” (May 2016).

This meant in turn that the EU never seriously considered what, if any, compromises it
could make on free movement. Instead, it underlined the fact that free movement was an
integral part of Single Market membership, and that outside that a comprehensive free
trade agreement was the closest relationship available. In that sense, as Portes (2022)
argues, the politics of immigration and free movement were the key driver of the shape
of the UK–EU post-Brexit economic and trading relationship, as now enshrined in the
Trade and Cooperation Agreement, agreed in December 2020 and implemented on 1
January 2021.

The TCA, while providing for zero tariffs and quotas on traded goods, contains very few
provisions of any economic significance relating to the mutual recognition of regulatory
standards, regulatory equivalence for services (including financial services), or labour
mobility. Compared to membership of the EU (and of its Single Market and Customs
Union), it therefore implies a major increase in trade barriers and trade costs in goods
and services, as well as new restrictions on migration flows. As Adam Posen puts it in his
chapter, the UK was fare more open to trade and immigration, and attractive to FDI, before
Brexit. Or, as he pithily summarises, “Brexit means that the UK has declared a trade war
on itself”. And with the interregnum between the referendum and the implementation of
the TCA over, we can now look at data, albeit early and noisy, on the results.

TRADE IN GOODS

Two of the chapters focus on trade in goods – the key focus of most analysis before the
referendum, and also the topic on which the most detailed and timely data are available.
Rebecca Freeman, Kalina Manova, Thomas Prayer and Thomas Sampson analyse both
the period between the Brexit referendum and the end of 2020, and the first year of trade
under the TCA. So they address two analytically separable questions: the trade effects
of future, but uncertain, trade cost increases, and of the realised impact of higher trade
costs under the TCA.
 On the first, they find no evidence of a statistically or economically significant decline in
the UK’s trade with the EU relative to the rest of the world prior to the implementation of
the TCA. By contrast, the actual introduction of the TCA caused a major shock to UK–EU
trade, with a sudden and persistent 25% fall in UK imports from the EU, relative to the
4 rest of the world. There is only a smaller and temporary decline in relative UK exports to
the EU, but nevertheless a large and sustained drop in the number of trade relationships
between UK exporters and EU importers. This suggests that the introduction of the TCA
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

caused many small UK firms to stop exporting to the EU, but that larger firms were by
and large able to absorb any extra costs.

Similarly, Jan David Bakker, Nikhil Datta, Josh De Lyon, Luisa Opitz, and Dilan Yang
find that impacts were quite different across sectors and destinations. First, they note
that UK imports from the EU have fallen both in absolute terms since the referendum
and relative to imports from non-EU countries since the TCA’s implementation. This is
not the case for exports, where trade with the EU has followed a similar path to trade
with non-EU countries so far. However, a more detailed analysis reveals substantial
heterogeneity across products in the effect of Brexit on trade flows. In some sectors, there
is evidence that goods have been substituted from EU to non-EU countries, especially
in sectors where imports from the EU are used as inputs to production, such as motor
vehicles. In others, there is a fall in imports from the EU but no obvious increase in
imports from non-EU countries. This could represent either a fall in the overall volume of
consumption of the good or a switch to domestic providers.

Since these changes in import patterns are the result of increased costs resulting from new
barriers to trade – customs checks, increased waiting times, and additional paperwork
– they are likely, at least in part, to be passed on to consumers. Focusing on the food
industry, Bakker et al. show that products more reliant on imports from the EU in 2015
saw larger increases in prices than those less reliant on the EU both immediately after
the 2019 election – when it was confirmed that the UK would leave the Single Market and
Customs Union – and the implementation of the TCA in January 2021.

Using a differences-in-differences approach, they estimate a 6% increase in food prices

due to Brexit over the two years to the end of 2021. While high inflation is now a global
phenomenon, and the recent sharp rise in inflation in the UK is not obviously out of line
with the EU or US, the apparent upward pressure on food prices resulting from Brexit
is certainly a far cry from the claims by some proponents of Brexit that leaving the EU’s
Common Agricultural Policy would result in sharp falls in food prices (Sampson et al.
2018).

The chapter by Meredith Crowley, Lu Han and Thomas Prayer explores the implications
of preferential trade arrangements – and, in the case of Brexit, their removal – for market
power, pricing, and consumer welfare. They show that in imperfectly competitive markets,
tariff reduction impacts prices not just directly but also by encouraging the entry of new
exporters via competitive pressures. Corresponding, increased tariffs (or other trade
barriers) is likely to have the reverse effect; reducing the number of exporting firms and
raising markups and prices. While these effects might be expected to take some time to
materialise, and the data used by both Freeman et al. and Bakker et al. reflect only the
first year of the post-Brexit trading arrangements, their results are certainly consistent
with the predictions of Crowley et al. 5

INTRODUCTION | PORTES
TRADE IN SERVICES

Data on services trade are slower to arrive and less granular than those for goods; and
post-pandemic disruption continues to affect travel, tourism and some other ‘in-person’
services delivery. Nevertheless, Jun Du and Oleksandr Shepotylo show, using a synthetic
differences-in-differences approach, that, in contrast to trade in goods, the period
between the referendum and the implementation of Brexit did see a significant Brexit-
induced fall in UK service exports, amounting to about 6% in 2019. There is no evidence
to suggest that UK businesses have redirected exports from the EU markets to those
outside the EU. The apparent beneficiary is Ireland, which has experienced rapid growth
in its service exports over the same period. As Du and Shepotylo note, while worrying
enough, their analysis only covers the anticipation and deterrence effect up to the end of
2019; it will take some time for the full impact of Brexit on UK services to emerge

Focusing on financial services, Sarah Hall also finds some evidence of a decline in UK
exports to the EU, and of the relocation of some financial service activity to locations
elsewhere in the EU, although no one centre has gained disproportionately. And while
perhaps 10% of total assets of the UK banking system have moved, a far smaller proportion
of jobs or value added has been lost.

Looking forward, the key issue is the extent to which the UK regulatory regime diverges
from that in the EU, and the likely consequences. While some divergence is likely – for
example, in insurance – there is little appetite in London for a ‘race to the bottom’; instead,
gradual and piecemeal divergence is more likely. Over the medium term, the implication
is that London will retain its prominence as Europe’s leading financial centre for the
foreseeable future, but this dominance will be gradually eroded over time.

MIGRATION AND THE LABOUR MARKET

If the politics of immigration and free movement were the key drivers of both the Brexit
vote and the decision to leave the Single Market, what of the economic impacts of the
new post-Brexit immigration policy? As with trade, the consensus of economists was
that ending free movement and introducing a more restrictive immigration policy overall
would reduce growth, certainly in GDP terms, and likely in terms of GDP per capita,
as well as having a negative impact on the public finances. This was the conclusion of
 the government’s own modelling described earlier (DExEU 2018), as well as independent
analyses (notably Forte and Portes 2017), which correctly forecast a sharp decline in EU
migration to the UK even before free movement legally ended.

In my chapter, I describe the new system, which does indeed represent a very significant
6
tightening of controls on EU migration compared to free movement. Migrants coming to
work in lower-skilled and lower-paid occupations are, in principle, no longer able to gain
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

entry. However, compared to the current system – and in contrast to earlier predictions
– the new proposals represent a considerable liberalisation for non-EU migrants, with
lower salary and skill thresholds and no overall cap on numbers. This implies that about
half of all full-time jobs in the UK labour market could in principle qualify an applicant
for a visa. This represents a very substantial increase – perhaps a doubling compared
to the previous system – and also makes the new system considerably more liberal with
respect to non-European migrants than that of most EU member states, which typically
apply much more restrictive (de facto and/or de jure) skill or salary thresholds, and often
enforce a resident labour market test. The provisions for international students after
completing their studies are also relatively liberal.

So, the new system does not represent an unequivocal tightening of immigration
controls; rather, it rebalances the system from one which was essentially laissez-faire for
Europeans while quite restrictionist for non-Europeans, to a uniform system that, on
paper at least, has relatively simple and transparent criteria. And this analysis appears to
be born out in data on the operation of the system in its first year, where there has been
a significant rise in work visas issues compared to pre-pandemic levels, particularly in
the health sector, and an even larger rise in the number of international student visas. I
conclude that there is some cause for optimism about the economic impacts of the new
post-Brexit UK immigration system.

Madeleine Sumption concurs with my analysis of the overall impact of the new system,
and focuses in particular on the impact of this ‘rebalancing’ on low-wage sectors that were
previously very dependent on EU workers who could move to the UK under free movement
rules. She finds that many employers in industries that previously relied heavily on EU
migration are finding it difficult to recruit, although Brexit and the end of free movement
are not the only factor at play here, with broadly similar patterns also observed in other
high-income countries.

She also notes that, in theory, there are a number of ways employers could respond to
these shortages:

• attracting more workers from the resident, for example by raising wages, improving
working conditions or being more flexible on hours and contracts;

• recruiting workers from abroad, usually by sponsoring them on a work visa;

• reducing the need for workers, for example through automation, switching to less
labour-intensive goods and services, or cutting back production in the UK.
While some employers have turned to overseas recruiting under the new immigration
system, this has not been the norm in lower-wage industries, likely as a result of the new
system’s eligibility criteria and costs (although the extension of the Health and Care visa to
low-paid social workers may make this sector an exception). Perhaps surprisingly, despite
anecdotal evidence, there is very little evidence so far that shortages are translating into 7
sustained higher wages in the affected sectors. Indeed, Sumption notes that the hospitality
industry, which saw the largest percentage declines in EU citizen employment from June

INTRODUCTION | PORTES
2019 to June 2021, was also one of the industries with the lowest wage growth during that
period. The adjustment process has some way to run.

CONCLUSIONS AND THE WAY FORWARD

The analyses in this eBook are very much a preliminary and incomplete account of the
economic impacts of Brexit. In some cases, they raise as many questions as they answer.
For example, why have UK imports of EU goods fallen so sharply, while UK exports are
much less affected, when (in contrast to the EU) the UK has not yet introduced the full
panoply of import controls provided for under the TCA? Why has the large fall in the
number of EU workers in some sectors – and a corresponding rise in vacancies – not
translated into higher wages, at least in relative terms? Nevertheless, the overwhelming
weight of the evidence presented suggests that – very much as economists predicted
– Brexit has made the UK a less open economy, reduced UK trade in both goods and
services, and increased prices for some products. Moreover, despite public scepticism
of economists and their forecasts, our verdict is increasingly shared by the wider public
(Surridge 2022).

However, as Fetzer points out, aggregate impacts are not the whole story by any means.
His analysis suggests not only that the costs of Brexit are very unevenly distributed, but
that, perhaps paradoxically, those areas that voted most heavily for Brexit are the worst
affected, while London has escaped largely unscathed, at least so far. More broadly, while
austerity has been moderated, it has by no means gone away, with the current ‘cost of
living crisis’ likely not only to hit the poorest hardest, but to be exacerbated by inflation-
driven cuts in benefits and public services – “austerity by stealth”, as Portes (2022b) puts it.

Both Posen and Fetzer propose distinct, but complementary, strategies to address both
the economic and political implications of Brexit. Posen outlines four key points. First,
Posen argues for the UK to abandon our post-imperial delusions and to follow, rather
than try and inevitably fail to ‘lead’ globally in an economic sense. Second, and building
on the relative liberalism of the post-Brexit migration regime, we should focus on
attracting skilled labour and productive capital, where we still have potential advantages,
rather than obsessing about trade deals that will make little economic difference. Third,
we should stop trying to ‘defy gravity’, and re-engage with the EU, based on regulatory
convergence. Finally, we should seek to rebuild our position as a bridge between Europe
and the US, albeit now from outside the EU. Meanwhile, Fetzer suggests both much
 greater devolution and greater redistribution – geographically and between generations,
as well as from rich to poor. The alternative, both suggest, is a poorer, more insular and
more divided country.

8
REFERENCES

Armstrong A and J Portes (2016), “The Economic Consequences of Leaving the EU”,
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

National Institute Economic Review 236(1): 2-6.

Baldwin, R (ed.) (2016), Brexit Beckons: thinking ahead from leading economists, CEPR
Press.

Chadha, J, P Johnson and J Van Reenen (2016), “Leaving the EU would almost certainly
damage our economic prospects”, NIESR blog, 20 June.

DExEU – Department for Exiting the European Union (2018), “EU Exit: Long term
economic analysis”, November.

Forte, G and J Portes (2017), “The economic impact of Brexit-induced reductions in

migration”, Oxford Review of Economic Policy 33(suppl_1): S31–S44.

HM Treasury (2016), “The immediate economic impact of leaving the EU”, 23 May.

May, T (2016), “The new centre ground”, speech at the Conservative Party Conference,
2 October.

Menon, A (2016), “2016: A Review”, UK in a Changing Europe, 31 December.

Portes, J (2017), “The Weather is not the climate”, UK in a Changing Europe, 19 January.

Sampson, T (2016), “How economists for Brexit manage to defy the laws of gravity”,
VoxEU.org, 2 June.

Surridge, P (2022), “The evidence shows both Remainers and Leavers think Brexit has
increased the cost of living”, UK in Changing Europe, 5 May.

ABOUT THE AUTHOR

Jonathan Portes is Professor of Economics and Public Policy at King’s College London and
a Senior Fellow of the ESRC UK in a Changing Europe programme. Previously, he was
Director of the National Institute of Economic and Social Research; and prior to that
Chief Economist at the UK Cabinet Office, where he advised the Cabinet Secretary and 10
Downing Street on economic and financial issues. Before that he held a number of other
senior economic policy posts in the UK government. His particular interests include
immigration, labour markets, and poverty.
CHAPTER 1
A short history of the political economy 9

of Brexit

A SHORT HISTORY OF THE POLITICAL ECONOMY OF BREXIT | FETZER

Thiemo Fetzer
University of Warwick

The UK’s 2016 EU referendum was a watershed moment in European history. There has
been a lot of focus on trying to make sense of the cross-sectional variation in support for
‘Leave’ in the referendum across people and places. There has been much less focus on
the broader economic and political context that enabled an EU referendum to be held in
the first place, or on the political economic implications of the ‘hard Brexit’ that became
a political reality. This chapter aims to summarise work carried out across several papers
and multiple years providing a nuanced insight into the underlying political economy
forces at play and speculates about the future direction and challenges.

BREXIT HIGHLIGHTED REGIONAL AND SOCIOECONOMIC CLEAVAGES

Statistically speaking, it is not hard to ‘explain’ the dominant factors behind the different
levels of support for Leave. Looking at the 2016 EU referendum vote and its geographic
distribution or its distribution across individuals, we observe that people and places with
relatively poor economic fundamentals, often referred to as the places and people that
have been ‘left behind’, came out much more strongly in favour of Leave in 2016 (Alabrese
et al. 2019, Becker et al. 2017).

Support for Leave, however, was not made up of a cohesive social or demographic group,
as is widely suggested. Most of the political science research and popular discourse has
focused on characterising the average Leave supporter as a white English male, with
relatively low educational attainment, nearing or in retirement. This average Leave
supporter was indeed dominant, easily accounting for two out three Leave votes. Yet,
the crucial voters that swung the referendum (and likely played a major role in bringing
about a referendum in the first place) were a mixed bag of protest voters that make up the
remaining third.

This polity of marginal Leave voters is crucial to understanding how the referendum
became a political reality in the first place and why Leave eventually won. EU scepticism
was always more widespread among the average Leave supporter, not just in 2016 but
also over previous decades. It thus is vital to understand who, over time, became a Leave
supporter and why they did so. Unfortunately, we do not have good historical data that
enable an exploration of this question directly. The next best window through which we
 can study the build-up of Leave sentiment – which, in turn, may help in understanding
the political realities that enabled an EU referendum to be called in the first place –
is using electoral data across local, European and general elections. In Fetzer (2019), I
document that the characteristics of Brexit began to appear in the drastic increase in
10 support for the UK Independence Party (UKIP), but only from 2010.
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

FIGURE 1 SUPPORT FOR UKIP ACROSS EUROPEAN, LOCAL AND GENERAL ELECTIONS
OVER TIME

30

20

10

2000 2005 2010 2015

European Local Westminster

The emergence of UKIP was consequential in terms of driving the UK’s electoral outcomes
in a direction that made an EU referendum a political possibility. The drastic increase in
UKIP support across all elections post 2010 was a prime driver of this.

The 2015 general election paved the path. It was predicted to result in another hung
parliament and a continuation of the Conservative/Liberal Democrat coalition
government, but this did not come to be. This was crucial given that, in 2013, David
Cameron had promised the EU-sceptic wing of his party a referendum on EU membership
should he win a majority in the 2015 election. This should have helped the Conservatives
to attract anti-EU voters, but instead Conservative support flatlined while support for
UKIP surged, with the party coming in third (with 12.7% of the vote) but winning only
a single seat in Parliament. The UKIP surge deprived Labour and the Liberal Democrat
candidates of crucial votes, while not hurting the Conservatives as much.
This voter movement helped the Conservatives to win seats in constituencies where
elections are typically fought out either between Labour and the Conservatives or between
the Liberal Democrats and the Conservatives. As a result, the 2015 general election saw
David Cameron gaining 24 seats with just a 0.7% swing, while Labour and the Liberal
Democrats lost a combined 79 seats. The prime minister ended up having to call an EU 11
referendum to honour his promise.

A SHORT HISTORY OF THE POLITICAL ECONOMY OF BREXIT | FETZER

It is therefore imperative to understand what caused this dramatic increase in electoral
support for UKIP between 2010 and 2015. And again, the answer must be found in
domestic politics.

AUSTERITY AND THE POLITICAL BACKLASH AGAINST IT ENABLED AN EU

REFERENDUM TO BE CALLED

Domestic politics were notably shaped by the UK coalition government’s drastic cuts to
public spending in the wake of the global financial crisis. There was a simple recurring
theme to these spending cuts: the more deprived an area, the more exposed it was to the
cuts (Beatty and Fothergill 2013, Innes and Tetlow 2015). The very same places that had
been falling behind economically, which are home to a lot of ‘left-behind’ voters, were
much more exposed to and affected by austerity. To a significant extent, austerity – even
within regions – exacerbated existing cleavages, with public spending becoming much
more age-biased. Figure 2 illustrates this by comparing real public spending per capita
on pensions, education, welfare and social protection. It is quite apparent that since 2010,
while expenditure on (state) pensions has continuously increased, spending on education
(i.e. benefiting the future generations) as well as spending on the current working age
adult population has drastically declined.

The sharp changes in public spending were not limited to central government spending.
At the local level, the adjustments were even more pronounced. Real spending per person
by local authorities decreased by around 25% between 2010 to 2015. Yet again, this
aggregate figure masks notable compositional changes. Figure 3 documents this for a few
classes of spending, with values indexed to 100% in 2007. Except for social care spending
– which, again, mostly benefits the current elderly population – spending on planning and
development, housing and culture and many others saw drastic cuts ranging from 20%
to 40% relative to 2007.
 FIGURE 2 EVOLUTION OF REAL SPENDING PER CAPITA ON PENSIONS, EDUCATION AND
WELFARE AND SOCIAL PROTECTION (£)

2500
12 Welfare & Protection
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

2000
Pension

1500 Education

1000
2000 2002 2004 2006 2008 2010 2012 2014

Source: Fetzer (2019).

FIGURE 3 LOCAL GOVERNMENT SPENDING INDEXED TO 2007, BY CATEGORY

index 2007=100

140

Social care
Housing
120

100 Culture

80

Planning & Development

60
2007 2009 2011 2013 2015

Source: Fetzer (2020).

These cuts have left their mark on politics, notably shaping contests across local,
European and general elections since. In Fetzer (2019), I document that a notable part
of the increase in electoral support for UKIP since 2010 can be attributed to austerity-
induced protest voting.
In fact, austerity-induced protest voting directly shaped the hugely consequential 2015
general election. Austerity hurt the Liberal Democrats at the ballot box much more than
it hurt the Conservatives. As the junior partner in the 2010 coalition government, in
essence the Liberal Democrats implicitly underwrote most of the public spending cuts
that appalled their respective voters. As a result, the shift of voters towards UKIP was 13
much more pronounced in places that were more exposed to austerity and also much
larger among voters that had supported the Liberal Democrats in 2010.

A SHORT HISTORY OF THE POLITICAL ECONOMY OF BREXIT | FETZER

As discussed, this was significant in constituencies that were traditionally contested
by the Liberal Democrats and the Conservatives. Out of the 49 seats that the Liberal
Democrats lost, 27 seats were lost to their Conservative coalition partners, enabling the
Conservatives to win an outright majority in the general election.

AUSTERITY BROKE THE IMPLICIT SOCIAL CONTRACT, INDUCING A PROTEST

VOTE

Austerity and the resulting cuts to social welfare spending brought to the surface a
broad set of existing economic grievances that had been festering for decades. The
recognition that structural change is generally desirable for economies and societies
is a key distinguishing feature of Western-style liberal market economic orders. This
is underpinned by an implicit social contract in most Western countries based on the
understanding that the welfare state will provide social insurance in the wake of large
structural changes that lead to adjustments in the economy.

While the broad evidence nowadays suggests that immigration is unlikely to have large
redistributional consequences (Becker and Fetzer 2018, Portes, 2019), there are many
other economic changes that can aggravate existing economic and social cleavages –
between the young and the old, between urban and rural areas, and between the high-
and low-skilled, among many others. The last decades have seen notable examples of such
changes, including structural transformation induced by globalisation (Autor et al. 2013,
Colantone and Stanig 2018, Dippel et al. 2015), broader skill-biased technological change
(Graetz and Michaels 2015, Rogerson 2008), the rise of the gig economy, growing market
power of firms and the proliferation of insecure employment practices, along with the
dislocations in housing markets resulting in protracted housing insecurity for large social
groups (Fetzer et al. 2019).

