BREXIT 2016
Policy analysis from the
Centre for Economic Performance
#CEPBREXIT
PAPERBREXIT
Disclaimer:
The Centre for Economic Performance (CEP) is a
politically independent Research Centre at the London
School of Economics. The CEP has no institutional
views, only those of its individual researchers.
Professor John Van Reenen who joined the CEP as
Director in 2003, did not (and does not) support joining
the Euro.
CEP’s Brexit work is funded by the UK Economic and
Social Research Council. As a whole the CEP, receives
less than 5% of its funding from the European Union.
The EU funding is from the European Research Council
for academic projects and not for general funding or
consultancy.
BREXIT 2016
Policy analysis from the Centre for Economic Performance
June 2016
Centre for Economic Performance
London School of Economics and Political Science
Houghton Street, London WC2A 2AE, UK
Tel: +44 (0)20 7955 7673 Email: [email protected]
Web: http://cep.lse.ac.uk Twitter: @CEP_LSE
Contents

#CEPBREXIT
PAPERBREXIT
BREXIT 2016
Introduction
i
Introduction
On June 23rd, the British people will vote in a referendum over whether or not to remain in the
European Union. It is the most important vote that most of us will have in our lifetimes. And
one that will have major repercussions for our country and the rest of the world for decades, if
not generations, to come.
Ever since David Cameron made his Bloomberg speech in January 2013 promising the
Referendum, I knew that this was likely to become the major issue. I was lucky enough to be
able to put a team together at the CEP of the world’s top researchers on international trade,
labour markets and growth. We were able to develop the new methods, theories and data to
address the deep and complex problem of the economic consequences of a decision to leave an
alliance we had been a key member of for over 40 years.
We published several reports over the last three years on the Brexit debate, especially in the
last three months, and this book is a selection of the fruits of our labour.
More information with the reports, blogs, and more technical details and so on can be found
here http://cep.lse.ac.uk/BREXIT/
The reports are self-contained and we have not re-written them from the originals.
Our conclusions are quite clear. Leaving the EU will make Britain poorer than it would be were
we to remain. There is simply no room for serious doubt.
The EU is far from perfect. It is over-bureaucratic and insufficiently democratic. However, the
more our research progressed, the more compelling the case for Remain became and the more
obvious it was that the Leave campaign had no coherent vision of life outside the EU.
In short, the UK will be poorer in the long-run from leaving because we will trade less with
our closest neighbours, losing full access to the largest Single Market on the planet. We will
have less foreign investment because of these weaker ties. And there will be an enormous
increase in uncertainty as we spend many, many painful years renegotiating the relationship
with Europe and the rest of the world.
The amount we save from paying less of an “entry fee” to Brussels is peanuts by comparison
to these losses. We know the “£350m a week” is a lie with Britain’s true net contribution less
than half of this. But this constitutes only about 0.4% of our national income, a trivial amount
compared to the estimated loss of 6% to 9% Brexit induced loss of national income.
The economic pain of Brexit is shared pretty evenly across households – the poor certainly do
not escape, although those on middle incomes are hit slightly harder than the rich.
The economic damage from Brexit can be reduced if we “do a Norway” and remain in the
European Economic Area. But this will mean we will have to continue to pay most of what we
currently do, and we will have to implement most of the Single Market rules without having
any voting rights on what these rules are.
ii
What makes this damage limitation exercise unlikely is that countries like Norway (and
Switzerland) also have to allow free EU migration. Immigration has dominated the last weeks
of the campaign, almost to the exclusion of all else.
Our research finds that EU immigration has benefited the UK. First, access to the Single Market
“buys” a big increase in real wages through higher productivity. Second, because EU
immigrants are more likely to be in work and are younger and better educated than the British
born, they pay more in tax than they take out in welfare. So immigrants have helped subsidise
the NHS and other public services for British people.
Finally, people born in the UK who live in areas of the country that have had big influxes of
EU migrants have not suffered lower wages or job opportunities. The only group which seems
to have a very small loss of wages from immigration are unskilled migrants.
To many people it seems obvious that migration is bad for jobs as we all know stories of how
a friend has gone for a job and a migrant got it. But there isn’t a fixed lump of jobs. Migrants
have to live, sleep, eat and drink so they increase demand and this increased expenditure creates
new jobs. This means that the net effect of immigration in an area turns out to be zero.
Similarly with public services – it seems hard to get a place at a school or a doctors’
appointment because of EU migrants. But since migrants pay more in tax than they take out,
there’s actually plenty of money to go around; it’s just that the government has not spent it
wisely on expanding services in the places they are needed.
People have suffered over the last decade. Real wages fell by over 8% in the 6 years after 2008.
But EU immigration was rising before 2008 and over the last two years when wages have
turned around. The pay cuts were due to the global financial crisis and a tough austerity package
– it was nothing to do with immigration. As with public services, EU migrants are part of the
solution, not part of the problem.
Our book concludes with a critique of the work of others. We give some comments on the
Treasury’s analysis of Brexit, which we think is overly cautious but comes to similar
conclusions to us over the harm of Brexit. We also include our commentary on the only
academic economist who has tried to make a semi-coherent case for Brexit, Professor Patrick
Minford. His case calls for ‘unilateral free trade’, the elimination of manufacturing jobs and an
enormous increase in wage inequality. His analysis is inconsistent with the most basic realities
of modern trade.