The retreat of the welfare state, which had helped tackle some of the dislocations caused
by these structurally induced economic changes, was akin to ripping the plaster from a
wound. It is not surprising that the resulting grievances eventually found their voice in
electoral protest votes. These protest votes were easily mobilised by a political campaign
that promised a change to a status quo that had ceased to work for many people. It is this
constituency that gave rise to the marginal Leave voter and the marginal UKIP supporter,
and which enabled the referendum to be held in the first place.
 FROM THE EU REFERENDUM VOTE TO A HARD BREXIT

The referendum was a direct result of the UK’s austerity-induced political dislocations,
which have been broadly aided by an electoral system that features a winner-takes-
14 all mentality and a political culture in which referenda are not widely used. Following
the EU referendum, a plurality of voters settled into a position that the result should
be implemented, despite many marginal Leave voters soon coming to regret their vote.
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

Interestingly, survey data suggest that this regret is most pronounced among those who
voted for Brexit but did not expect Leave to actually win the referendum (see Figure 4).

FIGURE 4 SHARE OF BRITISH ELECTION STUDY RESPONDENTS STATING THEY REGRET

HAVING VOTED TO LEAVE AS A FUNCTION OF THEIR EX-ANTE EXPECTATION
OF THE PROBABILITY THAT LEAVE WOULD WIN THE EU REFERENDUM

Note: Each data point refers to one decile of Leave voters. The value label on the horizontal axis is the expected probability
of a Leave victory in this decile of Leave voters.

This plurality of voters found itself sitting between the hardline ‘Brexiteers’ on one side,
who wanted to implement the hardest form of Brexit, and a group of staunch Remain
supporters who wanted to maintain the status quo of the UK as a member of the EU, or
at least supported another referendum on a well-defined plan for Brexit. Politically, the
momentum was with the Brexiteers.

As the 2015 election had provided the Conservatives with only a narrow majority in
Parliament, Prime Minister Theresa May called another general election in 2017. Her
intention was to ensure a stable majority to be able to see through Brexit in the years
to follow. May hoped the Conservative Party would make gains in the wake of the EU
referendum as die-hard Leave supporters who had supported UKIP in 2015 (not to be
confused with the marginal protest voters) returned to the Conservative Party fold.
Once again, however, protest voters complicated matters. Many of the UKIP protest
voters switched to supporting Labour under its new left-wing leader Jeremy Corbyn,
who represented to them an alternative version of a change to the social and economic 15
status quo.

A SHORT HISTORY OF THE POLITICAL ECONOMY OF BREXIT | FETZER

As a result, rather than winning more seats, the Conservatives lost seats and had to enter
into an agreement with the pro-Brexit Northern Irish Democratic Unionist Party (DUP).
Contrary to the original plan, this handed the Brexit hardliners in government even more
power. The natural consequence was a drawn-out political struggle over the next few
years, with Parliament unable to find a resolution to the crisis that was developing as each
compromise was voted down and, ultimately, sabotaged by the Brexit hardliners.

This reached the point when Theresa May was ousted by her party and Boris Johnson
installed as new prime minister to implement Brexit. After years of parliamentary
manoeuvres, a simple message of “getting Brexit done” resonated with voters and Johnson
was able to win an outright majority in the 2019 general election.

THE ECONOMIC DAMAGE OF BREXIT IS WIDENING EXISTING LEAVE VERSUS

REMAIN CLEAVAGES

The economic damage of Brexit is becoming increasingly paramount. Most of the studies
that have attempted to quantify the economic costs to date focus on national-level
estimates (Born et al. 2019). Yet, these costs are likely to have distinct regional flavours. In
Fetzer and Wang (2020), we construct synthetic control estimates at the subnational level
in the UK which show that the costs of Brexit are highly unequally distributed. Not only
do they seem to contribute to economic divergence between the UK’s constituent nations,
but there are also notable differences within England. Most notably, we observe that
London is far less affected than regions with a notable manufacturing sector presence
(see Figure 5).

This is interesting, as most of London came out strongly pro-Remain. At the regional
level, we find a significant correlation between economic damage that Brexit has caused
up to 2019 and the level of support for Leave in 2016. This goes quite contrary to the
political ambitions set out in the Conservative Party’s 2019 election manifesto, which
suggested that the UK could ‘level up’ or even out regional economic divides. Not only
has Brexit increased regional divergence, it has also made it harder to ‘level up’ as it has
reduced the overall size of the economy and thus the resources available.
 FIGURE 5 SYNTHETIC CONTROL ESTIMATES OF GDP ACROSS ENGLISH REGIONS VIS-À-
VIS THE ACTUAL PATTERN

Panel A: North East Panel B: North West Panel C: Yorkshire & the Humber

16
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

Panel D: East Midlands Panel E: West Midlands Panel F: East

Panel G: London Panel H: South East Panel I: South West

It is not clear how the economic and social challenges laid bare by austerity, the
resulting Leave vote and the subsequent process of Brexit will be addressed . It is even
unclear whether there are any honest political intentions to get to the root of the many
economic challenges, which are ultimately down to the UK’s own economic and political
institutions. The electoral system has enabled de facto minority rule and a culture that
breeds polarisation and zero-sum politics. The first-past-the-post system can create
perverse outcomes in a country whose population actually shows a demand for plurality
in politics. The present system cannot account for such plurality, and any outsider party
– such as UKIP or the subsequent Brexit Party – can notably affect an electoral outcome.

Going forward, the social and economic headwinds that the UK is facing are manifold. The
hard Brexit that became a political reality in 2021 is set to further exacerbate economic
inequalities. The higher cost of living that trade barriers imply are invariably passed on to
consumers, as are higher labour costs due to a more restrictive immigration regime and
likely increases in market concentration.

The UK is at a crossroads. It can choose to redesign its social contract to ensure that
the country works for young and working-age adults again. Core economic, but also
political, reforms are needed. On the political side, the UK urgently needs to reform its
electoral system to ensure that there is genuine political competition between ideas that
produces more consensual politics in which the focus is on finding policies and solutions
to increase the economic pie, rather than on in engaging in destructive zero-sum politics.
The UK should also enable meaningful devolution, empowering local decision makers
who are subject to political competition. This plurality within the country will enable
and encourage the type of experimentation and knowledge exchange that can produce 17
better public goods at the local level. Naturally, this has to be supported by a mechanism
for redistribution within the country – for example, through a regional cohesion fund that

A SHORT HISTORY OF THE POLITICAL ECONOMY OF BREXIT | FETZER

follows a clear policy function and is not subject to nepotism or political capture.

The alternative policy path that the UK could take is very concerning. The economic
headwinds are very strong and could produce new extreme policy swings or ruptures.
Yet, given the skewed political competition and the high degree of political power of the
incumbent Conservative Party, there is a genuine risk that, rather than making Britain
work for everyone, politicians find that repression is a viable option to quell social unrest
and dissent that cannot find its voice in the current political system.

REFERENCES

Alabrese, E, S O Becker, T Fetzer and D Novy (2019), “Who voted for Brexit? Individual
and regional data combined”, European Journal of Political Economy 56: 132–150.

Autor, D H, D Dorn and G H Hanson (2013), “The China Syndrome: Local labor market
impacts of import Competition in the United States”, American Economic Review 103(6):
2121–2168.

Beatty, C and S Fothergill (2013), Hitting the poorest places hardest: The local and
regional impact of welfare reform, Centre for Regional Economic and Social Research,
Sheffield Hallam Univeristy.

Becker, S O and T Fetzer (2018), “Has Eastern European Migration Impacted British
Workers?”, CAGE Working Paper.

Becker, S O, T Fetzer and D Novy (2017), “Who voted for Brexit? A comprehensive district-
level analysis”, Economic Policy 32(92): 601–650.

Born, B, G J Müller, M Schularick and P Sedláček (2019), “The Costs of Economic

Nationalism: Evidence from the Brexit Experiment”, Economic Journal 129(10): 2722–
2744.

Colantone, I and P Stanig (2018), “Global competition and Brexit”, American Political
Science Review 112(2): 201–218.

Dippel, C, R Gold and S Heblich (2015), “Globalization and Its (Dis-)Content: Trade
Shocks and Voting Behavior”, NBER Working Paper No. 21812.

Fetzer, T (2019), “Did Austerity Cause Brexit?”, American Economic Review 109(11): 3849–
3886.
 Fetzer, T (2020), “Austerity and Brexit”, Intereconomics 55(1): 27–33.

Fetzer, T and S Wang (2020), “Measuring the Regional Economic Cost of Brexit: Evidence
up to 2019”, CAGE Working Paper No. 486.

18 Fetzer, T, S Sen and P C Souza (2019), “Housing insecurity, homelessness and populism:
Evidence from the UK”, CAGE Working Paper.
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

Graetz, G and G Michaels (2015), “Robots at work”, Review of Economics and Statistics,
forthcoming.

Innes, D and G Tetlow (2015), “Delivering Fiscal Squeeze by Cutting Local Government
Spending”, Fiscal Studies 36(3): 303–325.

Portes, J (2019), What Do We Know and What Should We Do About Immigration?, Sage
Publishing.

Rogerson, R (2008), “Structural Transformation and the Deterioration of European

Labor Market Outcomes”, Journal of Political Economy 116(2): 235–259.

ABOUT THE AUTHOR

Thiemo Fetzer is a Professor in the Department of Economics at the University of

Warwick and a Visiting Fellow at the London School of Economics. He is also affiliated
the Pearson Institute at University of Chicago, the Spatial Economics Research Group
at London School of Economics and the Centre for Competitive Advantage in the Global
Economy (CAGE) at Warwick. He completed his PhD in Economics at the London School
of Economics and held a visiting appointment at the Harris School of Public Policy at the
University of Chicago. Thiemo’s research interest lie in the fields of political economy and
economic development. His work has been published in journals, including the American
Economic Review, the Economic Journal, Review of Economics and Statistics and the
Journal of the European Economics Association.
CHAPTER 2
Brexit Britain in a changing global 19

economy

BREXIT BRITAIN IN A CHANGING GLOBAL ECONOMY | POSEN AND RENGIFO-KELLER

Adam S. Posen and Lucas Rengifo-Keller
Peterson Institute for International Economics (PIIE) and CEPR; PIIE

After the 2016 referendum but prior to the implementation of Brexit, two main points
about its coming economic impact were evident (Posen 2017, 2019b). First, gravity matters.
One of the few things that economics can treat as nearly a physical law is that economies
trade and invest primarily with the economies that are closest to them geographically and
historically. Brexit was running in the face of that. Second, Brexit would affect more than
just international trade. Instead, it had to be seen in a political economy context as well
as an economic context of broader EU–UK commercial interactions involving foreign
direct investment, financial flows, information networks, and immigration. As has been
amply documented, most economic trends post-Brexit have come out pretty much as the
mainstream economists expected (Ayele et al. 2021, Bakker et al. 2022, Crowley et al.
2022, Freeman et al. 2022, Portes 2021, Sumption 2022). These studies emphasise the
econometric identification of specific Brexit impacts, in keeping with current trends in
economic research.

Our purpose in this chapter is to put Brexit in more of a global context as a regime change
in the UK’s approach to the global economy. We begin by assessing how British economic
openness compares to that of other large free market economies before and after Brexit.
We then set out the changing nature of the global economy – what one of us has called
the ‘corrosion of globalisation’ (Posen 2019a, 2022a) – which Covid-19 and the Russian
invasion of Ukraine have accelerated. Both are important additions because the Brexit
discussion has quite understandably tended to focus on the domestic impact and politics,
with little attention paid to the diplomatic response from the EU. The global context,
however, will shape Britain’s outcomes and policy options as much as the immediate
economic effects of Brexit implementation. Finally, we offer a ‘Global Britain’ strategy
which is actionable and economically constructive in this changed context.

A SUDDEN SHARP ECONOMIC WITHDRAWAL BY THE UK

As we will show below, during the 1990s through the mid-2010s until the Brexit
referendum, the UK was one of the most open large economies in the world. On trade, it
was comparable to France, Germany, Italy, and Spain, and it had larger inflows of FDI,
immigration, foreign students, and financial capital; it far outstripped the US, Australia
 and Canada proportionally for its size on the same measures. Thus, Brexit was not a small
adjustment but potentially a shift in the entire orientation of the UK economy, because it
would impose a huge terms-of-trade shock to trade with its primary trading partner. The
UK was going to be “mugged by economic reality” (Posen 2017) because the basic idea of
20 defying economic gravity was always self-defeating.1 Put differently in economic terms,
once you defy gravity, you end up lost in space.
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

In order to disentangle Brexit from Covid-19 and some other factors, we analyse its effects
on the UK’s openness primarily through a comparative lens. The advanced economies
that we compare Britain against – namely, the US, Canada, Australia, Japan, Germany,
France, Italy, and Spain – have all experienced roughly simultaneously the same shocks
brought on by the pandemic, reopening, and the disruption to energy markets from Russia’s
invasion of Ukraine. Informally, we are doing a difference-in-difference comparison of a
policy shift. We see the UK as fitting within two groups for comparison: the European
group, which includes the European countries in the sample; and the ‘Liberal Pacific’
group, which includes the English-speaking countries in the sample plus Japan.

We examine time series data on trade in goods and services as a percent of GDP in order
to discern openness to trade. We rely on quarterly data (running until the end of 2021)
from the OECD’s Quarterly National Accounts Database. On the immigration front, we
use data compiled by the United Nations Department of Economic and Social Affairs’
Population Division. The data show the average five-year average percent change in each
country’s migrant stock. The use of a five-year average helps to smooth out volatility and
delivers a clear look at trends in immigration growth. The dataset regrettably ends in
2020, so we turn to more recent UK data for a more up-to-date picture. The UK’s Office
of National Statistics publishes annual spreadsheets; we merged these separate ONS
spreadsheets to create an annual time series dataset which runs until 2021. We use World
Bank data on foreign direct investment inflows as a share of GDP to get a sense of how
open and appealing each country is to foreign investment. We use four-year averages to
smooth because such data can be noisy and easily influenced by large transactions in any
given year.

In Figure 1, we show the change in the measure of trade openness based on standard
imports plus exports as a share of GDP, from the start of 2017 to the latest available data
at the end of 2021. The idea of making this comparison is not just that it is since Brexit,
but also so that we ignore the downward valley of trade that Covid-19 prompted. What we
can see is that the overall decline in trade for the UK is much sharper than for anyone else.
Canada suffers a trade decline largely because it was exporting energy and there was a
period of low energy use and prices, which have since risen again. We can see that on net,

1 In the long-running West End musical Wicked, there is a triumphant song by the lead character at the end in which she
sings: “I’m through with playing by the rules of someone else’s game…It’s time to try defying gravity, and you can’t pull
me down!” She rises up (on her broom) over the mindless throng, while her sister pleads, “Can’t I make you understand
you’re having delusions of grandeur?” Of course, we all know that however understandably frustrated she may have
been, a couple of decades later another sister is under a house and then she herself gets dissolved by water in total
defeat of her plans to defy gravity.
and again smoothing out Covid-19, trade has continued to grow for the main economies
of Europe. It has also continued to grow for Australia and Japan. Even for the US, with
President Trump’s term and current President Biden doing nothing to reverse Trump on
trade, trade shrunk less than it did for the UK.
Figure 1
21
UK trade
FIGURE 1
openness has fallen more sharply
UK TRADE OPENNESS HAS FALLEN MORE SHARPLY THAN THAT OF OTHER
than other advanced economies since Brexit

BREXIT BRITAIN IN A CHANGING GLOBAL ECONOMY | POSEN AND RENGIFO-KELLER

ADVANCED ECONOMIES SINCE BREXIT

Percent change in trade in goods and services as a share of GDP from Q1 2017 to Q4 2021

a. UK

UK -6.2%

b. Europe

France 3.8%
Germany 8.3
Spain 9.3
Italy 14.5

c. Liberal Pacific

-4.4% Canada
-3.7 US

Australia 0.6
Japan 17.7

Note: Liberal Pacific group refers to large, advanced economies in the Pacific region, namely the United
States, Canada, Australia, and Japan.

Sources: Organization for Economic Cooperation and Development (OECD), Quarterly National
Accounts Database.

In Figure 2, the story gets a little more complicated. The basic message is that, except for
Canada, everybody sees a recovery in trade following Covid-19, while the UK basically
sits flat. There is no simple mapping between openness and per capita GDP growth, so
this is not necessarily a welfare loss. It is a statement, however, that at least top-down,
post-Brexit Britain does look different from its peers: it’s trade has recovered less from
Covid-19 than others, which represents a sort of difference in difference approach.2

2 A similar argument can be made about the transmission and persistence of inflation shocks in 2021 and early 2022 in the
UK compared to its EU neighbours (Posen 2022b).
 Figure 2

UK trade
FIGURE 2 hasHASnot
UK TRADE recovered
NOT RECOVERED from
FROM THE theASpandemic
PANDEMIC QUICKLY AS
as quickly as other
OTHER ADVANCED advanced economies
ECONOMIES

Quarterly trade in goods and services as a share of GDP, Q1 2017–Q4 2021

a. Europe b. Liberal Pacific
22
100% 100%100

Germany
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

80 80 80

Spain Canada
France
60 60 60

UK
Italy
UK
Australia

40 40
Japan
40

US

20 20
Q1 2018 2019 2020 2021 Q4 Q1 2018 2019 2020 2021 Q4
20

2017 2021 2017 2021

Note: Liberal Pacific group refers to large, advanced economies in the Pacific region, namely the United
States, Canada, Australia, and Japan.

Sources: Organization for Economic Cooperation and Development (OECD), Quarterly National
Accounts Database.

Looking at receptiveness to foreigners, the UK had a long period, starting in the early
2000s, when it was attracting and retaining a large number of foreign-born people. Figure
3 shows the growth in the foreign-born population, which is a proxy for immigration. After
Brexit, EU immigration tails off and becomes negative. This should be no surprise, but
the sharpness of the drop seems to be unrecognised. As Portes (2021, 2022) has pointed
out, importantly, the UK has not seen a decline in non-EU immigration and in other
nationally collected data (not shown), there has been even a significant surge. It is hopeful
that, when so much of the pro-Brexit referendum campaign seemed to have been run on
xenophobia (if not outright anti-migrant claims), we see a shift in terms of migration from
EU sources to elsewhere. Whatever the expansion in non-EU migration, however, it is not
strictly compensatory for the lost EU workers; importing more doctors and students does
not make up for workers in agriculture, transport, hospitality and construction.
 Figure 3

UK foreign-born
FIGURE 3 population
UK FOREIGN-BORN POPULATION GROWTHgrowth
HAS SLOWED has
SINCE THE BREXIT
slowed since the Brexit vote
VOTE

Average annual growth of foreign-born population in the UK by place of birth, 2001–2021

25%
23
EU
20

BREXIT BRITAIN IN A CHANGING GLOBAL ECONOMY | POSEN AND RENGIFO-KELLER

15

10

Non-EU 1.7%

0
-0.7%

-5
2001 2005 2010 2015 2021

#PIIECharts Learn more at piie.com/research/piie-charts

Note: Foreign-born population refers to migrants born abroad but living in the UK, including those on
temporary visas.

Source: UK Office of National Statistics.

Looking at the left-hand panel of Figure 4, which shows comparable Europe, Spain and
Italy had major surges in migrants from the Mediterranean early in the 2000s. They
capped these off very quickly, and growth then fell back down. On average, if you were
to leave out that period, the UK accepted more migrants on average than the rest of
the large European economies over this period. Post-Brexit (unfortunately there are
comparable data only up to 2020), the UK is on a downwards trend while everybody
else in ‘Rich Europe’ was back on an upwards trend before the invasion of Ukraine.
When we look at the Liberal Pacific economies, the picture is a little more mixed but
the UK had been in pole position (although, contrary to a lot of expectations, there were
significant periods, including of late, in which Japan was challenging the UK for the
highest migrant population growth among the Liberal Pacific economies and in which
the US was steadily going down).3 Using this dataset and splitting the UK’s foreign-born
population into EU and non-EU buckets helps show that non-EU immigrant population
growth remained positive during the Covid-19 pandemic, which suggests that the decline
in EU immigration stems from policy – rather than public health – changes. The UK’s
immigration slump was predictably exacerbated by the end of free movement. So, the UK
has ceased to be more open than others to migration, and it has not replaced what EU
migrants specifically brought to the economy.

3 Posen (2021) documents how strong and how long the US withdrawal from globalisation has been, including the reduction
in migration since the mid-90s. The UK remains much more open than the US, but that is not necessarily the standard
that you want, especially when even Japan and Australia, which have their own very fraught racist histories, have opened
up significantly to migrants in recent years.
 Figure 4

UK immigrant
FIGURE 4
population growth has fallen
UK IMMIGRANT POPULATION GROWTH HAS FALLEN BEHIND OTHER
behind otherECONOMIES
ADVANCED advanced economies
Annual growth in number of migrants, 1990–2020
a. Europe b. Liberal Pacific
24 20% 5%
Spain
UK
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

15 4
Japan
Italy

10 3
US

5 UK 2
Canada
Germany
Australia
0 1
France

-5 0
1990 1995 2000 2005 2010 2015 1990 1995 2000 2005 2010 2015
–95 –2000 –2005 –10 –15 –20 –95 –2000 –2005 –10 –15 –20

Note: Liberal Pacific group refers to large, advanced economies in the Pacific region, namely the United
States, Canada, Australia, and Japan. Date ranges show data from July 1 of the first year in range to
June 30 of the last year.

Sources: United Nations Department of Economic and Social Affairs, Population Division (2020). International
Migrant Stock 2020.

There has been a lot of attention in the UK on financial capital flows, whether it be the
City of London managing others people’s money, oligarchs from Russia and Middle East
putting money into London real estate, or part of the long history of worrying about the
pound. What often gets overlooked is the issue of how much foreign direct investment the
UK used to attract (both brownfield and greenfield). There is clear evidence that inward
foreign direct investment is extremely positive for growth, tending to lead to higher-wage
jobs, more advancement in innovation and technology, more transfer of skills, and more
sales of associated business services (Mathew et al. 2021, Moran et al. 2005).

Yet, already by 2017, it was clear that while the UK auto industry (meaning Toyota, Nissan,
Ford, Vauxhall, and Land Rover) would not immediately stop producing cars in the UK,
it would soon stop producing them for export to Europe. Therefore, the plants and the
associated investment would decline over time and would not be replaced (Posen 2017).
This is what we are seeing across other industries as well as autos.

Figure 5 shows multi-year average data on foreign direct investment inflows, going back
to a global perspective. By their nature, FDI flow data are extremely choppy because one
very large telecom deal or financial market deal can bias the data for a given year. Similar
to immigration, but over a longer period in the 1990s and early 2000s, the UK was a
leading recipient of inwards FDI. It is no coincidence that this period coincides with
the Maastricht Treaty allowing the UK a prime role as a bridge into European markets
 Note: Liberal Pacific group refers to large, advanced
for American, Chinese, and Japanese companies. However, FDI inflows to the UK have
States, Canada, Australia, and Japan.

Sources: Organization for Economic Cooperation and D

fallen back in recent years. There is one particular big transaction in financialAccounts
services in
Database.

2016 that drives up the penultimate data point, but taking that out, the UK is no longer
defying gravity or a positive outlier compared to other, similar economies in terms of
inward foreign direct investment. 25
Figure 5

Foreign direct investment into the UK has

BREXIT BRITAIN IN A CHANGING GLOBAL ECONOMY | POSEN AND RENGIFO-KELLER

FIGURE 5FOREIGN DIRECT INVESTMENT INTO THE UK HAS FALLEN SINCE THE BREXIT
fallen VOTE
since the Brexit vote
4-year average foreign direct investment inflows as a share of GDP, 1973–2020
a. Europe b. Liberal Pacific
10% 10%

8 8
UK UK

6 6

4 4

France Australia
2 2
Spain Germany
Italy US Japan
Canada
-0 0
1973 1985 1993 2001 2009 2017 1973 1985 1993 2001 2009 2017
–76 –88 –96 –04 –12 –20 –76 –88 –96 –04 –12 –20

Note: Liberal Pacific group refers to large, advanced economies in the Pacific region, namely the United
States, Canada, Australia, and Japan.

Sources: World Bank, World Development Indicators database.

SHRINKAGE AFTER THE COLD DIP OF BREXIT

Our international comparison is along three dimensions: immigration, foreign direct

investment, and volume of trade. We turn to spider charts in an effort to bring all three
of our openness categories together. Spider charts, sometimes called radar charts, are
often used to compare individual entities across several unique dimensions. In our case,
we compare the UK to its large European neighbours and to the countries of the Liberal
Pacific across the three categories discussed previously: trade in goods and services as
a share of GDP, the growth of the migrant stock, and inward FDI as a share of GDP.
Each of these represents an important aspect of openness – the more open a country
 is relative to its peers, the larger its triangle will be. The shrinkage of the UK’s triangle
therefore illustrates the extent to which Brexit has reduced its openness vis-à-vis the
other countries in the sample.4

Our methodology is simple but deserves some attention. We place the countries in their
26
respective samples, standardise the data for each country along each dimension, and then
plot the standardised values. The length of a triangle’s vertex in a given dimension – i.e.
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

the distance from the centre of the chart – indicates the size of a country’s z-score in
that dimension. Germany, for instance, has maintained a higher level of trade as a share
of GDP than its neighbours. This explains the long length of its trade dimension with
respect to the other European countries.

The left two triangles in Figure 6 compare the UK to the rest the other major economies
in Europe (top) and in the Liberal Pacific (bottom) before Brexit. The red triangles show
just how much more open the UK was just a decade or even seven years ago. Looking at
the areas of the triangles in the bottom-left chart, the UK was far more open in economic
terms than Canada, Australia, the US, and Japan. In comparison to Europe, the UK was
the most open by far in terms of FDI and immigration, and was roughly average in trade.

If we shift to the right column of Figure 6, we are looking at post-Brexit and the red
triangles have dramatically shrunk. Now the UK is essentially the lowest or second-lowest
in Europe on inward FDI, immigration, and trade. It remains noticeably more open on
trade than the Liberal Pacific economies, but not more so on immigration or inward FDI.

In Figure 7, the dark red triangle in is the current state of the UK, while the hot pink is the
previous state of the UK pre-Brexit. The lines in the background are based on the average
of countries in the sample. Whether you look at rich countries, EU members, or even non-
EU countries, the shrinkage economically of the British world is evident.

4 In work underway, we have been trying to measure what tariffs the UK faces, i.e., measures of not so much how open the
UK, is but what access it has to the rest of the world. This is another dimension along which Brexit may well have shifted
the economic integration of the UK with the rest of the world.
 Figure 6

The UK was far more open to trade and

immigration,
FIGURE 6
and attractive for FDI pre-Brexit
THE UK WAS FAR MORE OPEN TO TRADE AND IMMIGRATION, AND
ATTRACTIVE FOR FDI PRE-BREXIT

Trade and inward FDI as a share of GDP and immigrant population growth, relative values
a. Europe

Pre-Brexit Post-Brexit 27
Trade Trade
UK
Germany

BREXIT BRITAIN IN A CHANGING GLOBAL ECONOMY | POSEN AND RENGIFO-KELLER

France
Spain
Italy

Immigration Inward FDI Immigration Inward FDI

b. Liberal Pacific

Pre-Brexit Post-Brexit

UK Trade Trade
Canada
US
Australia
Japan

Immigration Inward FDI Immigration Inward FDI

Note: Values show relative position of each country in sample. Pre-Brexit trade is Q1 2016 trade as a share of
GDP; inward FDI is 4-year average inward foreign direct investment as a share of GDP, 2013–16;
immigration is 6-year average immigrant population growth, 2010–15. Post-Brexit trade is Q4 2021
trade as a share of GDP; inward FDI is 4-year average inward foreign direct investment as a share of
GDP, 2017–20; immigration is 6-year average immigrant population growth 2015–20.

Sources: World Bank, World Development Indicators; Organization for Economic Cooperation and
Development (OECD); United Nations Department of Economic and Social Affairs.
 Figure 7

Brexit has stifled UK trade, FDI inflows, and

immigration
FIGURE 7 BREXIT HAS STIFLED UK TRADE, FDI INFLOWS, AND IMMIGRATION

UK trade and inward FDI as a share of GDP and immigrant population growth pre- and
post-Brexit, relative values

a. Relative to all countries b. Relative to EU countries b. Relative to non-EU

in sample in sample countries in sample
28
Trade Trade Trade
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

Post-Brexit

Pre-Brexit

Immigration Inward FDI Immigration Inward FDI Immigration Inward FDI

Note: Values show relative position of UK in each sample. Pre-Brexit trade is Q1 2016 trade as a share of GDP;
inward FDI is 4-year average inward foreign direct investment as a share of GDP, 2013–16; immigration
is 6-year average immigrant population growth, 2010–15. Post-Brexit trade is Q4 2021 trade as a share
of GDP; inward FDI is 4-year average inward foreign direct investment as a share of GDP, 2017–20;
immigration is 6-year avarage immigrant population growth 2015–20. Sample consists of EU countries
(France, Germany, Italy, Spain) and non-EU countries (Australia, Canada, Japan, United States).

Sources: World Bank, World Development Indicators; Organization for Economic Cooperation and
Development (OECD); United Nations Department of Economic and Social Affairs.

A GRAVITY-RESPECTING STRATEGY FOR GLOBAL BRITAIN

What remains to play for is what economists refer to as the dynamic effects of trade and
investment. The dynamic effects are essentially what affects the trend growth rate going
forward – it is not just the volume of what you are doing and what inputs you have, it
is how well you are using them. The basic concern when one shrinks one’s triangle of
trade and openness is that you are losing competition internally. Therefore, you get less
innovation and less turnover and dynamism in your corporate sector, your investments,
and your labour force. Your dynamism also tends to decrease when you lose diversity
of talent coming in, diversity of corporate cultures coming in, and so on.5 Again, in the
empirical literature we have, the direct effect of trade on productivity growth is there
but not terribly clear. The direct effect of immigration and FDI on productivity growth
is quite clear and strong. To the degree that you end up with less competition in the
economy, whether through trade restrictions or other means, that definitely negatively
affects productivity growth over the long run. The collapse in corporate investment in
the UK since Brexit, not recovering as it has in peer countries post-Covid, indicates this
coming to pass (Parker and Giles 2022).

Various British politicians and pundits, including the current prime minister, have at
times invoked ‘Global Britain’ as an ideal. But what would it mean to be a Global Britain
in the current context? The first thing to be said is that, going back to gravity, the UK is

5 See, among others, Campo et al. (2018), Costas-Fernandez (2018), Mathew et al. (2021), Moran and Oldenski (2013), Posen
(2021, 2022), and Sumption (2022).
not likely to expand that triangle of openness much, and certainly not the apex of trade,
by making deals with a bunch of non-European economies, at for a decade or more. While
the situation has changed a little since observed in 2017, the UK still has more trade
with Ireland than with all of the BRICS except China (and even China has only recently
just passed Ireland in terms of its share of UK trade). So, it is useful for UK consumers 29
and businesses to do a deal with Australia and New Zealand where the UK gives them
side payments to get back to where it was when it was a member of the EU. It is okay to

BREXIT BRITAIN IN A CHANGING GLOBAL ECONOMY | POSEN AND RENGIFO-KELLER

fantasise about a trade deal with India. The UK can certainly, for a bunch of strategic
and economic reasons, apply to join the Comprehensive and Progressive Agreement for
Trans-Pacific Partnership (CPTPP). But this is all just ‘small beer’, ‘small potatoes’, small
whatever you want to call it.

So, what should be the Global Britain strategy going forward? The UK is now a relatively
small economy in a big world, and it has to adapt. Adapting will be more challenging now
than it would have been even a few years ago. As former Prime Minister Harold MacMillan
once said, “events, dear boy, events” are what presents difficulties for political leaders.
Covid-19 and the Russian invasion of the Ukraine are real events. The anti-globalisation
politics in the US that have been building for more than 20 years (Posen (2018, 2021)
are a real trend, an event. The rise of China, and with that (by Chinese Communist
Party choice, not inevitably) the country’s increasingly autocratic actions at home and
aggressive actions in global economic affairs, are a real event. When the UK government
thinks about what the UK can and should do, it is not just a case of “we hold all the cards
in the negotiation” or “we’ll just make the best deals we can” – there is a real environment
that must be taken into account to determine what is possible and constructive.

As argued in Posen (2022a), a corrosion of globalisation is underway. In a world where

China and the US are becoming more domineering and the international rules-based
institutions are less reliable, many countries will end up having to choose sides – not
entirely like with sanctions on Russia, but similarly – and this will be even more evident
in the years to come. After Covid, a very reasonable re-evaluation of global supply chains
was already underway due to issues of shortages and resilience. Then comes Russia’s
invasion of Ukraine, and politicians take it too far but, quite reasonably, businesses
decide that they may not want all of their sourcing from one place, or certainly not from
places that are geopolitically or ethically suspect. The result of all this is what the Biden
administration calls ‘friend-shoring,’ the EU (or at last France) calls ‘strategic autonomy’,
and what China calls a ‘dual circulation strategy.’

The current situation is in some ways analogous to Brexit, in the sense that you can set
out the pros and cons and it is entirely reasonable for elected officials or voters to say, “I do
not care about the economic consequences as much as I do my values, my sovereignty, my
security”. But it is also similar to Brexit in that this is a world where trading opportunities
are going to be more constrained, where investments will be losing some economies of
scale, where you are taking out more self-insurance (which is costly, even if the rational
thing to do). All of this lowers your returns on capital; it also reduces competition,
 where self-interested special interests and companies will be increasingly able to exploit
positions by saying “I am your national (or regional) champion and you have to protect
me”. At the same time as Britain wants to be self-consciously global, it is getting harder
to be so.
30
Defying gravity leaves you even further out in space if you are not part of one of the
increasingly closed or separated big economic blocs. The UK’s chances of being ignored,
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

left out, of floating away are increasing. It is increasingly unanchored to anything in

economic terms, and the things to which it is anchored, like the WTO or the its alliance
with the UK, are becoming politicised and less valuable anchors. This is directional. It is
not like the UK is going to disappear as an economy, but it is no longer just about being
excluded from some access to the EU market. The UK is certainly not part of the EU’s
strategic economy plans; and for the US, particularly for manufacturing, it is not clear
what advantage the UK has over Mexico or other developing economies that have cheaper
wages and potentially more growth.

Given this reality, the way forward for a Global Britain in this world has four elements.
First, UK decision makers and informed public opinion have to resist the delusions of the
‘imperial hangover’ even more so than before. This is a problem for the US, it is a problem
for France, it is a problem for every major economy; but it will require more change for
the UK. The idea that there is going to be a renewed economic Commonwealth, a special
relationship – let alone a free trade agreement – with the US, or a special relationship
with India is just fanciful. Worse would be letting the UK’s security ambitions and
foreign policy grandeur, which still have more justification than its economic ones, drive
economic decisions. National security and economics will be joined together in coming
years, unfortunately, but probably not in ways that are favourable to the UK. The UK
suddenly announced it would be a leader in the WTO or in other economic organisations,
whatever that would mean, when no one else was asking for UK leadership. Better to
follow constructively – try to make sure that you get to join the right clubs from here on
out, and be a good member of those clubs you are in (PIIE 2020, Posen 2021).

Second, to make Global Britain work economically today is to not think so much about
trade deals but fundamentally about attracting good labour and capital to the UK.
The prospects have not been as badly damaged on these fronts as on trade, despite the
shrinkage of openness on these dimensions. We have not seen the total anti-immigration
turn that might have followed Brexit; nor have we seen the financial race to the bottom
in deregulation that Brexit might have portended. We have even seen, due to tragic other
events, at least some sense of turning away from capital inflows that are just forms of
money laundering. This is something the UK can build on, and where it can be a leader.
For one thing, attracting the students of the world is a very high value proposition. If the
US continues to behave as it is currently behaving, the UK’s market share here should only
grow, and it will mean more to the UK than the US because the UK market is smaller.
Similarly, attracting FDI rather than short-term flows of capital may no longer be feasible
in the auto sector, for example, but it is feasible in R&D and specialised services.
Third, let gravity pull the UK economy back to earth before its industrial structure ends up
floating away. The UK really should go back to a soft Brexit deal, the European Economic
Area (EEA) or European Free Trade Association (EFTA). There is no substitute for this
– and the corrosion of globalisation is making this even more the case than it used to be.
Essentially, regulatory convergence on the European model and accepting that in trade 31
issues the UK is going to become like Switzerland or Norway is really what it should be
doing. Go ahead with CPTPP accession, go ahead with other bilateral deals, but what is

BREXIT BRITAIN IN A CHANGING GLOBAL ECONOMY | POSEN AND RENGIFO-KELLER

really going to bring the UK back to earth is – by hook or by crook, quietly or openly –
reversing some economic aspects of Brexit.

Finally, the one place where the UK has often lived up to its economic self-image and
ideals is being a mediator of sorts between the US and the EU in international economic
rule making. Not a mediator in the literal sense of mediating disputes, although that did
happen on occasion, but essentially a middle force intellectually between American and
European policy proclivities in the economic arena. This is still reflected in the figures
above depicting UK openness. While the UK (the shrinking red triangle) has become
more closed compared to the big economies of Europe, it remains at least as open, and
in some ways still much more open, than the US and the other Liberal Pacific nations.
There is room – in particular, in the setting of standards over technology transfer, foreign
direct investment, human rights, investment, and financial services – for the UK to play
a constructive role that would be in its own self-interest and in the world’s self-interest.

REFERENCES

Ayele, Y, G Larbalestier and N Tamberi (2021), “Post Brexit: Trade in Goods and Services”,
UK Trade Policy Observatory Working Paper 63.

Bakker, J D, N Datta, J De Lyon, L Opitz and D Yang (2022), Post-Brexit imports, supply
chains, and the effect on consumer prices, UK in a Changing Europe, 27 April.

Campo, F, G Forte and J Portes (2018), “The Impact of Migration on Productivity and
Native-Born Workers’ Training”, IZA Discussion paper No. 11833.

Costas-Fernandez, J (2018), Examining the Link between Migration and Productivity,

Report for the Migration Advisory Committee.

Crowley, M, L Han and T Prayer (2022), “The Pro-competitive Effects of Trade

Agreements”, UK in a Changing Europe.

Freeman, R, K Manova, T Prayer and T and Sampson (2022), “Unravelling deep integration:
UK trade in the wake of Brexit”, Centre for Economic Performance Discussion Paper 1847.

Mathew, A, K Koo, E Hatwell, M Johnson, S Hamroush, M Needham and T Johnsen

(2021), Understanding FDI and its impact in the United Kingdom for DIT’s investment
promotion activities and services, Phase 2 Analytical Report, Department for
International Trade.
 Moran, T, E Graham, and M Blomstrom (2005), Does Foreign Direct Investment Promote
Development?, PIIE Press.

Moran, T and L Oldenski (2013), Foreign Direct Investment in the United States: Benefits,
Suspicions, and Risks with Special Attention to FDI from China, PIIE Press.
32
Parker, G and C Giles (2022), “The deafening silence over Brexit’s economic fallout”,
Financial Times, 20 June.
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

PIIE – Peterson Institute for International Economics (2020), “Trade Winds” podcast,
with Anabel Gonzalez, Adam Posen, and Elizabeth Truss, 8 July.

Portes, J (2021), “Immigration and UK Economy after Brexit”, UKICE Working Paper, 15
June 15.

Portes, J (2022), “Immigration Statistics: Looking Forwards, Looking Backwards”,

UKICE Commentary, 26 May 26.

Posen, A (2017), “Posen Discusses the Damage of Brexit to the British Economy”, Peterson
Institute for International Economics.

Posen, A (2018), “The Post-American World Economy,” Foreign Affairs, 13 February.

Posen, A (2019a), “The Future of the Global Economy: Integration or Fragmentation?”,

panel held at the 2019 Allied Social Science Association (ASSA) Annual Meeting, Atlanta,
GA, 5 January.

Posen, A (2019b), “The Macroeconomic Impact of Trade Wars (and Brexit counts as one)”,
King’s College London, Business School Centre for Data Analytics, 24 March.

Posen, A (2021), “The Price of Nostalgia”, Foreign Affairs, 20 April.

Posen, A (2022a), “The End of Globalization? What Russia’s War in Ukraine Means for
the Global Economy”, Foreign Affairs, 17 March.

Posen, A (2022b) “Brexit reality bites as stagflation looms”, Financial Times, 12 May.

Sumption, M (2022), “What Has the Impact of Ending Free Movement Been So Far?”,
Migration Observatory.

ABOUT THE AUTHORS

Adam Posen is President of the Peterson Institute for International Economics and a
CEPR Distinguished Fellow. From 2009-12, during the global financial crisis, he served
as an external voting member of the Bank of England’s Monetary Policy Committee. He
is the author or editor of eight books, and has served as a visiting scholar at or consultant
to central banks throughout the G20 economies, as well as to the British and Japanese
Cabinet Offices. He received his PhD in Political Economy from Harvard University,
and has been the recipient of fellowships and major research grants from the American
Academy in Berlin, Bank of England, Brookings Institution, European Commission, Ford
Foundation, Sloan Foundation, and the US National Science Foundation.

Lucas Rengifo-Keller joined the Peterson Institute for International Economics as a

33
research analyst in January 2022. He works with Executive Vice President and Director of
Studies Marcus Noland on issues related to gender diversity, illicit drug trade, and North

BREXIT BRITAIN IN A CHANGING GLOBAL ECONOMY | POSEN AND RENGIFO-KELLER

Korea. Prior to joining the Institute, Rengifo-Keller was an associate at Charles River
Associates, where he worked on antitrust litigation and research. He also spent time as a
research assistant at the Georgetown University Initiative on Innovation, Development
and Evaluation (gui2de), a development economics research center. Rengifo-Keller
earned a bachelor’s degree in economics and a minor in mathematics from Georgetown
University
CHAPTER 3
The impact of Brexit on UK–EU trade 35

Rebecca Freeman, Kalina Manova, Thomas Prayer and Thomas Sampson

THE IMPACT OF BREXIT ON UK–EU TRADE | FREEMAN, MANOVA, PRAYER AND SAMPSON
London School of Economics; University College London and CEPR;
University of Cambridge; London School of Economics and CEPR

Brexit has led to an unprecedented unravelling of deep economic integration. Although

the UK voted to leave the EU in June 2016, its economic relationship with the Union
did not change until 1 January 2021, when the Trade and Cooperation Agreement (TCA)
governing future UK–EU relations came into effect.

Under the TCA, UK–EU trade is tariff and quota free, but the UK is no longer a member of
the EU’s Single Market or Customs Union. Consequently, implementation of the TCA has
led to higher trade costs due to the re-establishment of a customs and regulatory border
between the UK and the EU (Dhingra and Sampson 2022).