I hope that you enjoy the work here, find it thought provoking and that it helps in your decision
over the next few days.
John Van Reenen, Director of the Centre for Economic Performance and Professor of
Economics, London School of Economics
#CEPBREXIT
PAPERBREXIT
Summary Points
i
CEP BREXIT ANALYSIS
Life after Brexit: What are the UK’s options outside the European Union?
 It is highly uncertain what the UK’s future would look like outside the European Union
(EU), which makes ‘Brexit’ a leap into the unknown. This report reviews the advantages
and drawbacks of the most likely options.
 After Brexit, the EU would continue to be the world’s largest market and the UK’s
biggest trading partner. A key question is what would happen to the three million EU
citizens living in the UK and the two million UK citizens living in the EU?
 There are economic benefits from European integration, but obtaining these benefits
comes at the political cost of giving up some sovereignty. Inside or outside the EU, this
trade-off is inescapable.
 One option is ‘doing a Norway’ and joining the European Economic Area. This would
minimise the trade costs of Brexit, but it would mean paying about 83% as much into the
EU budget as the UK currently does. It would also require keeping current EU regulations
(without having a seat at the table when the rules are decided).
 Another option is ‘doing a Switzerland’ and negotiating bilateral deals with the EU.
Switzerland still faces regulation without representation and pays about 40% as much as
the UK to be part of the single market in goods. But the Swiss have no agreement with
the EU on free trade in services, an area where the UK is a major exporter.
 A further option is going it alone as a member of the World Trade Organization. This
would give the UK more sovereignty at the price of less trade and a bigger fall in income,
even if the UK were to abolish tariffs completely.
 Brexit would allow the UK to negotiate its own trade deals with non-EU countries. But as
a small country, the UK would have less bargaining power than the EU. Canada’s trade
deals with the United States show that losing this bargaining power could be costly for
the UK.
 To make an informed decision on the merits of leaving the EU, voters need to know more
about what the UK government would do following Brexit.
 This is the first in a series of briefings analysing the economic costs and benefits of Brexit
for the UK.
Centre for Economic Performance
London School of Economics and Political Science
Houghton Street, London WC2A 2AE, UK
Tel: +44 (0)20 7955 7673
Email: [email protected] Web: http://cep.lse.ac.uk
ii
CEP BREXIT ANALYSIS
The consequences of Brexit for UK trade and living standards
 The European Union (EU) is the UK’s largest trade partner. Around a half of the UK’s trade
is with the EU. EU membership reduces trade costs between the UK and the EU. This makes
goods and services cheaper for UK consumers and allows UK businesses to export more.
 Leaving the EU (‘Brexit’) would lower trade between the UK and the EU because of higher
tariff and non-tariff barriers to trade. In addition, the UK would benefit less from future
market integration within the EU. The main economic benefit of leaving the EU would be a
lower net contribution to the EU budget.
 Our analysis first quantifies the ‘static’ effects of Brexit on trade and income. In an
‘optimistic’ scenario, the UK (like Norway) obtains full access to the EU single market. We
calculate this results in a 1.3% fall in average UK incomes (or £850 per household). In a
‘pessimistic’ scenario with larger increases in trade costs, Brexit lowers income by 2.6%
(£1,700 per household).
 All EU countries lose income after Brexit. The overall GDP fall in the UK is £26 billion to
£55 billion, about twice as big as the £12 billion to £28 billion income loss in the rest of the
EU combined. Non-EU countries experience some smaller income gains.
 If the UK unilaterally removed all its tariffs on imports from the rest of the world after
Brexit, UK incomes fall by 1% in the optimistic case and 2.3% in the pessimistic case.
 In the long run, reduced trade lowers productivity. Factoring in these effects substantially
increases the costs of Brexit to a loss of 6.3% to 9.5% of GDP (about £4,200 to £6,400 per
household).
 Being outside the EU means that the UK would not automatically benefit from future EU
trade deals with other countries. This would mean missing out on the current US and
Japanese deals, which are forecast to improve real incomes by 0.6%.
 After Brexit, would the UK obtain better trade deals with non-EU countries? It would not
have to compromise so much with other EU states, but the UK would lose bargaining power
as its economy makes up only 18% of the EU’s ‘single market’.
 It is unclear whether there are substantial regulatory benefits from Brexit. The UK already
has one of the OECD’s least regulated product and labour markets. ‘Big ticket’ savings are
supposedly from abolition of the Renewable Energy Strategy and the Working Time
Directive – both of which receive considerable domestic political support in the UK.
Centre for Economic Performance
London School of Economics and Political Science
Houghton Street, London WC2A 2AE, UK
Tel: +44 (0)20 7955 7673
Email: [email protected] Web: http://cep.lse.ac.uk
iii
The impact of Brexit on foreign investment in the UK
 Foreign direct investment (FDI) raises national productivity and therefore output and
wages. Multinational firms bring in better technological and managerial know-how,
which directly raises output in their operations. FDI also stimulates domestic firms to
improve – for example, through stronger supply chains and tougher competition.