Previous research has found that products more exposed to Brexit uncertainty experienced
lower trade growth before and immediately after the referendum (Crowley et al. 2020,
Graziano et al. 2021). In recent work, we shed new light on the trade effects of Brexit
by analysing changes in UK trade with the EU relative to the rest of the world since the
referendum (Freeman et al. 2022).

We analyse both the period between the Brexit referendum and the end of 2020, when
there was uncertainty over what form Brexit would take, and the first year of trade
under the TCA. This allows us to study the trade effects both of news about a future, but
uncertain, trade cost increase, and of the realised impact of higher trade costs under the
TCA.

We find no evidence of a statistically or economically significant decline in the UK’s trade

with the EU relative to the rest of the world prior to the implementation of the TCA.
However, the introduction of the TCA caused a major shock to UK–EU trade.

We estimate that the implementation of the new trade relationship led to a sudden
and persistent 25% fall in UK imports from the EU, relative to the rest of the world. In
contrast, we find a smaller and only temporary decline in relative UK exports to the EU,
but nevertheless a large and sustained drop in the number of trade relationships between
UK exporters and EU importers. This suggests that the introduction of the TCA caused
many UK firms to stop exporting to the EU.
 How do we reach these conclusions? The main challenge is to disentangle the Brexit effect
from other causes of changes in trade, such as supply shocks that affect trade with all
countries and shifts in demand across products and destinations. For example, if the
Covid-19 pandemic caused import demand to fall more in the EU than in the rest of the
36 world, the UK’s relative exports to the EU may have declined for reasons unrelated to
Brexit.
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

To address this challenge, we use a difference-in-differences event-study methodology.

We compare growth in the UK’s trade with the EU to the UK’s trade with the rest of the
world, and to US and EU trade with countries other than the UK. This approach allows
us to account for changes in exporter supply and importer demand conditions in the UK
and abroad and, consequently, to isolate how Brexit has affected relative UK–EU trade. In
addition, to ensure our estimates are not biased by changes in the sectoral composition of
trade, we analyse trade at the product level for around 1,200 products.

Figure 1 shows our main results for UK exports to the EU (panel a) and UK imports from
the EU (panel b). For each quarter from 2013 Q1 through 2021 Q4, we plot the estimated
percent change in the UK’s trade with the EU relative to the rest of the world, compared
to the pre-referendum period of 2016 Q2. The shaded grey area shows the 95% confidence
interval for the estimates to indicate the degree of uncertainty about their precision. In
the figure, vertical lines identify the dates of the referendum, the UK’s withdrawal from
the EU in January 2020, and the start of the TCA. As the timing of the UK’s departure
from the EU coincided with the Covid-19 pandemic, we also denote the start of the first
and second Covid-19 waves in the UK in March and September 2020 with two green virus
icons along the horizontal axis.

FIGURE 1 UK TRADE WITH EU VERSUS REST OF THE WORLD

(a) Export Values - βt (b) Import Values - βt

Referendum Brexit TCA Referendum Brexit TCA
30% 30%

20% 20%

10% 10%

0% 0%

-10% -10%

-20% -20%

-30% -30%

-40% c c -40% c c
2013q1

2014q1

2015q1

2016q1

2017q1

2018q1

2019q1

2020q1

2021q1

2013q1

2014q1

2015q1

2016q1

2017q1

2018q1

2019q1

2020q1

2021q1

Notes: Panels (a) and (b) plot the estimated percentage changes in UK exports and imports with the EU versus the rest of
the world relative to 2016 Q2. 95% confidence intervals are calculated using standard errors clustered at the HS4 product-
region level.
Source: Freeman et al. (2022).
What do we learn from Figure 1? Panel (a) shows that, although relative exports to
the EU fluctuate from one quarter to the next, there is no sign that Brexit has led to a
sustained change in the geography of UK exports, either following the referendum or
after the introduction of the TCA. Total UK export growth compared to that of other
advanced economies was weak in 2021 (Springford 2022), but our analysis suggests that 37
this weakness cannot be explained by a decline in relative exports to the EU.

THE IMPACT OF BREXIT ON UK–EU TRADE | FREEMAN, MANOVA, PRAYER AND SAMPSON
Turning to UK imports, Panel (b) shows that, except for a temporary dip at the onset
of the Covid-19 pandemic, relative imports from the EU were unaffected prior to the
introduction of the TCA. On the other hand, the introduction of the TCA led to a deep
and sustained fall in relative UK imports from the EU. UK imports from the EU abruptly
declined by about 25% more than UK imports from the rest of the world after the TCA
came into effect, and this decline persisted throughout 2021.

These results provide novel evidence that trade flows are relatively unresponsive to
anticipated, but uncertain, increases in trade barriers. In contrast, we uncover patterns
consistent with the introduction to the TCA causing a substantial increase in UK–EU
trade costs and shifting UK imports away from the EU.

It is surprising that the TCA had a greater effect on imports than exports in 2021,
particularly since the UK delayed the introduction of many customs checks on EU imports
until 2022. We plan to study the reasons for this asymmetry in future research. For now,
we posit that it may relate to UK firms making interdependent input sourcing decisions
across origin countries, yet independent sales decisions across destination markets. It
may also reflect asymmetric market size effects and changes in fixed trade costs between
the UK and the EU.

To further understand the impact of the TCA, we analyse the extensive margin of trade
by studying the evolution of the number of trade relationships between the UK and the
EU. Although we do not observe the number of firms that trade with the EU, we measure
the number of trade relationships by counting the number of 8-digit products traded with
each country in the EU (and in the rest of the world) each quarter.

Figure 2 displays the estimated effects of Brexit on the number of trade relationships
using the same identification approach as in Figure 1 to isolate the ‘Brexit effect’. We
find that the introduction of the TCA led to a fall of around 30% in the number of export
relationships (or export ‘varieties’) with the EU relative to the rest of the world and a
smaller, but still significant, drop in the relative number of import relationships. The
decline in export relationships is driven by the exit of low-value relationships.
 FIGURE 2 NUMBER OF UK TRADE RELATIONSHIPS WITH EU VERSUS REST OF THE
WORLD

(a) Export Varieties - βt (b) Import Varieties - βt

Referendum Brexit TCA Referendum Brexit TCA
30% 30%

38 20% 20%

10% 10%
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

0% 0%

-10% -10%

-20% -20%

-30% -30%

-40% c c -40% c c
2013q1

2014q1

2015q1

2016q1

2017q1

2018q1

2019q1

2020q1

2021q1

2013q1

2014q1

2015q1

2016q1

2017q1

2018q1

2019q1

2020q1

2021q1
Notes: Panels (a) and (b) plot estimated percentage changes in the number of UK trade relationships with the EU versus
the rest of the world relative to 2016 Q2. A trade relationship (or variety) is defined as a CN8 product traded with a partner
country in a quarter. 95% confidence intervals are calculated using standard errors clustered at the HS4 product-region
level.
Source: Freeman et al. (2022).

Taking these results together, we conclude that the apparent stability in the relative value
of UK exports to the EU under the TCA, documented in Figure 1, hides a steep decline in
the number of export relationships, driven by the exit of small exporters. Consequently, it
would be a mistake to interpret our results as evidence that UK exporters were unaffected
by the introduction of the TCA. Instead, we conjecture that the TCA has increased the
fixed costs of exporting to the EU, causing small exporters to exit small EU markets, but
not (or at least not yet) severely hampering exports by the large firms that drive aggregate
export dynamics.

Our analysis has also uncovered substantial variation across products in how the TCA has
affected UK trade. We find that relative exports to the EU fell more in 2021 for products
with higher EU trade policy barriers, particularly on the extensive margin.

Moreover, when products are classified into capital, intermediate, and consumption
goods, we find that the value of UK exports to the EU relative to the rest of the world
increased under the TCA for capital and, to a lesser extent, intermediate goods, but there
was an offsetting reduction in exports of consumer goods. By contrast, the decline in the
UK’s relative imports from the EU is broad-based across all three types of goods.

While our research provides a rich description of how Brexit has affected UK trade, there
are a number of caveats to consider when interpreting the results. The sample ends in
2021, meaning that we only analyse the first year of trade under the TCA and do not
capture its long-run effects. Our findings may thus, in part, reflect temporary changes as
firms adjust to the new trading environment. We intend to update the paper and expand
our analysis as more data become available.
In addition, our estimation strategy is designed to capture the direct effect of Brexit on the
UK’s trade with the EU relative to its trade with the rest of the world. We do not analyse
whether Brexit has indirectly affected the UK’s trade with the rest of the world through
general-equilibrium adjustments or supply-chain linkages. Finally, we only study trade
in goods, and leave the equally important question of how Brexit has affected trade in 39
services to future work.

THE IMPACT OF BREXIT ON UK–EU TRADE | FREEMAN, MANOVA, PRAYER AND SAMPSON
REFERENCES

Crowley, M A, O Exton and L Han (2020), “The Looming Threat of Tariff Hikes: Entry
into Exporting under Trade Agreement Renegotiation”, AEA Papers and Proceedings 110:
547–51.

Dhingra, S and T Sampson (2022), “Expecting Brexit”, CEPR Discussion Paper 16970.

Freeman, R, K Manova, T Prayer and T Sampson (2022), “Unravelling Deep Integration:

UK Trade in the Wake of Brexit”, CEP Discussion Paper 1847.

Graziano, A G, K Handley and N Limão (2021), “Brexit Uncertainty and Trade

Disintegration”, The Economic Journal 131(635): 1150–1185.

Springford, J (2022), “The Cost of Brexit: December 2021”, Centre for European Reform
Insight.

ABOUT THE AUTHORS

Rebecca Freeman is a Trade Associate at the Centre for Economic Performance at the
London School of Economics. Her research focuses on empirical international trade,
with an emphasis on economic integration, global supply chains, and firm heterogeneity.
Previously, she worked at the World Bank, UN-ECLAC, the OECD, and the International
Labour Organization. Rebecca received her PhD in Economics from the Graduate Institute
in June 2019. She was awarded an MA in International Economics and International
Relations from Johns Hopkins University (SAIS) in 2014 and a BA in Economics from
Smith College in 2009.

Kalina Manova is Professor of Economics at UCL. She received her education at Harvard,
and previously taught at Stanford, Princeton, and Oxford. She is a CEPR Research Fellow,
Consultant at Bank of England and Inter-American Development Bank, Council Member
of the European Economic Association, and Editorial Board member at Review of
Economic Studies and American Economic Journal. She has received the UK Leverhulme
Prize in Economics, a €1.5 million grant from the European Research Council, and a
£750,000 ESRC Governance After Brexit grant. Her research in international economics
examines global value chains; financial frictions; and firm productivity, quality, and
management.
 Thomas Prayer is a PhD student at the University of Cambridge and an occasional
research assistant in CEP’s trade programme. His research interests lie in international
trade, trade policy, firm dynamics, and competition.

Thomas Sampson is an Associate Professor of Economics at the London School of

40
Economics. His research studies the impact of globalisation on workers, firms and
productivity. Thomas is also an academic consultant to the Bank of England and an
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

associate at the Centre for Economic Performance where he has worked extensively on
the economic consequences of Brexit. Prior to joining the London School of Economics,
Thomas worked as an Overseas Development Institute Fellow at the Bank of Papua New
Guinea and obtained a PhD in Economics from Harvard University.
CHAPTER 4
Post-Brexit imports, supply chains, 41

and the effect on consumer prices

POST-BREXIT IMPORTS, SUPPLY CHAINS AND THE EFFECT ON CONSUMER PRICES | BAKKER, DATTA, DE LYON, OPITZ AND YANG
Jan David Bakker, Nikhil Datta, Josh De Lyon, Luisa Opitz, and Dilan Yang
Bocconi University and Centre for Economic Performance (CEP); University College London
and CEP; University of Oxford and CEP; CEP; CEP

The UK’s exit from the EU led to a large increase in barriers to trade with its largest trading
partner. January 2021 marked the end of the UK’s participation in the Single Market and
the Customs Union, and the entry into force of the Trade and Cooperation Agreement
(TCA). Goods continue to be traded without tariffs and quotas but the regulatory and
customs framework for trade has changed, causing an increase in trade frictions between
the UK and the EU, set to be introduced gradually over time.

Current research suggests that Brexit precipitated a 25% fall in imports from the EU
(Ayele et al. 2021, Freeman et al. 2022). As around two-thirds of international trade is
in intermediate products – inputs to the production of other products (Datt et al. 2011,
Johnson and Noguera 2012) – this has major implications for UK firms and workers.
Costa et al. (2019) show that changes in intermediate imports due to the depreciation of
sterling reduced wages and training in affected sectors, highlighting the importance of
intermediate imports in international trade for worker outcomes. Furthermore, imports
increase the variety of products available to consumers (Karlsson 2011) and reduce their
prices (Jaravel and Sager 2019).

In a recent study (Bakker et al. 2022), we document that UK imports from the EU have
fallen both in absolute terms since the referendum and relative to imports from non-EU
countries since the TCA’s implementation. This is not the case for exports, where trade
with the EU has followed a similar path to trade with non-EU countries so far.

There is substantial heterogeneity across products in the effect of Brexit on trade flows,
which can be grouped into three common groups of patterns. First, some sectors, such
as textiles, experienced a slower rate of growth, if not decline, in UK imports from EU
countries compared to that of non-EU countries even before the implementation of the
TCA in 2021. This is suggestive of anticipatory adjustments by firms in some sectors.
Second, some products experienced a sizeable drop in the UK’s trade volume with EU
countries compared to that of non-EU countries after the TCA implementation, such as
fats and oils, which includes products such as olive oil, sunflower oil, and margarine.
Some of these products have experienced changes in their relative trade flows both before
 and after the implantation of the TCA . These include cameras, photographic plates, and
film. Third, there are some products for which there has been no observable difference
in UK–EU trade compared to UK non-EU trade, such as coal, gas, and petroleum oils.

Among the commodities that experienced a disproportionate dip in imports from the EU
42
compared to from non-EU countries, varying patterns are seen in the persistence and
timing of the dip . For some products, such as various types of meat including beef and
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

lamb, the dip is followed by a quick recovery back to pre-TCA trade levels. Other products
experienced a persistent drop or continued to decline over time, such as coffee, tea, and
spices.

In some cases where there is divergence between imports from the EU and from outside
the EU, there is evidence that goods have been substituted from EU to non-EU countries.
For example, vegetables and textile products experienced a downward trend in imports
from EU countries but an upward trend of imports from non-EU countries. In other
cases, there is a fall in imports from the EU but no obvious increase in imports from non-
EU countries (such as for fruits and nuts). This could either represent a fall in the overall
volume of consumption of the good or a switch to domestic providers.

Imports from the EU, relative to imports from non-EU countries, have fallen for goods
used as inputs to production in many UK supply chains. Since many firms initially chose
to source their inputs from the EU, it is likely to have been their most efficient option,
so this reallocation is likely to represent an increase in production costs and a fall in
efficiency, as cheaper access to intermediate inputs promotes productivity (Amiti and
Konings 2007).

Input-output data from the Office for National Statistics (ONS) provides measures of the
importance of each imported input to production for each sector in the UK. Matching
this with data on international trade, we construct measures of changes in imports of
intermediate inputs from the EU and from outside the EU for each UK sector. Some
sectors that rely heavily on imported inputs, such as motor vehicles, have experienced a
clear divergence in the source of its imports from the EU to outside the EU.

There is also evidence of stockpiling of imported intermediate products just before the
implementation of the TCA for some sectors, such as agriculture, fish-related products,
pharmaceuticals, and the automotive industry. Industries that were able to stockpile
their inputs may have been able to ease the initial frictions of the TCA entering into force.

Additional barriers at the border, such as customs checks, increased waiting times, and
additional paperwork, are likely to increase costs to UK importers or EU exporters. These
costs may then be passed on to consumers, causing an increase in prices. To quantify the
changes in prices due to the rise in trade barriers, we develop a novel dataset that matches
ONS micro price data with international trade data for highly detailed products.
Focusing on the food industry, we show that food products that were more reliant on
imports from the EU in 2015 saw larger increases in prices than those less reliant on the
EU both immediately after the 2019 election – when it was confirmed that the UK would
leave the Single Market and Customs Union – and the implementation of the TCA in
January 2021. Using a differences-in-differences approach, we estimate a 6% increase in 43
food prices due to Brexit, over the two years to the end of 2021. Products that had high
exposure to imports from the EU, such as fresh pork, tomatoes, and jams, experienced a

POST-BREXIT IMPORTS, SUPPLY CHAINS AND THE EFFECT ON CONSUMER PRICES | BAKKER, DATTA, DE LYON, OPITZ AND YANG
larger increase in prices than those with low EU import shares such as tuna and exotic
fruits like pineapple.

Figure 1 shows the results from the event study, which traces out the change in food prices
in percentage terms for food products with a high EU import exposure in comparison to
those with low EU import exposure, against a reference date of May 2016. The treatment
measure is the share of imports in 2015 that came from the EU for each product. The
method controls for macroeconomic shocks that affects all food prices, such as Covid-19.

FIGURE 1 IMPACTS OF BREXIT ON FOOD PRICES OVER TIME: EVENT STUDY ESTIMATES
.15
Estimated Effect Of Log Import Shares

.05 .1
On Log Prices
-.1 -.05 0 -.15

-45 -40 -35 -30 -25 -20 -15 -10 -5 0 5

Quarter Relative To Brexit

Notes: The figure presents estimates on the relationship between EU import exposure and food prices over time relative
to Brexit. Vertical lines represent Brexit-related events: the black line represents the month before the EU referendum, the
red line the 2019 general election, and the blue line the commencement of the TCA. Black dots are the estimates of the
interaction of the exposure measure and the quarter dummies, and vertical bars represent 95% confidence intervals.

The event study shows that in all quarters leading up to the Brexit referendum and in
almost all leading up to the 2019 election, there was no statistically significant difference
in price changes between food products that were more or less exposed to imports
from the EU. These results suggest that the parallel-trends assumption, required for
a causal interpretation using this method, is not violated during the ‘before’ periods.
 The implementation of the TCA aligns perfectly with a sharp, statistically significant
increase in prices for more-exposed products relative to those less-exposed. Although
the introduction of the TCA coincides with the third national UK lockdown, there is no
relationship between the exposure-induced price changes and the timing of the other
44 two lockdowns, nor the easing of restrictions from the third, suggesting that Covid-19 is
not driving the results. The statistically significant negative estimates found soon after
the referendum result are likely to be driven by the depreciation of the sterling relative to
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

other countries, which was stronger for non-euro currencies (Breinlich et al. 2018).

Despite a strong impact of Brexit on food prices, we do not detect impacts when examining
all products together. Impacts are likely more visible on food prices given the nature of the
products. First, food products tend to be more perishable, so they are more vulnerable to
border delays. Second, food products are close to being finished products in that they do
not tend to require many intermediate inputs or labour costs, meaning that there is more
likely to be observable passthrough for direct product-level matched data. Third, they are
among the most exposed product categories to Brexit-related non-tariff trade barriers,
such as sanitary and phytosanitary (SPS) measures which are set to be introduced.

Leaving the Single Market and Customs Union marked a major disruption in the UK’s
import patterns. In our study, we used data up to 30 September 2021 to analyse changes
in trade flows, supply chains, and consumer prices. While full customs checks will be
phased-in during 2022 and more data will be collected, we can already conclude that
Brexit had a significant short-run effect on consumer prices. The long-run impacts of
Brexit, on the other hand, are beyond the scope of this article and are yet to be determined.

REFERENCES

Amiti, M and J Konings (2007), “Trade Liberalization, Intermediate Inputs, and

Productivity: Evidence from Indonesia”, American Economics Review 97(5): 1611-1638.

Ayele, Y, G Larbalestier and N Tamberi (2021), “Post-Brexit: Trade in goods and services
(II)”, UK Trade Policy Observatory Briefing Paper No. 63.

Bakker, J D, N Datta, J De Lyon, L Opitz and D Yang (2022), Post-Brexit Imports, Supply
Chains, and the Effect on Consumer Prices, UK in a Changing Europe, 27 April.

Breinlich, H, E Leromain, D Novy, T Sampson and A Usman (2018), “The Economic

Effects of Brexit: Evidence from the Stock Market”, Fiscal Studies 39: 581– 623.

Costa, R, S Dhingra and S Machin (2019), “Trade and worker deskilling”, CEP Discussion
Paper 1622, Centre for Economic Performance, LSE.

Datt, M, B Hoekman and M Malouche (2011), “Taking Stock of Trade Protectionism Since
2008”, Economic Premise No. 72, World Bank.
Freeman, R, K Manova, T Prayer and T Sampson (2022), “Unravelling deep integration:
UK trade in the wake of Brexit”, CEP Discussion Paper 1847, Centre for Economic
Performance, LSE.

Jaravel, X and E Sager (2019), “What are the Price Effects of Trade? Evidence from the US
45
and Implications for Quantitative Trade Models”, CEP Discussion Paper 1642, Centre for
Economic Performance, LSE.

POST-BREXIT IMPORTS, SUPPLY CHAINS AND THE EFFECT ON CONSUMER PRICES | BAKKER, DATTA, DE LYON, OPITZ AND YANG
Johnson, R C and G Noguera (2012), “Accounting for intermediates: Production sharing
and trade in value added”, Journal of International Economics 86(2): 224-236.

Karlsson, T (2011), “Imports, Product Variety and the Extensive Margin: Some Stylized
Facts”, EcoMod2011 3003.

ABOUT THE AUTHORS

Jan David Bakker is an Assistant Professor at Bocconi University and a Research

Associate at the Centre for Economic Performance. Prior to joining Bocconi, he was a
Post-Doctoral Research Fellow at University College London and obtained his DPhil/Phd
from the University of Oxford.

Nikhil Datta is a final year PhD candidate at University College London and a Researcher
at the Centre for Economic Performance at LSE. From August 2022, he will be an
Assistant Professor at the University of Warwick

Josh De Lyon is a final year PhD candidate at the University of Oxford, a Researcher at
the Centre for Economic Performance at LSE, and a Research Fellow at the Institute for
the Future of Work. He completed his MPhil in Economics at University of Oxford and
his BSc in Economics at University of Bristol. In summer 2022, he will join the OECD in
the Productivity, Innovation, and Entrepreneurship division.

Luisa Opitz is a Research Assistant at the Centre for Economic Performance at LSE. She
holds an MSc in Economics from LSE, and in August 2022 she will start her PhD at
Bocconi University.

Dilan Yang is a Research Assistant at the Centre for Economics Performance at LSE. She
holds a Bachelor’s in Economics from University of Cambridge and was a Schwarzman
Scholar at Tsinghua University. She is currently reading MPhil in Economics at the
University of Oxford.
CHAPTER 5
The price impacts of trade agreements 47

Meredith A. Crowley, Lu Han, and Thomas Prayer

THE PRICE IMPACTS OF TRADE AGREEMENTS | CROWLEY, HAN AND PRAYER

University of Cambridge and CEPR; University of Liverpool Management School and CEPR;
University of Cambridge

The United Kingdom’s withdrawal from the European Union and subsequent entry
into the Trade and Cooperation Agreement (TCA) represents a fundamental shift in the
UK’s trade relationship with its largest and most important international partner. With
inflation rising steeply in 2022, many have asked whether recent price increases can be
attributed to Britain’s departure from the European Union’s Single Market. Although
there are no tariffs on merchandise imported from the EU, non-tariff barriers ranging
from onerous customs paperwork to establishing compliance with rules of origin have
raised the cost of doing business in the UK for EU firms. In this chapter, we examine how
changes in market access under a preferential trade agreement impact the market power
of exporting firms and the prices they charge.

To gain a better understanding of what is happening in the UK today, we examine the

economic impacts of the complex, overlapping network of preferential trade agreements
(PTAs) that have come to criss-cross the world over the last 40 years. Economists have
employed a range of different methodologies to estimate the trade and welfare effects
of these policies. One popular method, the estimation of structural gravity models, has
shown that PTAs can increase trade by 44% (Limao 2016) to 58% (Baier and Bergstrand
2007). An alternative methodology, popular among government ministries for ex ante
analysis of potential PTAs, is the computable general equilibrium (CGE) model. In
the wake of the Brexit referendum, CGE modelling by the UK government found that
the losses to the UK’s GDP from withdrawing from the EU ranged between 0.6% and
7.7%, with smaller losses associated with scenarios in which the UK maintains a closer
relationship with the EU (HM Government 2018).

Both gravity and CGE models, however, overlook the impact that PTAs have on the
nature and extent of competition among firms. If a PTA induces more firms to start
exporting and so increases competitive pressures in a market, it can reduce the market
power of incumbent exporters and bring about price reductions for consumers. This
is a pro-competitive effect of PTAs that leads to welfare gains beyond those estimated
from structural CGE models. While there is evidence on the competitive effects of trade
liberalisations on domestic firms’ prices and markups (Konings and Vandenbussche
2005, Amiti and Konings 2007, Pierce 2011, Edmond et al. 2015, De Loecker et al. 2016)
 and average prices per unit (Bown and Crowley 2006, Amiti et al. 2019, Fajgelbaum et al.
2020), there is almost no existing work on how the prices and markups of exporting firms
change under a PTA.

48
EMPIRICAL FINDINGS ON PRICE-COST MARKUPS OF EXPORTERS

In new work (Crowley et al. 2022), we examine the pro-competitive gains from trade
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

from a perspective emphasising foreign exporting firms and how the structure of
market competition changes under a PTA. Empirically, we integrate information
from 257 preferential trade agreements from the World Bank’s Deep Trade Agreement
database with the universe of customs transactions from eleven low- and middle-income
economies to examine firms’ entry into foreign markets, market shares, export prices
and markups. We find that preferential trade agreements have pro-competitive effects on
foreign producers’ prices and markups. A PTA which reduces tariffs by 10% leads to a 22%
increase in the number of exporting firms, and a 4% reduction in the markups charged
by exporters. For highly differentiated consumption goods, the magnitude of the markup
reduction increases to 10%.

A NEW MODEL OF TRADE WITH PRO-COMPETITIVE IMPACTS ON PRICES

Based on these empirical findings, we develop a new model of multi-country trade in

which price-cost markups vary according to the market structure and the intensity of
competition among firms from different countries. In this model, changes in a firm’s
price-cost markup induced by a trade liberalisation depend on a firm’s elasticity of
demand – i.e. how much consumers’ demand for a firm’s output will change when the
firm changes its price by 1%. Building on Atkeson and Burstein’s (2008) seminal work on
international pricing in oligopolistic markets, we show that an exporting firm’s elasticity
of demand depends on the market structure in the destination, which can be summarised
by two different market share measures. The ‘within-origin’ market share captures the
share of an individual exporting firm relative to the total value of exports from the origin
to the destination. The ‘cross-origin’ market share captures the country’s market share in
a destination relative to the destination’s total imports from all origins.

When countries sign a PTA to reduce tariffs or non-tariff barriers to trade, this leads
more firms from the PTA’s origins to start exporting to the PTA’s destinations. This, in
turn, causes the average within-origin market share to fall. This means each exporting
firm will have less market power relative to competing firms from the same origin
country, creating a force that tends to depress price-cost markups. At the same time,
an origin country which signs a PTA will enjoy greater market access and gain cross-
origin market share relative to non-PTA countries. This implies more market power for
the origin country as a whole, which tends to allow exporting firms from this origin to
raise price-cost markups.
With these two offsetting forces pushing prices in opposite directions, the net effect
ultimately depends on three parameters that capture the substitutability of goods (i)
across firms within an origin and product, (ii) across origins within a product, and
(iii) across products. Crucially, for realistic values of key model parameters, we find
that preferential trade liberalisations lead to intense entry that generates a significant 49
loss of market power for individual firms relative to other firms from the same origin.
The within-origin reallocation effect dominates the cross-origin effect, resulting in a

THE PRICE IMPACTS OF TRADE AGREEMENTS | CROWLEY, HAN AND PRAYER

reduction in price-cost markups.

THE IMPACT OF SIGNING A PREFERENTIAL TRADE AGREEMENT ON A

COUNTRY’S MARKET SHARE

We use our newly developed model to examine how tariff liberalisations under a PTA
impact market structure and price-cost markups. To illustrate how our quantitative
model works, we present results from an experiment based on a world with five symmetric
countries, which have identical distributions of costs and technology. Each country
produces 4,000 distinct goods with about ten domestic firms operating in each goods
market, each making slightly different varieties of the good. Initially, all five countries
set tariffs of 10% for all goods. We consider the impacts of a simple policy experiment
in which two of the five countries sign a PTA that removes all import tariffs on trade
with the other country, but leaves tariffs against and among the other three countries
unchanged. For concreteness, we will denote the PTA partner countries as 1 and 2 and
the non-PTA countries as 3, 4 and 5.

DISTRIBUTION OF FIRMS’ WITHIN-ORIGIN MARKET SHARES

Figure 1 depicts the market shares of firms from each of the five countries in country two’s
domestic market. The left-hand pie chart shows the cross-origin market shares before
the PTA and the chart on the right-hand side shows the cross-origin market shares after
the PTA. We see that the preferential tariff liberalisation between countries 1 and 2 led
to an increase in country 1’s market share in country 2, from 15.9% to 18.7%. This gain in
market share comes at the expense of competing exporters in countries 3, 4 and 5 as well
as country 2’s own domestic firms.
 FIGURE 1 AGGREGATE MARKET SHARES IN COUNTRY 2

Before After

50
15.9% 15.9% 15.2% 18.7%
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

15.8% 15.4%

36.4%
15.3% 35.4%
15.9%

Origin:
1 2 3 4 5

In Figure 2, we present two graphs of changes to the within-origin market shares of firms
from country 1 that export to country 2. Beginning with the top panel, which depicts
changes in within-origin market shares under free entry, the horizontal axis depicts the
market shares of individual firms from country 1 exporting to country 2 and the vertical
axis measures the frequency with which firms fall into each market share bin. There are
two distinct empirical distributions presented in this chart: the blue distribution depicts
within origin market shares before the PTA is signed, while the red distribution depicts
the same variable under the PTA. The places where the two frequency distributions
overlap are rendered in grey. On the right side of the chart, the mass around 1 tells us
that there are a few products where only one firm from country 1 is exporting. Most of the
mass in both distributions lies between 30% and 60%. This could arise, for example, for
products where two firms from country 1 export to country 2, but also for cases with one
larger firm and two smaller ones.

Overall, we can see that the red distribution is shifted to the left of the blue. This tells us
that after the PTA induced tariff liberalisation, firms’ market shares within an origin have
fallen. This is mainly driven by the entry of new firms from origin 1. In our simulated data,
we find the number of firms exporting from country 1 to country 2 has risen by more than
10%. This shift is summarised by the mean market share of each distribution, highlighted
by the red and blue dashed lines, which falls from 44% before the tariff liberalisation to
about 40% after the tariff liberalisation.

The bottom panel of Figure 2 shows the within-origin market shares from a different
policy experiment in which, under the PTA, tariffs between countries 1 and 2 fall, but
market participation by firms everywhere is fixed. That is, the analysis is conducted under
an extreme assumption that no firms enter or exit the market in any country in response
to the new PTA. We can see there is virtually no change in the within-origin market
shares under this assumption. This extreme and unrealistic policy experiment is useful
in that it allows us to see the important role of firms’ entry decisions and consequent
changes in market structure on firms’ markups, shown in Figure 3.
51

FIGURE 2 DISTRIBUTION OF FIRMS’ WITHIN-ORIGIN MARKET SHARES OVER 4,000

THE PRICE IMPACTS OF TRADE AGREEMENTS | CROWLEY, HAN AND PRAYER

PRODUCTS

3500
Before
3000 After
2500
Frequency

2000

1500

1000

500

0
0 .2 .4 .6 .8 1

3500
Before
3000 After
2500
Frequency

2000

1500

1000

500

0
0 .2 .4 .6 .8 1

HOW PRICE-COST MARKUPS CHANGE WITH CHANGES TO MARKET

STRUCTURE

The final figure in our chapter presents two graphs of markups charged by firms located
in country 1 that export to country 2. The top panel presents markups before and after the
introduction of the PTA under free entry, while the bottom panel depicts markups when
firm participation is fixed. The horizontal axis measures the markup as a ratio relative to
production costs, so that a value of 1.9 indicates a price-cost markup of 90%. The vertical
axis measures the frequency of exporting firm observations at each level.
 FIGURE 3 PRICE-COST MARKUPS OF COUNTRY 1 FIRMS SELLING IN COUNTRY 2

5,000
Before
After
4,000
52
Frequency 3,000
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

2,000

1,000

0
1.4 1.5 1.6 1.7 1.8 1.9

5,000
Before
After
4,000
Frequency

3,000

2,000

1,000

0
1.4 1.5 1.6 1.7 1.8 1.9

As in Figure 2, the markup distribution before the tariff liberalisation is shown in blue
and after the tariff liberalisation in red. Signing the PTA causes the markup distribution
to shift to the left, suggesting that country 1’s exporters lower their markups in response
to a bilateral tariff liberalisation. The mean markup has decreased from 54% before the
liberalisation (blue dashed line) to 52% after the liberalisation (red dashed line).

Finally, the bottom panel shows how markups would have adjusted under the PTA if
firms were not allowed to enter or exit in response to the new conditions of the trade
agreement. The two distributions almost completely overlap and the average change in
the markup is almost non-existent. The central insight from this panel is that entry and
exit are important aspects of firm responses to PTAs.

IMPLICATIONS FOR BRITAIN

The PTAs examined in our study reduced tariffs and non-tariff impediments to trade.
Because the TCA does the opposite, we might expect that the new hurdles it has introduced
could have an anti-competitive or price-increasing effect. Existing evidence shows that
the outcome of the Brexit referendum discouraged entry into exporting between the UK
and EU (Crowley et al. 2020) and between Portugal and the UK (Fernandes and Winters
2021), and that the EU stopped exporting many products to the UK altogether after the
introduction of the TCA (see the chapter by Rebecca Freeman, Kalina Manova, Thomas
Prayer and Thomas Sampson in this book). Together, these findings hint at the possibility 53
of rising market power and price-cost markups among EU firms that continue to export
to the UK. Of course, a full assessment will have to wait until firm-level data from 2021

THE PRICE IMPACTS OF TRADE AGREEMENTS | CROWLEY, HAN AND PRAYER

becomes available to researchers. However, at present, the risk of an anti-competitive rise
in UK prices brought about by Brexit remains a serious concern.

REFERENCES

Amiti, M and J Konings (2007), “Trade Liberalization, Intermediate Inputs, and

Productivity: Evidence from Indonesia”, American Economic Review 97(5): 1611-1638.

Amiti, M, S J Redding and D E. Weinstein (2019), “The Impact of the 2018 Tariffs on
Prices and Welfare”, Journal of Economic Perspectives 33(4): 187-210.

Atkeson, A and A Burstein (2008), “Pricing-to-Market, Trade Costs, and International

Relative Prices”, The American Economic Review 98(5): 1998-2031.

Baier, S L and J H Bergstrand (2007), “Do free trade agreements actually increase
members’ international trade?”, Journal of International Economics 71(1): 72 - 95.

Bown, C P and M A Crowley (2006), “Policy Externalities: How US Antidumping Affects

Japanese Exports to the EU”, European Journal of Political Economy 22(1): 696-714.

Crowley, M A, O Exton and L Han (2020) “The looming threat of tariff hikes: entry into
exporting under trade agreement renegotiation”, American Economic Association Papers
and Proceedings 110: 457-551.

Crowley, M A, L Han and T Prayer (2022), “The Pro-competitive Effects of Trade

Agreements”, mimeo, University of Cambridge.

De Loecker, J, P K Goldberg, A K Khandelwal and N Pavcnik. (2016), “Prices, Markups,

and Trade Reform”, Econometrica 84(2): 445-510.

Edmond, C, V Midrigan and D Y Xu (2015), “Competition, markups, and the gains from
international trade”, American Economic Review 105(10): 3183-3221.

Fajgelbaum, P D, P K Goldberg, P J Kennedy and A K Khandelwal (2020), “The Return to

Protectionism”, The Quarterly Journal of Economics 135(1): 1-55.

Fernandes, A P and L A Winters (2021), “Exporters and shocks: The impact of the Brexit
vote shock on bilateral exports to the UK”, Journal of International Economics 131,
103489.
 Freeman, R, KL Manova, T Prayer and T Sampson (2022), “The impact of Brexit on UK–
EU trade”, in The Economics of Brexit: What Have We Learned?, CEPR Press.

HM Government (2018), EU Exit: Long-Term Economic Analysis, November.

54 Konings, J and H Vandenbussche (2005), “Antidumping protection and markups of

domestic firms”, Journal of International Economics 65(1): 151 - 165.
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

Limao, N (2016), “Preferential Trade Agreements”, Chapter 6 in K Bagwell and R W

Staiger (eds), Handbook of Commercial Policy, Vol.1, North-Holland.

Pierce, J R (2011), “Plant-level responses to antidumping duties: Evidence from U.S.

manufacturers”, Journal of International Economics 85(2): 222 - 233.

ABOUT THE AUTHORS

Meredith A. Crowley is a Professor of International Economics and Fellow of St. John’s

College at the University of Cambridge. She is a Senior Fellow at the UK in a Changing
Europe (UKCE) and a Research Fellow at the Centre for Economic Policy Research (CEPR
- London). Her research, which focuses on international trade, has been published in
leading journals, including the American Economic Review. Prior to arriving at Cambridge
in 2013, Crowley worked at the Federal Reserve Bank of Chicago. She received her MPP
from Harvard University and her PhD in Economics from the University of Wisconsin-
Madison.

Lu Han is a Lecturer at the University of Liverpool Management School and a Research

Affiliate at the Centre for Economic Policy Research (CEPR - London). He obtained
his PhD in Economics from the University of Cambridge. Prior to joining Liverpool, he
worked as a Postdoctoral Research Associate in the Faculty of Economics at Cambridge.
He is a macroeconomist with research interests in international economics, open
macroeconomics, firms and trade.

Thomas Prayer is a PhD student at the University of Cambridge and an occasional

research assistant in CEP’s trade programme. His research interests lie in international
trade, trade policy, firm dynamics, and competition.
CHAPTER 6
Brexit and UK services trade 55

Jun Du and Oleksandr Shepotylo

BREXIT AND UK SERVICES TRADE | DU AND SHEPOTYLO

Aston Business School

It is impossible to ignore services trade when assessing the costs of Brexit for two
reasons. First, the UK is a superpower in services production, and trade and services are
too important to its economy to be overlooked. As the second largest service market in
the world and the largest in Europe, the UK’s services sectors represent 80% of all UK
businesses and contribute to around 80% of the UK’s total jobs, gross output, and value-
added (Douch et al. 2020). The UK’s services sectors rely heavily on services exports, with
the EU being the main customer (Hall et al. 2020). Financial, professional and business
services, and ‘other business services’ (including research and development, professional
and management consulting services, technical and trade-related services) generate a
significant trade surplus for the UK, compensating for the large trade deficit in goods.

While financial services are concentrated in London, all UK regions export various
types of services, more than half of which are supplied by small businesses (see Figure 1).
Importantly, negative shocks to services firms spill over to the wider economy. Depressed
business dynamism and eroded household income harm non-exporters as well as
exporters in all sectors and all regions. Moreover, weakened services sectors have adverse
effect on the networks of services and manufacturing sectors, whose competitiveness
increasingly rely on servitisation.

Second, Brexit meant the UK’s departure from the world’s deepest trade agreement –
an agreement which had led to much deeper integration not only of trade in goods, but
especially of trade in services. Unlike trade in goods, services trade has no clear fallback
position from deep integration: restrictions to services trade are typically not determined
by tariffs, but rather by qualifications, permissions, recognition of standards and other
non-tariff barriers that may eradicate, rather than reduce, trade in an extreme scenario
(Lawless 2019). This means some services might be decimated rather than just disrupted.
Given this, it was surprising that the UK’s strength in services was not reflected in the
government’s ambitions for the sector in the EU–UK trade negotiations that followed the
referendum (Lowe 2020).
 FIGURE 1 UK SERVICE EXPORTS: REGIONS, SECTORS, EXPORTER SIZE AND TRADING
PARTNERS, 2011–17 (£ MILLION)

56
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

Source: Authors’ compilation based on Office of National Statistics, International Trade in Services (ITIS), 2011-2017.
Notes: There are eight different types of services traded by firms (Professional, R&D, Financial, Telecommunications,
Construction, Cultural and Trade Related Services); there are five destinations for services trade (EU, USA, Other European
Countries, BRICS, and the Rest of the World); the dataset has three broad industry categories (Manufacturing, Professional
and Business Services, and Other Service industries); and services firms are located in the regions in the UK classified by
NUTS1. Total export and import statistics are in millions of pounds. The overall export value captured by the ITIS over 2011-
17 is £237. SMEs refer to businesses with fewer than 250 employees.

Since the 2016 referendum, the uncertainty over these trade negotiations damaged the
UK economy as a whole (Bank of England 2019: 38–43), reduced firms’ exports of goods
(Crowley et al. 2018) as well as aggregate trade flows (Graziano et al. 2021), and led to
trade redirection away from the close, rich and previously frictionless neighbouring EU
markets to places further afield (Douch et al. 2019). However, for services trade, there is
only limited evidence of the impact.

UK TRADE IN SERVICES

Using the harmonised OECD-WTO Balanced Trade in Services dataset (BaTIS), we

observe that the UK’s growth in services trade was on a par with major exporters such as
Germany, Spain and the US up to 2016. Over time, the UK appears to have lost headway,
most apparently after 2016. The UK had been considerably outperformed by Ireland and
the Netherlands since 2013 in the global competition for services trade expansion, and
the differences in services export growth between Ireland and the UK have only been
amplified and reinforced since 2016. By the end of 2019, Irish services exports were 68.6%
higher than in 2014, while the UK’s services exports had grown by only 5.8%. The UK has
also been outperformed by Germany, the US and the Netherlands. What might explain
this stunning success of Ireland as a services exporter compared to the rather moderate
performance of the UK, especially after 2015? A reasonable hypothesis is that the UK’s
EU referendum might have triggered the diversion of services from the UK to Ireland. 57

BREXIT AND UK SERVICES TRADE | DU AND SHEPOTYLO

FIGURE 2 INTERNATIONAL COMPARISON OF UK SERVICES EXPORTS
Export of services index, 1 = Export in 2012

2
Germany

France

Ireland

Spain
1.5

The Netherlands

UK

US

1
12

14

16

18

20
20

20

20

20

20

Year
Source: OECD BATIS 2019.
Source: OECD Balanced Trade in Services dataset 2019.

THE BREXIT EFFECT ON TRADE IN SERVICES

To test this hypothesis, our recent paper (Du and Oleksandr 2021b) analyses the causal
effect of the Brexit referendum on UK services trade between 2016 and 2019. We adopt a
novel method using a synthetic difference-in-differences (SDID) estimator (Archangelskiy
et al. 2019) to construct a counterfactual for the UK had it not voted Leave in 2016. The
assessment of the Brexit effect is then done by comparing the actual performance of the
UK with the modelled performance of this synthetic UK, which looks much like the UK
but did not vote to leave the EU.