 The UK has an FDI stock of over £1 trillion, about half of which is from other members
of the European Union (EU). Part of the UK’s attractiveness for foreign investors is that
it brings easy access to the EU’s Single Market. After Brexit, higher trade costs with the
EU would be likely to depress FDI.
 Our new empirical analysis looks at bilateral FDI flows between 34 OECD countries
(including the UK) over the last three decades. Controlling for many other factors, the
baseline estimate is that EU membership has raised FDI by about 28%.
 The positive effect of EU membership on FDI is robust, ranging between 14% and 38%
under different statistical assumptions. The size of these effects is also consistent with
comparisons between UK FDI flows and a set of matched control countries.
 Striking a comprehensive trade deal – for example, joining Switzerland in the European
Free Trade Association – would not significantly reduce the negative effects of Brexit on
FDI, according to the data.
 Assessing the impact of lower FDI on income is complex. We use existing
macroeconomic estimates of how FDI affects growth combined with a very conservative
estimate of the impact of Brexit – a 22% fall in FDI over the next decade. We calculate
that a Brexit-induced fall in FDI could cause a 3.4% decline in real income – about £2,200
of GDP per household. The income losses due to lower FDI are larger than our estimates
of static losses due to lower trade of 1.3% to 2.6%.
 Estimates of the impact of Brexit on the UK’s car industry imply that UK production
would fall by 181,000 cars (12%) and prices would rise by 2.5%. Even if the UK manages
a comprehensive trade deal and keeps tariffs at zero, production would fall by 36,000
cars.
 The UK’s financial services industry is the largest recipient of FDI. Restrictions on
‘single passport’ privileges following Brexit, would lead to big cuts in activity.
Furthermore, the UK would be unable to challenge EU regulations at the European Court
of Justice.
Centre for Economic Performance
London School of Economics and Political Science
Houghton Street, London WC2A 2AE, UK
Tel: +44 (0)20 7955 7673
Email: [email protected] Web: http://cep.lse.ac.uk
CEP BREXIT ANALYSIS
iv
CEP BREXIT ANALYSIS
Brexit and the impact of immigration on the UK
 Between 1995 and 2015, the number of immigrants from other European Union (EU)
countries living in the UK tripled from 0.9 million to 3.3 million. In 2015, EU net
immigration to the UK was 172,000, only just below the figure of 191,000 for non-EU
immigrants.
 The big increase in EU immigration occurred after the ‘A8’ East European countries joined
in 2004. In 2015 29% of EU immigrants were Polish.
 EU immigrants are more educated, younger, more likely to be in work and less likely to claim
benefits than the UK-born. About 44% have some form of higher education compared with
only 23% of the UK-born. About a third of EU immigrants live in London, compared with
only 11% of the UK-born.
 Many people are concerned that immigration reduces the pay and job chances of the UKborn due to more competition for jobs. But immigrants consume goods and services and this
increased demand helps to create more employment opportunities. Immigrants also might
have skills that complement UK-born workers. So we need empirical evidence to settle the
issue of whether the economic impact of immigration is negative or positive for the UK-born.
 New evidence in this Report shows that the areas of the UK with large increases in EU
immigration did not suffer greater falls in the jobs and pay of UK-born workers. The big falls
in wages after 2008 are due to the global financial crisis and a weak economic recovery, not
to immigration.
 There is also little effect of EU immigration on inequality through reducing the pay and jobs
of less skilled UK workers. Changes in wages and joblessness for less educated UK-born
workers show little correlation with changes in EU immigration.
 EU immigrants pay more in taxes than they take out in welfare and the use of public services.
They therefore help reduce the budget deficit. Immigrants do not have a negative effect on
local services such as crime, education, health, or social housing
 European countries with access to the Single Market must allow free movement of EU
citizens whether in the EU (like the UK) or outside it (like Norway and Switzerland).
 The refugee crisis has nothing to do with EU membership. Refugees admitted to Germany
have no right to live in the UK. The UK is not in the Schengen passport-free travel agreement
so there are border checks on migrants.
Centre for Economic Performance
London School of Economics and Political Science
Houghton Street, London WC2A 2AE, UK
Tel: +44 (0)20 7955 7673
Email: [email protected] Web: http://cep.lse.ac.uk
v
CEP BREXIT ANALYSIS
Who bears the pain?
How the costs of Brexit would be distributed across income groups
 All serious economic analysis finds that Brexit would have a negative impact on UK GDP
per capita. But a popular view is that membership of the European Union (EU) only benefits
elites and has not helped those in the middle or at the bottom of the income distribution.
 Our research uses data on household expenditure by different income groups and household
types combined with estimates of changes in the prices of goods and services after Brexit to
look at who would win and who would lose.
 We find that prices would go up most in transport (a price hike of between 4% and 7.5%),
alcohol (4% to 7%), food (3% to 5%) and clothing (2% to 4%). These product groups rely a
lot on imports. By contrast, prices for services would rise the least.
 We show that the living standard of every income group would be lower after Brexit due to
these higher prices. Those on middle incomes would suffer slightly more in proportionate
terms than the richest and poorest households.