As shown in Figure 3, we find that the referendum has caused a large negative effect on
the UK’s services trade. Exiting the EU has resulted in the UK experiencing an average
shortfall of £18.5 billion worth of services exports every year between 2016 and 2019
relative to had the UK remained in the EU. This is 5.7% lower than in 2019 and is a
more conservative estimate compared to the existing estimates using different data and
methodologies of between 7% and 9.2% (Douch and Edwards 2021, Du and Shepotylo
2021a).
 58
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

FIGURE 3 BREXIT EFFECT ON EXPORTS IN SERVICES, BY EU COUNTRIES AND SECTORS (BILLION USD)

Note: This is a simplified table for the SDID estimates as sectoral level for each country reported. The estimated coefficients reported in the table are statistically significant at p>0.05. Those not
statistically significant are replaced by zeros for visualisation purposes but do not necessarily mean zero effect.
This finding could be attributed to a deterrent effect and the anticipated trade costs of
goods (Crowley et al. 2008, Graziano et al. 2018, Douch et al. 2020, Douch and Edwards
2021), or other channels such as FDI relocation (Breinlich et al. 2020) and more generally
GDP effects (Born et al 2019), as well as shifts in consumer sentiment from EU towards
non-EU products and services (Douch and Edwards 2021). 59

There are large variations in impact across sectors. A statistically significant decline

BREXIT AND UK SERVICES TRADE | DU AND SHEPOTYLO

following the vote for Brexit is seen for exports in “Transport’, “Travel”, “Insurance” and
“Telecom” services in the run up to 2020. However, we did not find a significant decline
for “Business Services” and “Intellectual Property Services”, or in the “Financial” and
“Cultural” sectors.

At the same time, Ireland benefited significantly during this period by expanding its
services sector exports. We estimate a resulting growth in total services exports of £24
billion annually over 2016–19 in Ireland compared to a counterfactual scenario where
Brexit did not occur. This translates to an impressive 14.75% annual growth over the
period of 2016-2019 compared to Ireland’s 2019 total services exports level. This growth
is particularly concentrated in the “Telecoms”, “Business”, “Intellectual Property” and
“Insurance” sectors.

Ireland’s low corporate tax rates for R&D/intangibles encourages knowledge-intensive

multinationals. Its global connectivity and open society, and the fact that it is now the
only predominantly English-speaking country in the EU, mean that Ireland is attractive
to skills and talents migrating from the UK. A likely explanation for Ireland’s beneficial
effect of Brexit is that it has become an even more favourable location for investment for
businesses looking for frictionless access to the EU’s Single Market.

There are other winners besides Ireland in some service areas. The Netherlands has
increased its “Business” and “Intellectual property” exports considerably, Spain has
seen growth in “Travel and transport” services exports, while Germany has gained in
“Transport”, “Insurance”, “Telecom’ and “Intellectual property” services exports. Ireland
seems to have done exceptionally well in relation to the export of “Telecom” services, in
sharp contrast to the lost exports in this sector not just from the UK but also from the
Netherlands, Switzerland and France.

POLICY IMPLICATIONS

The economic de-integration resulting from the UK’s exit from the EU has led to the
erection of new trade barriers between major services-providing countries/regions. The
finding of a negative effect of Brexit on UK trade in services as a whole, and in particular in
the sectors that enjoyed deep trade liberalisation with the EU, is not unexpected. However,
this chapter only discusses the anticipation and deterrent effect of the uncertainty felt by
businesses up to the end of 2019. It will take some time for the full impact of Brexit on UK
services to emerge.
 Since 2019, trading conditions and market access for UK services in the EU have only
worsened with entering into the Trade and Cooperation Agreement, since the agreement
has a very limited coverage of trade in services. Freedom of movement has ended, which has
created hard barriers for many sectors. Trade costs are expected to rise, given regulatory
60 and other discriminatory costs of trading services with the EU. Some firms specialised
in certain service types and many small traders are expected to exit the export market.
Thus, the trade eco-system has been disturbed and a restructuring of new networks may
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

be underway. We are yet to understand the implications of this.

Further, there is still high uncertainty surrounding the future UK–EU services trade
relationship. The financial sector may not obtain equivalence to allow transactions to flow
freely. Digital data flows between the UK and EU, and digital flows between the UK and
third non-EU countries with EU businesses as related parties, could still be restricted in
the future. Embedding a provision on free data flow in future trade agreements would
encourage investment flows. Stability, transparency and regulatory consistency in
financial markets could be challenged. Engaging in regulatory liberalisation to enhance
the UK’s competitiveness as an investment destination and as a services provider will
therefore be crucial.

On the other hand, new opportunities might surface. Continued trade negotiations and
dialogues regarding trade liberalisation are essential with both the EU and large, fast-
growing markets beyond Europe. However, it is important to remember that, in general,
liberalising services trade more challenging than liberalising trade in goods. It is extremely
hard, if not impossible, to foresee future preferential trade agreements achieving new and
better market access in a significant way (Lowe 2021). Even for services trade, the gravity
rule, according to which trade declines rapidly with distance, still applies (Springford and
Lowe 2018). Preserving goodwill, rebuilding trust and keeping the dialogues open with
its nearest and the largest trading partners should clearly be the most important items on
the agenda for the UK to reduce barriers and promote mutual benefits in future trading
relationships.

Crucial to understanding these impacts are reliable data and rigorous analysis, which
have always been a challenge for services trade. It will take even longer for economists
to measure the full impact of transitioning to the new UK–EU trade relationship with
reliable services trade data, but at least the work has begun.

REFERENCES

Arkhangelsky, D, S Athey, D A Hirshberg, G W Imbens and S Wager (2019), “Synthetic

difference in differences”, NBER Working Paper No. 25532.

Bank of England (2019), Monetary Policy Report, November.

Breinlich, H, E Leromain, D Novy and T Sampson (2020), “Voting with their money:
Brexit and outward investment by UK firms”, European Economic Review 124: 103400.
Crowley, M, O Exton and L Han (2018), “Renegotiation of trade agreements and firm
exporting decisions: evidence from the impact of Brexit on UK exports”, Society of
International Economic Law (SIEL), Sixth Biennial Global Conference.

Douch, M and T H Edwards (2021), “The Brexit policy shock: Were UK services exports
61
affected, and when?”, Journal of Economic Behavior & Organization 182: 248-263.

Douch, M, J Du and E Vanino (2019), “Defying Gravity? Policy Uncertainty and Trade

BREXIT AND UK SERVICES TRADE | DU AND SHEPOTYLO

Diversion”, Lloyds Banking Group Centre for Business Prosperity Research Paper.

Douch, M, J Du, U Nduka and O Shepotylo (2020), “Ten facts about the UK professional
and business services sectors and their international traders”, Lloyds Banking Group
Centre for Business Prosperity Insight Paper.

Du, J and O Shepotylo (2021a), “Feeding the Celtic Tiger: Brexit, Ireland and Services
Trade”, Lloyds Banking Group Centre for Business Prosperity Research Paper.

Du, J and O Shepotylo (2021b), “Brexit and Services Trade New Evidence from Synthetic
Diff-in-Diff Approach”, UK in a Changing Europe Working Paper 08/2021.

Graziano, A G, K Handley and N Limão (2021), “Brexit uncertainty and trade

disintegration”, The Economic Journal 131(635): 1150-1185.

Lawless, M and E L Morgenroth (2019), “The product and sector level impact of a hard
Brexit across the EU”, Contemporary Social Science 14(2): 189-207.

Lowe, S (2020), “British services have played second filled in the Brexit debate”, UK in a
Changing Europe Commentary, 11 February.

Lowe, S (2021), “Keeping up appearances, what now for UK services trade?”, Centre for
European Reform, February.

Springford, J and S Lowe (2018), “Britain’s services firms can’t defy gravity, alas”, Centre
for European Reform Insight.

ABOUT THE AUTHORS

Jun Du is Professor of Economics at Aston Business School and Founding Director of

the Centre for Business Prosperity (CBP). Jun leads research on international trade
and productivity of the CBP and research on internationalisation of UK Enterprise
Research Centre (ERC). Her recent research focuses on how trade policy and firms’
internationalisation strategies affect their competitiveness and wider economic impact
in the economy. She tracks international trade dynamics and assesses how global value
chains impact on firms’ internationalisation strategy and performance.

Jun has a wide research interest and expertise on trade, investment, innovation and
productivity and her research has been supported by the Economic and Social Research
Council (ESRC), Leverhulme Foundation, NESTA, UK national and local governments,
 and private sectors. Jun is member of Innovation Caucus, an expert in UK Department
for Business, Energy & Industrial Strategy (BEIS) External Peer Review Group, and
ESRC Grant Assessment Panel member (Economics and Management).

She holds a PhD from theUniversity of Leicester and an MSc in Economics, Banking
62
and Finance from Cardiff University. Jun worked as a junior manager leading research
programs in Industrial and Commercial Bank of China prior to coming to the UK.
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

Oleksandr Shepotyl is an Associate Professor at Aston Business School. He received his

PhD in Economics from the University of Maryland, College Park and his Master’s in
Economics from the New Economic School (Russia). He previously worked at the World
Bank, DEC RG Trade group, Kyiv School of Economics, Higher School of Economics, and
the University of Bradford.

Oleksandr’s main research interests are in international trade, trade policy, and
globalisation. He has worked in a variety of topics such as the impact of trade liberalization
on exports and productivity and the use of the structural gravity model to evaluate the
gains from trade. He has extensive experience measuring and quantifying non-tariff
measures and analysing trade policy impacts, advising governments about the effects
of trade policy decisions in developing economies; more than 15 years of professional
experience in statistical analysis of trade, both of WTO accession and regional trade
agreements.
CHAPTER 7
The economics of UK financial services 63

post Brexit

THE ECONOMICS OF UK FINANCIAL SERVICES POST BREXIT | HALL

Sarah Hall
University of Nottingham

As an EU member state, the UK developed significant EU exports in financial services. By

2019, the EU accounted for 40% of UK financial services exports (£24 billion) and 32% of
UK financial services imports (£6 billion). This contributed to a trade surplus in financial
services of £41 billion. EU–UK financial services trade at this time was underpinned by
passporting arrangements. Passporting allowed firms to sell their services into the EU
from their UK base without the need for additional regulatory clearances.

The UK financial services sector was split with regards to Brexit. Whilst the majority of
financial service providers, particularly in the areas of banking and insurance, generally
favoured remaining in the EU, other parts of the sector – notably hedge funds – identified
opportunities through domestic regulatory control within the UK that leaving the EU
would bring. This division has been identified as a contributory factor to the very limited
provisions agreed between the UK and the EU on financial services trade in the Trade
and Cooperation Agreement (TCA) and associated declarations (e.g. James and Quaglia
2019).

Without Single Market access through passporting, financial services trade between the
UK and the EU has become much more difficult than it was before Brexit due to the
increase in non-tariff barriers (Heneghan and Hall 2021, Lavery et al. 2018). However,
financial services had a longer lead-in to Brexit than some other sectors, and this allowed
preparations to be made by businesses to avoid a cliff-edge Brexit and associated systemic
financial instability.

Following the Political Declaration of October 2019, agreed by the UK and the EU
alongside the Withdrawal Agreement, it was clear that UK financial services would
lose their passporting rights into the Single Market. It was also clear that these would
not be replaced under the terms of the TCA and that, instead, Single Market access for
UK firms — and access to the UK market for EU firms — would be governed through
equivalence decisions.
 Equivalence is not decided through bilateral negotiation but by each party independently
deciding what access it will grant. Equivalence does not provide a full replacement for
passporting. Even if it is granted, equivalence as set out by the EU does not cover the full
range of financial services. Core banking services, such as lending, payments and deposit
64 taking, are excluded. Neither does equivalence grant permanent access rights; the EU
can withdraw equivalence determinations with 30-days’ notice.
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

Given that the UK was equivalent to the EU up until the point it left the Single Market,
it could have technically expected the EU to make positive equivalence determinations.
However, the EU has taken a restrictive approach, reflecting its concerns that the UK is
seeking to obtain competitive advantage through divergence from EU regulation in the
future. The EU only granted time-limited equivalence decisions for derivatives clearing
(initially for 18 months but now extended until the end of June 2025) and settling Irish
securities (for an initial six months and now expired). Both these decisions were driven by
the need to avoid short-term disruption in cross-border UK–EU financial services trade.

In contrast, the UK Chancellor, Rishi Sunak, announced in November 2020 that the UK
would pursue a more liberal approach, granting equivalence from 1 January 2021 to EEA-
based financial services in a range of areas (as shown in Figure 1). This reflects the UK’s
wider strategy of significantly liberalising inbound services trade. The UK has also made
clear that its approach to equivalence differs from that of the EU, implicitly criticising
the EU’s approach in the process. For example, as part of a broader emphasis on the
importance of transparency and outcomes within the UK’s approach to equivalence, the
Treasury has said that the withdrawal of equivalence would be considered only “as a last
resort” and would be accompanied by adaptation periods, rather than the 30 days’ notice
period of the EU approach.

FIGURE 1 UK AND EU EQUIVALENCE DECISIONS WITH SELECTED JURISDICTIONS AT

THE POINT OF UK EXIT

EU Equivalence Decisions: 27 areas UK Equivalence Decisions: 32 areas

UK 2 EEA 28

USA 21 USA 21

Singapore 15 Singapore 15

Switzerland 13 Switzerland 13

Source: European Commission (10/02/2021); HM Treasury (14/01/2021.

The limited number of equivalence decisions granted by the EU means that UK financial
services trade has less Single Market access than other leading international financial
centres such as New York and Singapore. The profoundly different equivalence regimes
being applied to the UK suggests that the EU’s approach to the UK may be motivated
more by concerns about developing its own competencies in financial markets than by 65
the merits of the UK’s domestic approach to regulation.

THE ECONOMICS OF UK FINANCIAL SERVICES POST BREXIT | HALL

The UK has implemented a Temporary Permissions Regime, currently scheduled to run
for three years until the end of 2023 although it is scheduled for review and possible
extension. This allows EEA firms and funds that were using a passport to access the
UK market during the transition period to continue to do so, provided they notified the
Financial Conduct Authority that they would use the permissions regime before the end
of transition. There is no equivalent EU-wide scheme for UK firms operating in the EU,
although some member states, such as Ireland and Denmark, have established temporary
permissions for UK firms that were passporting into particular parts of financial markets
for specific periods of time from 1 January 2021.

As it became clear during the trade negotiations that the UK’s passporting rights would
be withdrawn and replaced by limited equivalence decisions, financial services’ activity
began to relocate from London to European cities, notably Frankfurt, Amsterdam,
Luxembourg, Paris and Dublin, as well as other global centres such as New York.
Although official statistics do not track these relocations systematically, a range of other
information sources reveals the changes to date.

Latest figures indicate that between 7,000 and 7,500 jobs in financial services have left
the UK for the EU since 2016 involving relocations from within 440 firms.1 This is a small
proportion of the 1.1 million jobs in financial services across the UK. The figures also
do not account for new jobs created outside the UK that would have been in the UK
without Brexit. Bank assets valued at £900 billion have been relocated, equal to around
10% of the UK’s banking system, although this does not translate into a commensurate
relocation of economic activity (as measured by employment or value-added).

Instead, there is evidence of different specialisations emerging, with Frankfurt

dominating in banking and Dublin in asset management relocations. Amsterdam has
dominated in the relocation of activity that supports financial market transactions —
so-called market infrastructure. Amsterdam overtook London as the primary location
for European share trading in January and London’s derivatives platforms lost three-
quarters of their euro volumes to Amsterdam and New York.

1 Source: New Financial (2021) and EY Financial Services Brexit Tracker 2022 (www.ey.com/en_uk/news/2022/03/ey-
financial-services-brexit-tracker-movement-within-uk-financial-services-sector-stabilises-five-years-on-from-article-50-
trigger).
 Recent figures from the Office for National Statistics show that a realignment of UK
trade in financial services away from the EU to other geographical markets appears to
be underway (Figure 2, ONS 2021). Between 2Q2019 and 2 Q2021, non-EU exports of
financial services increased by £0.5 billion (5.1%) and imports by £0.01 billion (0.4%).
66 Over the same period, EU exports declined by £2.0 billion (-30.6%) and EU imports by
£0.3 billion (-21.3%). ONS analysis suggests that the decline in financial services trade
with the EU is being driven by declines in exports to France, Ireland, Germany and the
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

Netherlands.

FIGURE 2 EXPORTS OF EXPLICITLY CHARGED AND OTHER FINANCIAL SERVICES TO EU

AND NON-EU (2016Q1–2021Q2)

EU27 Non-EU

10,000
£ million

7,500 Start of UK End of

national transition
lockdown period

5,000

2,500

0
2016Q2 2016Q4 2017Q2 2017Q4 2018Q2 2018Q4 2019Q2 2019Q4 2020Q2 2020Q4 2021Q2

Source: ONS (2021).

The prospects of this trade realignment reversing appear limited. The TCA is accompanied
by a non-binding Joint Declaration committing the UK and the EU to cooperation on
matters of financial regulation. This is intended to be facilitated by a memorandum
of understanding (MoU). However, although the UK government announced that the
technical discussions underpinning the MoU had successfully concluded in March 2021,
no formal text has emerged.

An MoU is not unique to UK–EU trade in financial services. Similar commitments

are in place between the EU and the US and Japan, for example. However, a general
commitment to regulatory cooperation is no replacement for certainty regarding Single
Market access for UK financial services firms.

Further equivalence decisions from the EU do not appear likely. Despite the UK being
equivalent at the point of departure, the EU has paused the process until it is satisfied
regarding the operation of the Withdrawal Agreement and the Northern Ireland Protocol.
The EU is also keen to attract more financial services activity into the Single
Market, partly in order to prevent possible systemic risks associated with not
having core financial market activity that it relies on outside its own regulatory
border. This is particularly true in relation to clearing. Currently, the one area of
equivalence granted to the UK by the EU is in clearing, which is due to expire in 67
June 2025. This allows EU traders to use the more liquid clearing infrastructure
available in London. The EU granted this to prevent market instability at the

THE ECONOMICS OF UK FINANCIAL SERVICES POST BREXIT | HALL

end of transition, but it needs to develop its own capabilities in this area. In a sign that this
has proved difficult, the European Commission has extended this equivalence decision
until the end of June 2025. The Commission intends to use this time to develop plans to
develop central clearing activities within the EU by developing the competitiveness, cost
efficiency and liquidity of the EU as a clearing hub. The extent to which this is possible
remains uncertain given the market benefits of the liquidity and legal framework for
clearing in London, which makes clearing more cost effective through the so-called
netting of transactions.

Meanwhile, there are indications that the UK will diverge from EU financial services
regulation. In his Mansion House speech in the summer of 2021, Rishi Sunak appeared to
confirm that the UK intends to prioritise regulatory freedom over potentially enhanced
Single Market access for financial services. Further details of how such regulatory
opportunities will be exploited by the UK were set out in Sunak’s “New Chapter for
Financial Services” published in July 2021. This identifies both sectoral and geographical
foci for financial services post Brexit, with Sunak arguing that the UK should “push
for closer co-operation and more cross-border access with other like-minded financial
centres in markets around the world”.

In terms of sectoral foci, the UK Treasury is keen to cement the UK’s position as
a global FinTech hub and is also working to make the UK the main global centre for
sustainable finance. Work on FinTech has coalesced around the Kalifa Review that seeks
to support the FinTech sector through 15 recommendations across five policy areas:
policy and regulation; skills and talent; investment; international attractiveness and
competitiveness; and national connectivity. Recent data show that in 2021 the UK was
second to the US in terms of investment in FinTech. Concerns remain, however, about
the supply of highly skilled labour in this area, as FinTech was particularly reliant on EU
nationals.

The UK government is also eager to develop the UK’s leadership position in green
and sustainable finance. The government published its Green Finance Framework in
June 2021, setting out plans to finance expenditures on sustainable technologies and
infrastructure through the issuance of green gilts and the retail Green Savings Bond
launched in October 2021. The inaugural sale of two green gilts raised £16 billion in
October 2021. The government has also published its Greening Finance Roadmap, which
sets out the plans for economy-wide Sustainability Disclosure Requirements (SDR) in
 which firms across sectors will need to report on sustainability. It also includes plans for
a UK ‘green taxonomy’ to list all economic activities that UK regulators consider to be
environmentally sustainable.

At one level, this roadmap reflects the UK’s post-Brexit ambitions to use its new-found
68
domestic regulatory control to set international financial standards through engaging
with and shaping international regulatory architectures. The roadmap includes
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

provisions to align the UK’s new SDR with the international Task Force of Climate-
Related Financial Disclosures framework, which was developed at an international level
with significant input from the UK through the Financial Stability Board.

The green taxonomy is essentially a classification system that set out the criteria that
economic activities need to meet to be classified as environmentally sustainable. The
structure of the UK’s taxonomy will draw on the existing EU green taxonomy to which
the UK contributed as a member state, but will “take an approach that is suitable for the
UK market and consistent with UK government policy”. This signals the possibility of
divergence in approach and/or scope from the EU.

In terms of geographical reach, the UK has adopted a bilateral strategy

based around regulatory dialogue. So far, the UK has agreed a UK–Singapore
Financial Partnership aimed at using regulatory cooperation to increase bilateral
financial services trade. The UK has also announced that it is pursuing mutual
recognition of regulatory standards with Switzerland. This is an important set
of negotiations because, if agreed, it would facilitate financial services trade
liberalisation similar to that within the EU — including passporting between
the UK and Switzerland — but on a bilateral basis and without the political architecture
of the EU. As such, it is seen as a potential template for UK financial services agreements
elsewhere.