 Looking solely at the ‘static’ short-run impact of trade, the income (not GDP) of the average
UK household would drop by 1.8% (£754) per year in our most ‘optimistic’ scenario where
the UK joins countries like Norway in the European Economic Area. Income would fall by
4% per year (£1,637) if the UK were to trade under World Trade Organization rules (in our
more realistic ‘pessimistic’ scenario). If we take account of the longer-run dynamic effects
of Brexit on productivity, the average household would lose between 6.1% and 13.5% of
their real incomes per year (£2,519 to £5,573).
 For the poorest tenth of households (the bottom decile), real income losses would be 1.7%
to 3.6% in the short run and 5.7% to 12.5% in the long run. For the richest households, the
short-run losses would be 1.8% to 3.9% and the long-run losses 6% to 13.4%. So the middle
class would lose out the most, but only by a bit.
 Looking at specific households such as pensioners, families with children and single people,
we find that the pain would also be widely shared. For example, even in the short run,
pensioners would lose between 2% and 4% of their income.
 Adding in the effects of reduced immigration and the differential effects of trade by industry
has no discernible effect on our analysis of inequality.
Centre for Economic Performance
London School of Economics and Political Science
Houghton Street, London WC2A 2AE, UK
Tel: +44 (0)20 7955 7673
Email: [email protected] Web: http://cep.lse.ac.uk
vi
CEP BREXIT ANALYSIS
True long-run costs of Brexit likely to be higher than Treasury estimates
Commentary from the Centre for Economic Performance
 Overly cautious assumptions in the Treasury’s recent report on the long-run
consequences for the UK economy of leaving the European Union (EU) mean that it
has probably underestimated the economic costs. That is one of the conclusions in a
commentary on the Treasury’s analysis published today by the Centre for Economic
Performance (CEP) at the London School of Economics.
 Forecasting the economic consequences of Brexit is a difficult challenge and all
estimates will be subject to a degree of uncertainty. But the CEP research team’s
overall assessment is that the Treasury Report is a credible analysis, which, for the
most part, uses the best available estimation methods.
 The Report’s headline forecast that Brexit would reduce long-run UK GDP by 6.2% in
the Treasury’s central case of a Canadian-style negotiated bilateral trade deal is
broadly consistent with CEP’s previous work and many other independent estimates.
For example, CEP’s dynamic estimates of the cost of Brexit indicate a GDP loss of
6.3% to 9.5% in the case of the UK moving from the EU to European Free Trade
Association. Treasury estimates are at the lower end of this range.
 CEP director Professor John Van Reenen concludes:
‘The Treasury’s findings reinforce the academic and business consensus that
Brexit would make the UK significantly poorer. The Report is a serious
contribution to the debate.’
 Swati Dhingra said:
‘The Treasury Report looks at the realistic options the UK will face after
Brexit and the cost of each. It takes a conservative approach to the potential
costs.’
Centre for Economic Performance
London School of Economics and Political Science
Houghton Street, London WC2A 2AE, UK
Tel: +44 (0)20 7955 7673
Email: [email protected] Web: http://cep.lse.ac.uk
vii
CEP BREXIT ANALYSIS
Economists for Brexit: A Critique
 Professor Patrick Minford, one of the ‘Economists for Brexit’, argues that leaving the
European Union (EU) will raise the UK’s welfare by 4% as a result of increased trade.
His policy recommendation is that following a vote for Brexit, the UK should strike no
new trade deals but instead unilaterally abolish all its import tariffs.
 Under this policy (‘Britain Alone’), he describes his model as predicting the
‘elimination’ of UK manufacturing and a big increase in wage inequality. These
outcomes may be hard to sell to UK citizens as a desirable political option.
 Our analysis of the ‘Britain Alone’ policy predicts a 2.3% loss of welfare compared
with staying in the EU. This is only 0.3 percentage points better than Brexit without
unilaterally abolishing tariffs which would result in a 2.6% welfare loss.
 Minford’s results stem from assuming that small changes in trade costs have
tremendously large effects on trade volumes: according to his model, the falls in tariffs
become enormously magnified because each country purchases only from the lowest
cost supplier.
 In reality, everyone does not simply buy from the cheapest supplier. Products are
different when made by different countries and trade is affected by the distance between
countries, their size, history and wealth (the ‘gravity relationship’). Trade costs are not
just government-created trade barriers. Product differentiation and gravity is
incorporated into modern trade models – these predict that after Brexit the UK will
continue to trade more with the EU than other countries as it remains our geographically
closest neighbour. Consequently, we will be worse off because we will face higher trade
costs with the EU.
 Minford’s assumption that goods prices would fall by 10% comes from attributing all
producer price differences between the EU and low-cost countries to EU trade barriers,
ignoring differences in quality.
 Single Market rules (for example, over product safety) facilitate trade between EU
members as it creates a level playing field. Minford’s assumption that the Single Market
merely diverts trade from non-EU countries is contradicted by the empirical evidence.
 Minford also overlooks the loss in services trade that would result from leaving the
Single Market, such as ‘passporting’ privileges in financial services.
 Minford’s approach of ignoring empirical analysis of trade data seems predicated on
the view that because statistical analysis is imperfect, it should all be completely
ignored. But such statistical biases may reinforce rather than weaken the case for
remaining in the EU. Theories need grounding in facts, not ideology.
Centre for Economic Performance
London School of Economics and Political Science
Houghton Street, London WC2A 2AE, UK
Tel: +44 (0)20 7955 7673
Email: [email protected] Web: http://cep.lse.ac.uk
Life after BREXIT: What are the
UK’s options outside the
European Union?
Swati Dhingra and Thomas Sampson
#CEPBREXIT
BREXITPAPER
Introduction
Suppose the UK votes to leave the European Union (EU): what happens next? Unfortunately,
no one knows for sure.
A vote to remain in the EU is a vote to maintain the status quo. The new settlement that the
government is negotiating with the EU leaves the UK’s current economic and political
relations with Europe broadly unchanged. But what happens in the aftermath of a vote to
leave is more uncertain.
Leaving the EU would not mean that the UK could wash its hands of dealing with the rest of
Europe. As Prime Minister David Cameron noted in his 2013 Bloomberg speech committing
the Conservative Party to holding a referendum, ‘If we leave the EU, we cannot of course
leave Europe. It will remain for many years our biggest market, and forever our geographical
neighbourhood’ (Cameron, 2013).
Yet neither the government nor the campaign to leave the EU has put forward clear and
concrete proposals for what comes after Brexit. In fact, the government has explicitly ruled
out making contingency plans to cope with Brexit (Parker, 2015). To shed light on the
possible aftermath of Brexit, this report outlines some of the options for the UK outside the
EU and discusses the costs and benefits of each alternative.
Formal procedures for leaving the EU were introduced by the Lisbon Treaty, which came
into force in 2009. A country wishing to leave the EU must notify the EU of its intention and
this notification would trigger negotiations over a withdrawal agreement between the country
and the remainder of the EU. The country would officially exit the EU on the date the
withdrawal agreement came into effect or, if no agreement is reached, the country could leave
two years after the date of notification.
What matters, of course, is the content of any withdrawal agreement. Several former colonies
and overseas territories of European countries, such as Algeria in 1962 and Greenland in
1985, left the European Economic Community (EEC), the predecessor of the EU. But no
independent European country has ever left the EEC or the EU. Therefore, there is no
relevant precedent that can be used to understand the details of how the withdrawal process
would work or to shed light on how the EU would treat the exiting country.
In the event of Brexit, the UK government and the EU would need to make decisions in five
main areas.
First, what happens to the UK businesses and the two million UK citizens that are resident in
the EU and to the EU businesses and the three million EU citizens that are resident in the UK?
For example, would Britons living or working in the EU retain the same rights that they
currently enjoy or would they be treated like migrants from outside the EU? Do migrants
from the EU have the right to stay in the UK?
There is a presumption in international law that when treaty rights have been executed, those
rights are unaffected by withdrawal from the treaty (House of Commons, 2013). This
1
suggests that individuals and businesses that have taken advantage of the Single Market1 to
move either from the UK to the rest of the EU or in the opposite direction would probably be
allowed to stay. But this outcome is not certain and would certainly be a subject addressed by
any withdrawal agreement.
Second, how would UK law change following withdrawal from the EU? Currently, in areas
where the UK has ceded sovereignty to the EU, such as regulation of the Single Market, UK
law is shaped by decisions made at the EU level. EU legal decisions enter UK law in two
ways. EU directives require member states to adopt policies or change laws to achieve the
outcome specified by the directive. By contrast, when the EU issues a regulation, it
immediately becomes law in all member states. Thus, directives are enacted through changes
to UK law, while regulations have legal force only because the UK is part of the EU.
Consequently, if the UK leaves the EU, then laws that were passed to implement EU
directives would be unaffected unless the government chooses to change them. But EU
regulations would immediately lose legal force. Since EU regulations govern many important
areas, such as food hygiene and safety, this would leave a gap in UK law.
To avoid this possibility, prior to leaving the EU, the government would need to pass
legislation setting UK law in areas currently subject to EU regulations. Whether this
legislation would simply transpose EU regulations into UK law or implement new regulatory
policy is uncertain.
Leaving the EU would also mean that the UK ceased to be subject to the Charter of
Fundamental Rights of the European Union. The government would need to decide whether
any of the economic, social and political rights guaranteed to EU citizens under this Charter
should be written into UK law.
Third, the UK government would need to decide what, if any, policies to adopt in areas that
currently fall under the authority of the EU. Of particular importance would be the
government’s regional and agricultural policies since these are the biggest components of the
EU budget. Less wealthy areas of the UK, such as Northern Ireland and Wales, receive
significant funding from the EU’s regional development programmes, which would cease
following Brexit. Brexit would mean leaving the Common Agricultural Policy (CAP). The
UK as a whole would benefit from this change (Philippidis and Hubbard, 2001), but unless
the government introduced new agricultural subsidies, farmers would be among the big losers
from Brexit.