In terms of new trade deals, the UK–Japan Comprehensive Economic Partnership

Agreement (CEPA) signed in October 2020 contains some important financial services
elements that are not generally contained in trade agreements. For example, it prohibits
requirements that mandate the local storage of data held by firms (with some exceptions
for public purposes) and includes measures to facilitate the free transfer of these data
between the two countries. Data mobility is important in banking and payment businesses
that operate across multiple jurisdictions and is becoming increasingly important with
the growth of digital payments. The free trade agreement with Australia, agreed in June
2021, includes an annex aimed at fostering regulatory cooperation on financial services
regulation.

However, there are some policy areas, notably in relation to EU regulation on bankers’
bonuses, where the UK could now diverge from the EU but has chosen not to do so – a
reflection of the domestic political optics of not being seen to use a Brexit regulatory
dividend to enhance bankers’ salaries, while at the same time emphasising government
ambitions to ‘level up’ the UK economy.
In sum, a deregulatory race to the bottom, or a form of widespread deregulation as
happened under the Big Bang in the 1980s, seems unlikely in UK financial services.
Piecemeal regulatory changes will and are being made, particularly where they link to
other key policy objectives in relation to digital and green growth. Even so, it seems likely
that these changes will not be sufficient to offset lost financial services exports to the EU. 69
It will take some time to assess this outcome, but it does seem likely that London will be
a smaller, albeit still large, financial centre in the medium term.

THE ECONOMICS OF UK FINANCIAL SERVICES POST BREXIT | HALL

REFERENCES

Heneghan, M and S Hall (2021), “The emerging geography of European financial centres:
fragmentation in the European Union and concentration in the UK”, European Urban
and Regional Studies 28(1): 40-46.

James, S and L Quaglia (2019), “Brexit, the city and the contingent power of finance”, New
Political Economy 24(2): 258–271.

Lavery, S, S McDaniel and D Schmid (2018), “Finance fragmented? Frankfurt and Paris
as European financial centres after Brexit”, Journal of European Public Policy 26(10):
1502-1520.

New Financial (2021), Brexit and the City: the impact so far.

ONS – Office for National Statistics (2021), “The impacts of EU exit and coronavirus
(COVID-19) on UK trade in services: November 2021”.

ABOUT THE AUTHOR

Sarah Hall is a Professor of Economic Geography at the University of Nottingham and a

Senior Fellow with the UK in a Changing Europe. Her current work focuses on Brexit and
the UK’s financial services sector. She has previously worked on London’s international
financial district and its relations with Europe, China and North America. She is the
author of Global Finance (Sage, 2017) and Respatialising Finance (Wiley Blackwell 2021).
Her research has been covered my leading media outlets including The Financial Times,
the BBC, The New York Times, The Daily Telegraph and Wired magazine. She regularly
provides evidence and analysis to policymakers and industry bodies and blogs at UK in
a Changing Europe.
CHAPTER 8
The economics of the UK’s post-Brexit 71

immigration system

THE ECONOMICS OF THE UK’S POST-BREXIT IMMIGRATION SYSTEM | PORTES

Jonathan Portes
King’s College London

Immigration was central to the politics of Brexit (Hobolt 2016) but was peripheral in
the pre-referendum discussion of its economic consequences (Portes 2016). Indeed, both
before and in the immediate aftermath of the referendum, the UK’s choice was often
framed as a trade-off between the economic costs of increasing trade frictions between
the UK and EU on the one hand, and the political benefits of ending free movement and
restoring ‘control’ over immigration on the other (e.g. Baldwin 2016). Many policymakers
believed it would be clearly in the UK’s economic interests after Brexit to retain most or all
of the benefits of membership of the Single Market – either by maintaining membership
of the European Economic Area (like Norway) or via a series of bilateral agreements (like
Switzerland).

But the UK government, first under Theresa May and then Boris Johnson, rejected such
an approach, making clear ‘we are not leaving the European Union only to give up control
of immigration’. This position meant in turn that the EU never seriously considered what,
if any, compromises it could make on free movement. Instead, the EU underlined the fact
that free movement was an integral part of Single Market membership. In that sense,
the politics of immigration and free movement were the key driver of the shape of the
UK–EU post-Brexit economic and trading relationship, as now enshrined in the Trade
and Cooperation Agreement.

Since the referendum, immigration has become a much less salient political issue and
public attitudes towards immigration have become more positive (Runge 2019). However,
its economic significance has become more apparent, first as migration flows from the EU
fell sharply and then, over the course of the Covid-19 pandemic, as substantial numbers of
EU-origin workers returned to their countries of origin. Forte and Portes (2017) forecast
that net EU migration could fall by up to 150,000 over the period between 2016 and 2020.
These forecasts proved broadly accurate; net EU migration did indeed fall by slightly
more than 150,000 by the end of 2019.
 FIGURE 1 NET MIGRATION BY CITIZENSHIP, UK, YEAR ENDING MARCH 2010 TO YEAR
ENDING DECEMBER 2019

Estimate
Unadjusted estimate
72 Thousands
300
Non-
EU
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

200

100

EU
Migration adding to UK population
0
Migration deducting from UK population

British
EU referendum
-100 (June 2016)

YE Dec 10 YE Dec 11 YE Dec 12 YE Dec 13 YE Dec 14 YE Dec 15 YE Dec 16 YE Dec 17 YE Dec 18 YE Dec 19
provisional

Source: UK Office for National Statistics, 2020

At the same time, there was also a significant rise in non-EU migration, facilitated by
government policy, with the cap on Tier 2 visas for non-EU migrants (that is, relatively
skilled or highly paid workers) being relaxed. This marked the end of the Theresa May
era in immigration policy, during which the overriding objective had been to reduce
numbers. So, at the beginning of 2020, net migration to the UK remained high, although
the post-2004 trend for EU migration to partially displace non-EU migration had in part
been reversed. The onset of the Covid-19 pandemic, however, led to a very sharp reversal
of migration flows.

While the International Passenger Survey, which forms the basis for migration statistics,
has been suspended, the Labour Force Survey provides data on non-UK-born people
resident in the UK, while HMRC publishes statistics on employees. The latest published
data suggest a fall of about 200,000 in the EU-born resident population, driven by a fall
in the number of EU-origin workers. While these estimates are highly uncertain, there is
little doubt that there has indeed been a large outflow.

Given the nature of the pandemic and its economic and social impacts, this is not surprising.
Migrants, especially from Europe, are disproportionately likely to be employed in the
hospitality sector and other service sectors that require face-to-face contact, so were
more likely to have been furloughed or lose their jobs. With many universities moving
wholly or largely to online teaching, many foreign students decided not to come to the
UK or to return here.
It is against this background that the UK introduced the new, post-Brexit immigration
system, concurrently with the implementation of the Trade and Cooperation Agreement
in January 2021. Portes (2022) discusses the new system in detail, and outlines some of
the potential implications. As set out in Portes (2020), it was shaped by two broad forces.
The first was the government’s commitment to ending free movement and to a system 73
which would treat EU and non-EU migrants similarly. But second were significant positive
shifts in both public opinion and government policy towards immigration in general.

THE ECONOMICS OF THE UK’S POST-BREXIT IMMIGRATION SYSTEM | PORTES

The replacement of Theresa May with Boris Johnson, who had adopted relatively liberal
positions on immigration during his tenure as Mayor of London, signalled a change in the
relative priorities within government attached to the economic benefits of immigration
compared to the political need to be seen to be controlling it.

The policy intent of the new system was therefore less about reducing migration and more
about making it both more geographically diverse and more selective. With the end of
free movement, the new system applies to all those moving to the UK to work, apart from
Irish citizens. EU (and EEA/Swiss) nationals resident in the UK prior to January 2021
were eligible to apply to remain indefinitely under the ‘settled status’ scheme; well over 5
million have done so.

The key provisions of the new system are that:

• New migrants should be coming to work in a job paying more than £25,600 or
the lower quartile of the average salary, whichever is higher, and in an occupation
requiring skills equivalent to at least A-levels (‘RQF3’).

• There is a lower initial threshold for PhDs, new entrants, and for those in shortage
occupations, meaning that for some occupations the salary threshold may be as low
as about £20,000.

• Medical professionals, including doctors, nurses, and some technical staff, can apply
for a new, simpler and cheaper, Health and Care Worker visa; this was extended to
social care staff in early 2022.

• There is an expanded Seasonal Agricultural Workers Scheme, but no general

sectoral schemes for other workers who do not meet the skill threshold. However, a
number of small ad hoc visa schemes have since been introduced

• The visa provisions for students have not changed significantly. However, Brexit also
means that EU-origin students are no longer treated similarly to UK-origin ones
but rather to non-EU ones, meaning that they have to pay much higher fees.

• Moreover, a new ‘Graduate’ visa has been introduced, similar to the ‘post-study
work route’ visa abolished by Theresa May, which allows international students
graduating from UK universities to stay on and work in essentially any job, for up
to two years.
 The new system represents a very significant tightening of controls on EU migration
compared to free movement. Migrants coming to work in lower-skilled and paid
occupations are, in principle, no longer able to gain entry. Even those who do qualify will
need their prospective employers to apply on their behalf, will have to pay significant fees,
74 and will, as is the case for non-EU migrants at present, have significantly fewer rights –
for example, in respect of access to the benefit system.
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

However, compared to the current system, the new proposals represent a considerable
liberalisation for non-EU migrants, with lower salary and skill thresholds and no overall
cap on numbers. Approximately 68% of UK employees work in occupations requiring
RQF3 level skills or above. Given the requirement for new migrants to be paid at or above
the lower quartile of earnings for that occupation, that implies about half of all full-time
jobs would, in principle, qualify an applicant for a visa. This represents a very substantial
increase – perhaps a doubling compared to the previous system for non-EU nationals,
which was also for most of the 2012-19 period subject to an overall quota and a resident
labour market test. It also makes the new system considerably more liberal with respect to
non-European migrants than that of most EU member states, which typically apply much
more restrictive (de facto and/or de jure) skill or salary thresholds, and often enforce a
resident labour market test. The provisions for international students after completing
their studies are also relatively liberal.

So, the new system does not represent an unequivocal tightening of immigration controls;
rather, it rebalances the system from one which was essentially laissez-faire for Europeans
while quite restrictionist for non-Europeans, to a uniform system that, on paper at least,
has relatively simple and transparent criteria.

We now have a full year’s worth of data on the operation of the new system.1 While
interpretation is clouded by the continuing impact of the pandemic and the subsequent
reopening, a number of key points emerge:

• Work-related migration has recovered strongly. As with the broader economy –

where worries about persistent long-term unemployment have been replaced with
concern about skill and labour shortages – there is no evidence that the pandemic
has had any permanent negative impact on immigration. And the liberalisation of
policy towards non-EU citizens is already evident. The introduction of the Health
and Care Worker visa has had a big impact, with about 65,000 such visas issued in
2021. As before the pandemic, India is by far the largest single source country for
work visas, making up about 40% of the total and dominated by the ICT sector.
But there are big increases for Nigeria and the Philippines, probably reflecting
the Health and Care Worker visa; the Philippines is the largest source country for
nurses in the world.

1 https://www.gov.uk/government/statistics/immigration-statistics-year-ending-december-2021
FIGURE 2 WORK-RELATED VISAS GRANTED BY VISA TYPE, YEARS ENDING BY QUARTER
BETWEEN 2012 AND 2021
160
Work-related visas (thousands)

140
75
120

THE ECONOMICS OF THE UK’S POST-BREXIT IMMIGRATION SYSTEM | PORTES

100

80

60

40

20

0
Dec Jun Dec Jun Dec Jun Dec Jun Dec Jun Dec Jun Dec Jun Dec Jun Dec Jun Dec
12 13 13 14 14 15 15 16 16 17 17 18 18 19 19 20 20 21 21
Year ending
Skilled Temporary High value Other
Source: Home Office.

• Student visa numbers have increased to above 400,000, up about 50% on pre-
pandemic levels. Particularly large increases are visible for Indian, Pakistani,
and Nigerian students. The UK has clearly gained in relative attractiveness to
international students; in part, this may be because of restrictive measures imposed
by competing markets, such as Australia and the US. The new ‘Graduate’ route may
well be perceived as attractive, especially to those coming from the countries which
have seen the largest increase. So far, only about 12,500 visas have been issued under
this route, but the numbers are likely to grow.

FIGURE 3 SPONSORED STUDY VISA GRANTS, YEAR ENDING BY QUARTER, 2009–2021

500
450
400
Visas granted (thousands)

350
300
250
200
150
100
50
0
Dec Dec Dec Dec Dec Dec Dec Dec Dec Dec Dec Dec Dec
09 10 11 12 13 14 15 16 17 18 19 20 21
Year ending

Source: Home Office.

 • EU citizens make up only a little more than one in ten work visas, and 5% of
student visas. In part, this may be because many EU citizens coming here had
already acquired settled status based on a past period of residence here, and some
of those intending to move here in 2021 would have tried to get here in 2020 so as
76 to be eligible for settled status. The proportion of work and study visas going to EU
citizens is likely to rise in future.
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

TABLE 1 TOP FIVE NATIONALITIES GRANTED SPONSORED STUDY VISAS, 2029, 2020,
AND 2021

Percentage Percentage
Nationality 2019 2020 2021 change change
2019/2020 2020/2021

China 119,825 81,268 119,334 -<1% +47%

India 37,396 52,010 98,747 +164% +90%

Nigeria 8,384 12,845 43,200 +415% +236%

Pakistan 4,927 6,432 17,533 +256% +173%

United States 14,837 8,339 15,623 +5% +87%

Other nationalities 99,352 68,072 137,842 +39% +102%

Total 284,721 228,966 432,279 +52% +89%

However, while the new system does appear to be, as planned, facilitating skilled and
student migration, it is also – as many had forecast – resulting in labour and skill shortages
in sectors and occupations. This has led to some modifications of the policy, with special
visa routes for HGV drivers and poultry workers, and most importantly the extension of
the Health and Care Worker visa to care workers, including those on relatively low pay, for
whom the salary threshold is now £10.10/hour (only about 6% more than the minimum
wage).

Portes (2022) discusses the potential economic impacts. Earlier views on the potential
impacts on the UK economy of the end of free movement and the transition to the new
system may have been too pessimistic, based as they were on the assumption the new
system would be more restrictive. For example, Forte and Portes (2019) estimated that the
new system would result in a reduction in UK GDP of up to 2% over ten years. However,
updating their estimates to reflect the new system results in a much smaller fall in GDP,
and indeed a small rise in GDP per capita.

More immediately, there have been numerous anecdotal reports of wage rises in sectors
and occupations experiencing skill and labour shortages, in particular some transport
and logistics roles and the hospitality sector. The new immigration system is unlikely to
be the main driver of UK wage growth, either overall or for these specific sectors – the
disruption resulting from the pandemic and subsequent reopening, and the sharp rise
in inflation, are likely to be far more important (e.g. Duval et al. 2022). However, raising
relative wages among low-skilled and low-paid workers most exposed to competition
from ‘cheap’ immigrant workers was a key objective – economic and political – of the new 77
system, so it is clearly of interest to examine whether this is in fact the result.

THE ECONOMICS OF THE UK’S POST-BREXIT IMMIGRATION SYSTEM | PORTES

So far – despite the anecdotes – there is little quantitative evidence to suggest that the new
system has in fact boosted wages for such workers. In particular, wages in the hospitality
sector (“accommodation and food services”), while they have grown strongly over the last
year, remain lower relative to median wages than before the pandemic, despite the large
shock to migrant labour supply. Meanwhile, the overall wage distribution has returned
to almost exactly what it was before the pandemic (in contrast to the US, where low wage
workers have made significant relative gains).

FIGURE 4 WAGE GROWTH AND CHANGE IN EU ORIGIN WORKERS DURING THE PANDEMIC

Information

0.10 Arts
Change in log median pay between

Professional activites
Feb 2020 and June 2021

Other services Financial

Administrative Human health
Agriculture Education
0.05 Electricity
Transport
Wholesale
Manufacturing Real estate
Water
Accomodation
Public admin.
Mining
Construction

0.00
-3 -2 -1 0 1
Change in the share of EU workers between Feb 2020
and June 2021 by sector
Source: Author’s calculations from HMRC data.

So, the current picture appears to be one of a UK immigration system rapidly (re)-
orientating from Europe to the rest of the world, especially South and Southeast Asia, and
an attendant reorientation between sectors away from hospitality and some agriculture
and manufacturing sectors towards health, care, and ICT. While this will alleviate, if not
solve, some of the extreme pressures on staffing in the health and social care system, it is
as yet far from clear that it will have much impact on wage growth in other sectors.
 Considerable uncertainty remains, with an additional dimension resulting from the
government’s decision to offer entry visas to the almost three million British National
(Overseas) passport holders from Hong Kong and their dependents (Home Office 2020).
Nevertheless, while the overwhelming consensus amongst economists that Brexit – and
78 in particular, the ‘hard Brexit’ pursued by the UK government – will have significant
negative impacts on trade and investment, and hence on the broader UK economy,
there is much more cause for optimism about the impacts of the new post-Brexit UK
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

immigration system.

REFERENCES

Baldwin, R (ed) (2016), Brexit Beckons: thinking ahead by leading economists, CEPR
Press.

Duval, R, Y Ji, L Li, M Oikonomou, C Pizzinelli, S Shibat, A Sozzi and M Tavares (2022),
“Labor Market Tightness in Advanced Economies”, IMF Staff Discussion Note, March.

Forte, G and J Portes (2017) “The economic impact of Brexit-induced reductions in

migration to the UK”, VoxEU.org, 5 January.

Forte, G and J Portes (2019), “Migration to Wales: the impact of post-Brexit policy
changes”, Welsh Centre for Public Policy, University of Cardiff, February.

Hobolt, S (2016), “The Brexit vote: a divided nation, a divided continent”, Journal of
European Public Policy 23(9): 1259-1277

Home Office (2020), “Impact Assessment: Hong Kong British National Overseas Visa”,
October.

Portes, J (2016), ‘Immigration after Brexit’, National Institute Economic Review 238:
R13-R21.

Portes, J (2020), “Between the Lines: Immigration to the UK between the Referendum
and Brexit”, DCU Brexit Institute Working Paper 12-2020, Bridge Network, December.

Portes, J (2022), “Immigration and the UK economy after Brexit”, Oxford Review of
Economic Policy 38(1): 82–96.

Runge, J (2019), “Overview of UK attitudes towards immigration”, NIESR Briefing,

National Institute of Economic and Social Research, August.
ABOUT THE AUTHOR

Jonathan Portes is Professor of Economics and Public Policy at King’s College London and
a Senior Fellow of the ESRC UK in a Changing Europe programme. Previously, he was
Director of the National Institute of Economic and Social Research; and prior to that 79
Chief Economist at the UK Cabinet Office, where he advised the Cabinet Secretary and 10
Downing Street on economic and financial issues. Before that he held a number of other

THE ECONOMICS OF THE UK’S POST-BREXIT IMMIGRATION SYSTEM | PORTES

senior economic policy posts in the UK government. His particular interests include
immigration, labour markets, and poverty.
CHAPTER 9
How is the end of free movement 81

affecting the low-wage labour force in

HOW IS THE END OF FREE MOVEMENT AFFECTING THE LOW-WAGE LABOUR FORCE IN THE UK? | SUMPTION
the UK?
Madeleine Sumption1
Migration Observatory at the University of Oxford

The post-Brexit immigration system was simultaneously liberal and restrictive. It

liberalised access to the UK labour market for non-EU citizens, lowering salary thresholds
and skills requirements for work visas. But it was much more restrictive that the previous
status quo for EU citizens, who had enjoyed full access to the UK labour market before
the end of the post-Brexit transition period in December 2020.

Since free movement ended at the end of December 2020, many employers in industries
that previously had relied heavily on EU citizens can no longer recruit newly arriving EU
citizens because the jobs are not eligible for work visas. Analysis conducted over the past
six years repeatedly showed that some industries would be particularly affected by the
introduction of a ‘skill-selective’ work visa system (Vargas Silva 2016, MAC 2018, Morris
2020) – including, for example, hospitality and retail, where employers relied heavily on
EU citizens and also offered relatively few jobs that met the skill and salary criteria for
work visas.

Over the course of 2021, headlines were filled with stories of labour shortages, including
HGV drivers driving fuel tankers and stocking up retail stores; pig farmers claiming
that they must slaughter thousands of animals because of a lack of abattoir workers and
skilled butchers; and demands from numerous other industries – from hospitality to the
care sector – for greater access to visas to supply their labour. So, what do we know so
far about the impacts of ending free movement on industries such as these? This chapter
examines the evidence from the first year of the post-Brexit immigration system.

1 Thanks to Zach Strain for Research Assistance. Research for this project was produced with the support of Trust for
London and Research England’s PSF allocation to the University of Oxford.
 HOW HAS THE EU MIGRANT WORKFORCE CHANGED SINCE BREXIT AND THE
PANDEMIC?

Brexit and the pandemic have together had a significant impact on the shape of the
82 migrant workforce in the UK. First, the pandemic led to substantial job losses, pushing
many EU citizens to leave the country (ONS 2021). Second, as the economy and labour
market started to recover, the new immigration system changed the way immigration
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

could respond. Whereas under free movement, we might have expected EU migration
to bounce back during the economic recovery, the new immigration system provides few
options for migration to take up low-wage work – as Jonathan Portes explains in his
contribution to this eBook.