In addition, the UK is the third largest recipient of EU research and innovation funding
(Ugwumadu, 2013). Following Brexit, the government would need to decide whether to
replace this funding. After leaving the EU, the government would also regain responsibility
for issues such as competition policy and international trade negotiations, which are currently
handled at the European level. There would be a cost of developing the competencies
necessary to manage these areas, since the required skills do not currently exist within the UK
civil service.
1 The ‘Single Market’ is the name given to the integrated European economy created by removing economic
barriers between EU member states. The Single Market is based on four freedoms: the freedoms of movement
of goods, services, people and capital within the EU.
2
Fourth, would there be a transition period after the UK exits the EU during which the UK’s
rights and obligations as an EU member are phased out or would the change happen abruptly?
A transition period would allow workers and companies that do business with the EU time to
adjust to changes in laws, regulations and market access resulting from Brexit.
Fifth and probably most importantly, a withdrawal agreement would need to determine the
future of the UK’s relationship with the EU. Would free trade between the UK and the EU
continue? Would free labour mobility between the UK and the EU continue? And would UK
companies continue to have the right to establish subsidiaries and do business in the EU?
This report describes alternative post-Brexit futures for UK-EU relations and summarises the
economic and political consequences of each option. It starts with the alternative that
maximises economic integration between the UK and the EU and then moves to options with
successively lower degrees of integration.
As will become clear, the key trade-off that the UK would face outside the EU would be the
same trade-off that has always dominated the country’s European policy. There are economic
benefits from integration, but obtaining these benefits comes at the political cost of giving up
sovereignty over certain decisions. Inside or outside the EU, this trade-off is inescapable.
The Norwegian model – joining the European Economic Area
The European Economic Area (EEA) was established in 1994 to give European countries that
are not part of the EU a way to become members of the Single Market. The EEA comprises
all members of the EU together with three non-EU countries: Iceland, Liechtenstein and
Norway. Members of the EEA are part of the European Single Market and there is free
movement of goods, services, people and capital within the EEA. Since EEA members are
part of the Single Market, they must implement EU rules concerning the Single Market,
including legislation regarding employment, consumer protection, environmental and
competition policy.
EEA membership does not oblige countries to participate in monetary union, the EU’s
common foreign and security policy or the EU’s justice and home affairs policies. EEA
members also do not participate in the CAP. While there is free trade within the EEA, EEA
members are not part of the EU’s customs union, which means that they can set their own
external tariff and conduct their own trade negotiations with countries outside the EU.
EEA members effectively pay a fee to be part of the Single Market. They do this by
contributing to the EU’s regional development funds and contributing to the costs of the EU
programmes in which they participate. In 2011, Norway’s contribution to the EU budget was
£106 per capita, only 17% lower than the UK’s net contribution of £128 per capita (House of
Commons, 2013). Becoming part of the EEA would not generate substantial fiscal savings
for the UK government.
Joining the EEA would allow the UK to remain part of the Single Market while not
participating in other forms of European integration. An important finding of research on the
economic consequences of leaving the EU is that although Brexit would harm the UK’s
economy through reduced trade, the cost is smaller when the UK remains more economically
integrated with the EU (Ottaviano et al, 2014). Consequently, EEA membership is an
3
appealing option for those attracted by the economic benefits of the EU, but who are not in
favour of ‘ever closer union’.
There are other downsides to joining the EEA in addition to the membership fee and the need
to follow EU regulations. While EEA members belong to the Single Market, they are not part
of the deeper integration that occurs within the EU. For example, as an EEA member Norway
does not belong to the EU’s customs union. This means Norwegian exports must satisfy
‘rules of origin’ requirements to enter the EU duty-free.2
With the growing complexity of global supply chains, verifying a product’s origin has
become increasingly costly. If the UK joined the EEA, part of this cost would be borne by
UK firms. Exporters would have to limit their use of inputs imported from outside the EU to
meet the EU’s rules of origin (Stewart-Brown and Bungay, 2012). The EU can also use antidumping measures to restrict imports from EEA countries, as occurred in 2006 when the EU
imposed a 16% tariff on imports of Norwegian salmon. Campos et al (2015) find that
Norway’s failure to undertake the deeper integration pursued by EU countries has lowered
Norway’s productivity.
While these consequences of EEA membership would increase the cost of doing business
with the EU, the more important drawbacks of adopting the Norwegian model would be
political. Non-EU members of the EEA must accept and implement EU legislation governing
the Single Market without having any part in deciding the legislation. The rules of the Single
Market are set by the EU not the EEA.
By leaving the EU to join the EEA, the UK would give up its influence over all EU decisionmaking, including how to govern the Single Market. In this sense joining the EEA entails
giving up even more sovereignty than being part of the EU. EEA members must agree to
implement legislation that they have no say in deciding.
For a relatively large country such as the UK, which is accustomed to having a prominent
voice in European and world affairs, this is likely to be a difficult position to accept. For
example, the government would have no opportunity to block proposals that it believed
harmed the UK’s national interest or to drive forward policies it generally supports, such as
further liberalisation of trade in services. If a vote to leave the EU is interpreted as a vote
against giving up UK sovereignty to the EU, then joining the EEA could easily be construed
as a betrayal of the spirit of the outcome of the referendum.
The Swiss model – bilateral treaties
Switzerland is not a member of the EU or the EEA. Instead, it has negotiated a series of
bilateral treaties governing its relations with the EU. Usually, each treaty provides for
Switzerland to participate in a particular EU policy or programme. For example, among many
others, there are treaties covering insurance, air traffic, pensions and fraud prevention.
Switzerland is also a member of the European Free Trade Association (EFTA), which
provides for free trade with the EU in all non-agricultural goods.
2 ‘Rules of origin’ are used to determine whether a product originated in a free trade area and is eligible to enter
a market duty-free. The precise specifications of rules of origin are complex and variable, but typically to
benefit from free trade a product must undergo a certain level of processing within a country that belongs to the
free trade area, or a certain proportion of its value-added must come from within the free trade area.
4
The bilateral treaty approach allows Switzerland the flexibility to choose the EU initiatives in
which it wishes to participate. Through EFTA membership and an agreement covering
technical barriers to trade, Switzerland has achieved a similar level of goods market
integration with the EU as EEA countries.
Currently, there is also free movement of people between Switzerland and the EU, although
in February 2014, Switzerland voted in a referendum to impose restrictions on immigration
from the EU that would violate its agreement with the EU on free movement of people. It
remains to be seen whether or how the Swiss government will implement this vote and what
will be the consequences for Swiss-EU relations.
Switzerland and the EU have not reached a comprehensive agreement covering trade in
services. Consequently, Switzerland is not part of the Single Market for services and Swiss
financial institutions often serve the EU market through subsidiaries based in London.
As with the EEA countries, Switzerland has almost no influence over the design of the EU
programmes in which it participates. It makes an in or out choice, but has no ability to shape
the content of the programmes. The treaties require Switzerland to implement policies and
legislation set by the EU.
In this sense, Switzerland also trades integration for sovereignty and for the most part,
Switzerland has chosen to remain relatively closely integrated with the EU by accepting most
EU economic regulation. Like the EEA countries, Switzerland makes a financial contribution
to the EU to cover regional funding and the costs of the programmes in which it participates.
Switzerland’s contribution in recent years has averaged around £53 per capita, 60% lower
than the UK’s net contribution per capita (House of Commons, 2013).
Adopting the Swiss model following Brexit could be appealing if the UK is looking for an ‘à
la carte’ approach to European integration. But there are drawbacks. The EU would be under
no obligation to serve the UK everything on the menu, which means that the Swiss model
would not provide the same guarantee of market access that EU or EEA membership offer.
For example, whether the UK could reach an agreement with the EU to participate in the
Single Market in services is uncertain and exclusion from the Single Market would be
detrimental to the UK’s ability to export financial and business services to the EU.
Overall, it is likely that the Swiss model would result in less economic integration between
the UK and the EU than EEA membership, leading to higher economic costs of Brexit. The
Swiss model would also entail giving up some sovereignty, since the UK would no longer
have a say in EU decision-making, but would have to adopt EU legislation to participate in
the Single Market.
Re-joining the European Free Trade Association
Following the Norwegian or Swiss models would allow the UK to remain economically
integrated with the rest of Europe and to participate in at least some parts of the Single
Market. But a vote in favour of Brexit could lead the UK to seek a more decisive break with
the EU. When the UK opted out of joining the EEC in 1957, it founded EFTA as an
alternative. EFTA is a free trade area covering all non-agricultural goods. EFTA also has free
trade agreements with the EU and numerous other countries.
5
Re-joining EFTA would guarantee UK goods tariff-free access to the EU and ensure the UK
did not impose tariffs on goods imported from the EU. But it would not provide for free
movement of people or free trade in services between the UK and the EU. Since the UK
would not belong to the Single Market, re-joining EFTA would also probably result in a
gradual divergence between economic regulation in the UK and the EU. This would increase
‘non-tariff barriers’ to trade between the UK and the EU.3
Ottaviano et al (2014) estimate the costs of Brexit to the UK economy would come primarily
from increases in non-tariff barriers between the UK and the EU, not from changes in tariffs.
This suggests there would be an economic price to pay for joining EFTA.
In 1960, when EFTA came into being, reducing tariffs was the primary goal of efforts to
lower trade costs and promote international economic integration. But the success of the
World Trade Organization (WTO), the EU and other regional and bilateral trade agreements
in lowering tariffs has shifted the focus of today’s trade negotiations – such as the
Transatlantic Trade and Investment Partnership (TTIP) – towards non-tariff barriers and trade
in services and capital. EFTA is not designed to promote integration in these areas.
Consequently, all EFTA members have either left to join the EU or sought greater integration
with the EU through other channels.
At present, the members of EFTA are Iceland, Liechtenstein, Norway and Switzerland. All
these countries are either members of the EEA (Iceland, Liechtenstein and Norway) or have
their own bilateral agreements with the EU (Switzerland). Unless the UK wishes to opt out of
all forms of economic integration except tariff removal, re-joining EFTA is not a stand-alone
solution to the problem of what should follow Brexit.
World Trade Organization – the fallback option?