So, how has the end of the free movement combined with the pandemic affected the
migrant workforce? The best data to answer this question come from tax records and are
published by HMRC and the Office for National Statistics (ONS), and show the number
of jobs held by people of different nationalities (if the same person has two different jobs,
they are counted twice). At the time of writing, they cover the period up to June 2021, i.e.
the first six months of the new immigration system. These data show that, overall, the EU
citizen workforce had not returned to its pre-pandemic size by mid-2021. In June 2021,
it remained 6% smaller than two years previously. By contrast, the number of jobs held
by UK workers had almost returned to pre-pandemic levels (99%), and among non-EU
citizens the figure was 9% higher.

However, the picture varies a lot by industry. In fact, the overall decline in the employment
of EU workers was primarily driven by two industries: hospitality, which saw a net decline
of just over 98,000 EU citizen jobs in the two years to June 2021; and administrative and
support services (a category that includes a range of mostly low-wage service positions
such as building cleaning and maintenance, as well as some agency work), where the same
figure fell by just under 64,000.

Some industries have therefore seen a much sharper-than-average decline in the EU

migrant workforce, while in others the number of jobs held by EU migrants increased
despite the pandemic and post-Brexit immigration system (Figure 1).
These patterns are partly driven by the fact that some industries expanded while
others contracted during this two-year period during the pandemic, but that is not the
whole story. Different industries’ relative reliance on EU migrant workers also shifted.
In hospitality, for example, the EU workforce recovered much slower than the UK or
non-EU workforce, as a result of which the EU share of jobs held decreased by four
percentage points. By contrast, jobs held by EU migrants increased faster than for other
groups of workers in construction and in transport and storage, despite Brexit and the
pandemic (Figure 2).
FIGURE 1 CHANGE IN EU EMPLOYEE JOBS, JUNE 2019 TO JUNE 2021

Construction
Transport & storage

Health and social work

83
IT
Professional, scientific & technical

HOW IS THE END OF FREE MOVEMENT AFFECTING THE LOW-WAGE LABOUR FORCE IN THE UK? | SUMPTION
Education

Retail
Finance

Manufacturing
Support services
Hospitality

Agriculture

-35% -30% -25% -20% -15% -10% -5% 0% 5% 10% 15%

Source: ONS (2022e).

The data do not distinguish between people who have recently arrived in the UK versus
those who have lived in the country for some time. The variable picture by industry is
consistent with the expectation that some industries, such as hospitality, have traditionally
relied more heavily on newly arriving migrants, while others recruit in larger numbers
from the existing workforce. In 2019, for example, hospitality was the industry whose EU
workforce was most likely to comprise recent arrivals. An estimated 35% of EU-born
workers in hospitality had arrived in the previous five years, well above the average across
all industries of 22% or the 18% share in the construction industry (author’s calculations
from the Annual Population Survey). This would imply that the impact of Brexit and the
pandemic on industries like construction may simply take longer to emerge.

ARE EMPLOYERS EXPERIENCING SHORTAGES?

There are certainly signs that the labour market in early 2022 is tight, i.e. that it is more
challenging than usual for employers to find candidates. Vacancy rates have risen sharply
in the UK, particularly in hospitality and health (ONS 2022a), and one third of businesses
with at least ten employees said that they were facing staff shortages in early April 2022
(ONS 2022b). The number of people quitting their jobs and moving to another one reached
a record of more than one million in the fourth quarter of 2021, with 41% moving between
industries (ONS 2022b); voluntary job moves are an indicator of a tight labour market
(Cominetti 2022). And unemployment in the three months to February 2022 was only
3.8%, close to the record lows seen in 2019 (ONS 2022c).
 FIGURE 2 CHANGE IN PAYROLLED EMPLOYMENTS INDEXED TO JUNE 2019 BY
INDUSTRY AND NATIONALITY, UK, JUNE 2019 TO JUNE 2021

!" #! $%&'#!

84 !"#$% &''"((")$#*"+,$+),-"") &)(*+*.#0$#*1/,$+)

0123456789:',- #D31&E5%F ./01*'/. .233"0#,./01*'/.
;%<=>%?& 2@3421<41%&3
,*+
9&>@AB5()&@
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

*+,-5C5,++
,++

/+

.+
()&'*+,- ()&'*+*, ()&'*+,- ()&'*+*, ()&'*+,- ()&'*+*,

&0#.4,/+#/0#$*+(/+#,$+) 5"+.#02'#*"+ 6)2'$#*"+

0/'0/$#*"+
,*+

,++

/+

.+
()&'*+,- ()&'*+*, ()&'*+,- ()&'*+*, ()&'*+,- ()&'*+*,

7/$%#8,$+),."'*$%,9"0: ;+-"0($#*"+,$+) <$+2-$'#20*+=

'"((2+*'$#*"+
,*+

,++

/+

.+
()&'*+,- ()&'*+*, ()&'*+,- ()&'*+*, ()&'*+,- ()&'*+*,

>0"-/..*"+$%4,.'*/+#*?' !0$+.3"0#$#*"+,$+) @8"%/.$%/,$+),0/#$*%A

$+),#/'8+*'$% .#"0$=/ 0/3$*0,"-,("#"0,1/8*'%/.
,*+

,++

/+

.+
()&'*+,- ()&'*+*, ()&'*+,- ()&'*+*, ()&'*+,- ()&'*+*,

Source: © Office for National Statistics, 2022. This information is licensed under the Open Government Licence v3.0. To
view !"#$%&'()*(+&,&-#&(.-/(0#12"31(4(5.6(71(8"#(9.$-(+&.:(;<3&(=->"$3.2<"-(?-"-@
this licence, visit https://www.nationalarchives.gov.uk/doc/open-government-licence/.
1&.1"-.::6(./A#12&/B(.-/(*<C$.-2(D"$E&$(!%.-
The rise in vacancies has been highest in jobs that relied most heavily on EU workers pre-
pandemic (Joyce et al. 2022). However, high vacancy rates are not unique to the UK and
there are other competing explanations, such as high job turnover following the pandemic
recession and larger numbers of people becoming economically inactive (for example,
retiring early) (IES 2022). Many high-income economies are currently experiencing high 85
vacancy rates during the recovery from the pandemic, which suggests that the end of free
movement is by no means the only cause (Duval et al. 2022).

HOW IS THE END OF FREE MOVEMENT AFFECTING THE LOW-WAGE LABOUR FORCE IN THE UK? | SUMPTION
Disentangling the impacts of the end of free movement from other quirks of the post-
pandemic labour market is difficult. Indeed, where employers face difficulties recruiting
sufficient staff, often it is because multiple factors come together. For example, the
Migration Advisory Committee’s review of the social care workforce (MAC 2022) found
that the industry was facing serious staff shortages that were having a significant impact
on the quality and availability of adult social care, and that these shortages had been
exacerbated by the end of free movement but was fundamentally driven by other factors
– most notably, years of insufficient funding. Similarly, a high-profile shortage of HGV
drivers in the second half of 2021 had multiple potential causes; these included a decline in
EU driver numbers, but also declines in the attractiveness of the industry to UK workers
over the years, and a sharp decline in driving tests due to Covid-19 related restrictions.

HOW ARE EMPLOYERS RESPONDING TO THE END OF FREE MOVEMENT?

In principle, there are various different ways employers could respond to the end of free
movement:

• attracting more workers from the UK workforce (whether UK-born or migrants who
already live in the UK), for example by raising wages, improving working conditions
or being more flexible on hours and contracts;

• recruiting workers from abroad, usually by sponsoring them on a work visa;

• reducing the need for workers, for example through automation, switching to less
labour-intensive goods and services, or cutting back production in the UK.

Wages
Policy debates about migration often focus on the first of the adjustments outlined above
– attracting more UK workers – and ignore the fact that other responses also exist. In
theory, we should not expect employers’ primary response to be to raise wages and attract
more UK workers, since past research has generally found that immigration has limited
impacts on wages (e.g. Vargas-Silva 2020). Where there are impacts on wages, they are
expected to be relatively short lived and disappear within a few years as the economy
adjusts to accommodate newly arrived workers (Peri 2010).
 In general, tighter labour markets – with high vacancy rates and low unemployment –
are expected to cause higher wage growth at least in the short term, as employers seek
to attract or retain staff (Cominetti et al. 2022, Duval et al. 2022). However, during the
pandemic period and the first months of the post-Brexit immigration system in the UK,
86 analysis from the Institute for Fiscal Studies (Joyce et al. 2022) suggests that there has
been no correlation between rising vacancies (which have been higher in jobs previously
reliant on EU workers) and wage growth. It concludes that vacancies did not push up
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

wages from 2019 to 2022. In fact, the hospitality industry, which saw one of the largest
percentage declines in EU citizen employment from June 2019 to June 2021 (see Figure 1),
was one of the industries with the lowest wage growth during that period (ONS 2022d).

In other words, the early figures have not shown any evidence that tight labour markets
– whether due to lower EU migration or other factors – have increased wage growth in
low-wage jobs. Also unclear is how long any wage effects of immigration would last even if
they do start to emerge. In theory, increased wages are expected to encourage employers
to reduce the number of people they employ, either by turning to alternatives (such as
automation) or by passing on costs to consumers via higher prices, thus experiencing
lower demand for their goods or services. As more data on wages become available in
the coming months, the short-term effects of lower immigration on wages may start to
become clearer, although it could be some years before it would be possible to confirm the
expectation that ending free movement has not had a large impact on wages.

Recruiting migrants from overseas

Can employers who might previously have hired EU workers under free movement rules
recruit under the new immigration system instead, for example by sponsoring EU and
non-EU workers? For the most part, the immigration system is only designed to permit
this in skilled jobs – i.e. those that meet skill requirements and salary requirements of
at least £25,600 (with some exceptions, such as rates starting from £20,480 in shortage
occupations). Most low-wage jobs are not eligible, although there is a work visa route for
agriculture, and care workers can qualify for skilled worker visas if they earn at least
£10.10 per hour, which is £0.60 above the minimum wage.

Most of the industries that had above-average shares of EU migrants before the pandemic
– such as hospitality, transportation and storage, manufacturing, construction, and admin
and support services – were low users of Skilled Worker Route visas in 2021 (Figure 3a).
Looking at the results by occupation (Figure 3b), it is clear that many occupation groups
that have previously relied heavily on EU migrants sponsored no migrants on skilled
work visas at all, because they were ineligible. However, even where jobs are eligible and
occupations have in the past relied significantly on EU workers, there are often very low
rates of sponsorship. For example, just under 8% of workers in the skilled construction
and building trades were EU born in 2018-2020 (82,000 EU migrants), but employers
sponsored only 123 people in this occupation group in 2021.
 CoS as share of 2018-2020 workforce (blue bar)

0.0%
0.2%
0.4%
0.6%
0.8%
1.0%
1.2%
1.4%
1.6%
1.8%
2.0%
Elementary Admin & Services
Elementary Trades And Related
Process, Plant & Machine Operatives
Sales
Leisure, Travel And Related Personal Services
Transport & Mobile Machine Drivers And…
Protective Services

IT
Health & social work
Finance
Manufacturing
Education
Construction
Hospitality
Agriculture
Real estate
Transport & storage
Admin and support

Professional, scientific, technical

Arts & Leisure
Other Service Activities
Retail

Administrative Occupations

Share of the 2018-2020 workforce

0.0%
Customer Services
Secretarial And Related

CoS
Skilled Agricultural & Related Trades

Source: HMRC payroll data; FOI data from Home Office.

0.2%
Skilled Construction & Building Trades
Skilled Electrical & Metal Trades
% of total employment in the industry (Jan 2021)

Health & Social Care Associate Professionals

0.4%

Notes: Data show Certificates of Sponsorship for out of country visas.

Other Managers And Proprietors

FIGURE 3B SKILLED SPONSORED WORKERS IN 2021

Science, Engineering & Tech Associate…

EU worker share
Teaching & Educational Professionals
0.6%

Culture, Media And Sports Occupations

Corporate Managers And Directors
Caring Personal Services
0.8%

Food-related & other skilled trades

FIGURE 3A SKILLED SPONSORED WORKS (EU AND NON-EU) IN 2021

Business & Public Service Associate Professionals

Business, Media & Public Service Professionals
1.0%

Science, Research, Engineering & Tech

Health Professionals
1.2%

2.0%
6.0%

0.0%
4.0%
8.0%
12.0%
16.0%

10.0%
14.0%
18.0%

EU born as share of 2018-2020 workforce

87

HOW IS THE END OF FREE MOVEMENT AFFECTING THE LOW-WAGE LABOUR FORCE IN THE UK? | SUMPTION
 Low sponsorship rates in most middle-skilled jobs that previously relied on EU citizens is
likely to result from the costs and bureaucracy associated with the using the immigration
system. Some smaller employers also perceive sponsoring migrants on work visas as risky
– concerned, for example, that inadvertent non-compliance with rules on monitoring
88 employees could mean they lose their license (MAC 2022).

Because there are start-up costs involved in becoming a sponsor and learning how to
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

sponsor workers, it is possible that larger numbers of employers will turn to the work visa
system over time.

From 2019 to 2021, the EU migrant workforce fell at the same time as the number of
non-EU workers increased (ONS 2022e), raising the question whether employers have
substituted from EU to non-EU workers. For the most part, however, the industries
that have driven the increase in non-EU citizen employment are not the same ones that
drove the decrease in EU citizen employment. In particular, the biggest decline in EU
employment was in hospitality, where very few jobs are eligible for work visas, while the
increase in non-EU workers, by contrast, was driven by the health sector (ONS 2022e).

One notable exception is the agricultural industry, where employers have been able to
recruit non-EU citizens using the Seasonal Workers visa. In June 2021, the number of
EU citizen-held jobs in agriculture was 13,000 (28%) lower than it had been in the same
month two years earlier (Figure 1). At the same time, agricultural employers turned in
large numbers to the Seasonal Workers visa, which expanded in 2021 to 30,000 places
(with 29,600 visas granted).

Another industry where there is some evidence of substitution from EU to non-EU

workers is health, although this trend predates the post-Brexit immigration system and
instead started at around the time of the EU referendum. Before the referendum, new
staff joining the NHS had increasingly come from EU countries and by 2016 roughly half
of new joiners from outside the UK were EU nationals (excluding those whose nationality
is not known) (Figure 4). This trend reversed from late 2016 onwards, and by the year
ending September 2021 only 21% of non-UK new joiners were from EU countries.

In other words, there have been some cases where there is clear evidence that employers
have been able to respond to lower EU migration by recruiting staff on work visas. In most
of the jobs that were previously particularly reliant on EU workers, however, employers
have not turned to the work system in large numbers, whether because they are not
eligible or because of the costs and perceived risks of doing so.
FIGURE 4 EU AND NON-EU STAFF JOINING THE NHS, AS SHARE OF NON-UK JOINERS

FTE, excluding those of unknown nationality

90%
80%
70% 89
60%

HOW IS THE END OF FREE MOVEMENT AFFECTING THE LOW-WAGE LABOUR FORCE IN THE UK? | SUMPTION
50%
40%
30%
20%
10%
0%
200909 201009 201109 201209 201309 201409 201509 201609 201709 201809 201909 202009
to to to to to to to to to to to to
201009 201109 201209 201309 201409 201509 201609 201709 201809 201909 202009 202109

EU Non-EU

Source: NHS Digital.

Note: figures are presented as a percentage of EU or non-EU new joiners. This is because from 2019 to 2021 the share of
workers with ‘unknown’ nationality substantially decreased due to improvements in data collection; therefore presenting
the figures as absolute numbers can distort the overall picture.

Reducing demand for workers

In the medium to long run, one of the main impacts of the changing shape of the workforce
under the new immigration system should be that industries that were most reliant on
EU workers grow less quickly than they would have done if free movement had continued
(MAC 2018). There is some early qualitative evidence of this in practice since the end of
free movement. For example, research in the horticultural sector found that UK growers
facing staff shortages had in some cases moved production overseas or invested in
automation in pack houses to reduce their demand for workers (Barbulescu et al. 2021).
The Migration Advisory Committee’s (2022) report on social care found evidence that
worker shortages (not primarily resulting from the end of free movement but exacerbated
by it) had meant that less care could be provided, with some care providers handing back
contracts that they could not fulfil. In February/March 2022, among employers who said
they were facing worker shortages in the ONS Business Insights survey in February/March
2021, around half said that they were “unable to meet demands” as a result, suggesting
that business growth was constrained (ONS 2022b).

CONCLUSION

It is still early to assess the full impacts of the end of free movement and the introduction
of the new immigration system. The data available at the time of writing in May 2021
suggest that many employers in industries that previously relied heavily on EU migration
are finding it difficult to recruit, but that this has not necessarily translated into high wage
 growth for low-wage workers. The end of free movement is likely to have played some role
in the tight labour market, but is not the only factor and broadly similar patterns have
played out in other high-income countries.

While some employers have turned to overseas recruiting under the new immigration
90
system, this has not been the norm in lower-wage industries, likely as a result of the new
system’s eligibility criteria and costs. The consequences of employers’ greater difficulty
THE ECONOMICS OF BREXIT: WHAT HAVE WE LEARNED?

recruiting are still playing out. In theory, shortages are expected to be short term and to
translate into slower employment growth in lower-wage jobs or a reduction in demand
for these goods and services as any wage increases that do take place translate into higher
prices. There is some emerging, but still limited, evidence that these adjustments are
taking place, with some employers exploring automation or reducing the amount of
activity in response to staff shortages.

REFERENCES

Barbulescu, R, C Vargas-Silva and B Robertson (2021) “Written Evidence submitted by

Dr Roxana Barbulescu, Prof. Carlos Vargas-Silva and Dr Bethany Robertson (LS0035)”,
written evidence to the Environment, Food and Rural Affairs Committee, October.

Cominetti, N, K Handscomb, H Slaughter and G Thwaites (2022), The Resolution

Foundation Labour Market Outlook, London: Resolution Foundation.

Duval, R, Y Ji, L Li, M Oikonomou, C Pizzinelli, I Shibata, A Sozzi and M Tavares

(2022), “Labor market tightness in advanced economies”, IMF Staff Discussion Note
SDN/2022/001.

IES – Institute for Employment Studies (2022), “Labour market statistics: March 2022”,
IES briefing note, Institute for Employment Studies.

Joyce, R, F Postel-Vinay, P Spittal and X Xu (2022), “Job opportunities after the pandemic”,
IFS briefing note BN342, Institute for Fiscal Studies.

MAC – Migration Advisory Committee (2018), EEA migration to the UK: Final report,
London: Migration Advisory Committee.

MAC (2022), Review of adult social care 2022, London: Migration Advisory Committee.

Morris, M (2020), Building a Post-Brexit Immigration System for the Economic Recovery,
London: IPPR.

ONS – Office for National Statistics (2021), “Long-term international migration,

provisional: year ending December 2020”.

ONS (2022a), “VACS02: Vacancies by Industry”, Dataset (accessed 12 April 2022).

ONS (2022b), “Business insights and impact on the UK economy”, Dataset (accessed 21
April 2022).
ONS (2022c), “A02 NSA: Employment, unemployment and economic inactivity for people
aged 16 and over and aged from 16 to 64 (not seasonally adjusted)”, Dataset.

ONS (2022d), “Earnings and employment from Pay As You Earn Real Time Information,
seasonally adjusted”, Dataset ().
91
ONS (2022e), “Changes in payrolled employments held by non-UK nationals during the
coronavirus (COVID-19) pandemic and EU Exit periods”.

HOW IS THE END OF FREE MOVEMENT AFFECTING THE LOW-WAGE LABOUR FORCE IN THE UK? | SUMPTION
Peri, G (2010), The impacts of immigrants in recession and economic expansion,
Washington DC: Migration Policy Institute.

Vargas-Silva, C (2016), Potential implications of admission criteria for EU nationals

coming to the UK, Oxford: Migration Observatory.

Vargas-Silva, C (2020), The labour market effects of immigration, Oxford: Migration

Observatory.

ABOUT THE AUTHOR

Madeleine Sumption is the Director of the Migration Observatory at the University of

Oxford, which provides impartial analysis to inform public debates about migration in
the UK. Madeleine’s research focuses on labour migration, investor visas, the economic
and social impacts of migration policies, and immigrant integration. Madeleine is a
member of the Migration Advisory Committee, an independent panel that advises the UK
Government on migration issues. She is also Chair of Migration Statistics User Forum,
which brings together producers and users of migration data. In 2017, she received an
MBE for services to social science.
 On 1 January 2021, the UK implemented the biggest
changes to its trade policy in 40 years, as well as
introducing an entirely new immigration system; all in
the middle of a global pandemic. Eighteen months on,
what have we learned about the impacts of Brexit on the
UK economy?

This eBook, produced by VoxEU/CEPR and UK in a

Changing Europe, brings together leading academic
researchers on trade, immigration and the political
economy of Brexit to present new research and evidence
on different aspects of this question. It shows that
Brexit has already had measurable and meaningful
impacts on the structure and volume of UK trade and
migration. While some of these impacts are consistent
with economists’ predictions, others are surprising and
unexpected.

As well as explaining what has happened so far,

contributors also suggest how the UK can adapt to the
implications of new trading arrangements. Brexit will
reshape the UK’s place in the global economy. This book
both analyses how those changes are playing out and
helps to point the way forward.

ISBN: 978-1-912179-61-9

ISBN 978-1-912179-61-9 33 Great Sutton Street | LONDON EC1V 0DX | UK

187 boulevard Saint-Germain | 75007 Paris | France
TEL: +44 (0)20 7183 8801 | FAX: +44 (0)20 7183 8820
EMAIL: CEPR@CEPR.ORG
9 781912 179619 WWW.CEPR.ORG