Suppose the UK leaves the EU without putting in place any of the alternative arrangements
discussed above. Then the country’s trade with both the EU and almost all the rest of the
world would be governed by the WTO. As of 2015, the WTO has 161 members comprising
all major economies and most minor ones. Under WTO rules, each member must grant the
same ‘most favoured nation’ (MFN) market access, including charging the same tariffs, to all
other WTO members. The only exceptions to this principle are that countries can choose to
enter into free trade agreements such as the EU or EFTA and can give preferential market
access to developing countries.
As a WTO member, the UK’s exports to the EU and other WTO members would be subject
to the importing countries’ MFN tariffs. Compared with EU or EFTA membership, this
would raise the cost of exporting to the EU for UK firms (Ottaviano et al, 2014). The UK’s
services trade would also be subject to WTO rules. Since the WTO has made far less progress
than the EU in liberalising trade in services, this would mean reduced access to EU markets
for UK service producers.
3 ‘Non-tariff barriers’ is a catch-all term referring to any measure that raises the costs of trade but does not take
the form of a tariff. It covers everything from quantitative trade restrictions such as import licensing to border
costs of complying with customs procedures and behind the border costs caused by regulatory or product
standard differences across countries. The EU Single Market has reduced non-tariff barriers between member
states by removing customs procedures and harmonising regulations and product standards.
6
The WTO has no provisions for free movement of labour, so under this scenario, free labour
mobility between the UK and the EU would cease. But free movement of capital between the
UK and EU would probably continue, as the EU prohibits restrictions on capital mobility not
only within the EU, but also with countries outside the EU.
After leaving the EU, the UK would no longer be bound by the EU’s common external tariff,
but would be free to set its own MFN tariffs on imports. As a starting point, the UK would be
most likely to inherit the EU’s tariff commitments, but it could then choose to reduce its
import tariffs below EU levels to lower import costs for UK consumers and firms and
increase the competition faced by UK businesses.
But since the average tariff charged on imports to the EU is only 1% (World Bank, 2015),
there is limited scope for further tariff reductions. There is also limited scope to lower nontariff barriers through unilateral action since reducing non-tariff barriers often requires
harmonising policies, regulations or product standards across countries, which requires
international agreement.
The pay-off for the lack of economic integration would be greater political sovereignty.
Being outside the Single Market would enable the UK government to set economic policy
and regulatory standards without taking account of the preferences of other EU members. But
any divergence in regulation between the UK and the EU would still act as a non-tariff barrier
to trade and raise the cost of doing business with Europe.
Overall, it is uncertain how leaving the Single Market would affect the UK’s economic
policies and regulations and whether any changes would be beneficial. The OECD has found
that, even as a member of the Single Market, the UK’s labour and product markets are
substantially less regulated and more flexible than those of other EU countries (Koske et al,
2015).
In fact, the UK’s labour and product markets exhibit similar levels of flexibility to Canada
and the United States and are much less regulated than those of non-EU countries such as
Norway and Switzerland. This shows that the Single Market does provide scope for countries
to adapt economic regulations to suit national preferences.
Table 1 summarises the costs and benefits of the alternative paths the UK could follow after
leaving the EU.
Looking away from Europe
EU members have a common trade policy and are represented by the EU in all international
trade negotiations. After Brexit, the UK would become an independent player in trade
negotiations. This means that the country would not only need to determine its future
economic relations with the EU, but also with the rest of the world. Instead of entering into
new agreements with the EU following Brexit, the UK could opt to seek closer integration
with countries outside Europe.
7
Table 1: Options for the UK outside the EU

8
For example, the UK could propose a free trade area among Commonwealth countries or
could attempt to join Canada, Mexico and the United States as a member of the North
American Free Trade Agreement (NAFTA). Of course, the EU is also working to dismantle
trade barriers with the rest of the world, such as through the TTIP agreement currently being
negotiated with the United States. It is uncertain whether leaving the EU would enable the
UK to negotiate more and better trade agreements than it can as part of the EU.
Even without the UK, the EU is the world’s second largest exporter behind China and the
world’s second largest importer behind the United States. This makes the EU a desirable trade
partner and gives the EU an important voice in trade negotiations. Since the UK is a much
smaller market than the EU, the country alone would have less bargaining power in
international trade negotiations than the EU currently has.
On the other hand, Brexit would enable the UK to seek trade agreements tailored to the
interests of UK businesses and consumers rather than having to make compromises to meet
the needs of other EU countries.
Whether the benefits from greater autonomy in trade negotiations would outweigh the costs
from reduced bargaining power is hard to predict, but some insight into how the UK may fare
following Brexit can be gained by looking at the experience of Canada – another mediumsized developed economy in close proximity to a much larger market.
Under NAFTA, there is free trade between Canada, Mexico and the United States, but one of
the costs of obtaining access to the US market is adoption of the provisions of the ‘investment
state dispute settlement’ (ISDS). ISDS clauses are almost always included in US trade
agreements (Poulsen et al, 2013) and they allow US investors to bring claims directly against
the Canadian government (and vice versa). By contrast, under the WTO’s dispute settlement
mechanism, investors must go through their home government to bring a claim against
another country.
Cases brought against Canada under the ISDS have covered issues such as the decision to
introduce plain packaging of tobacco products and ‘anti-graft’ rules that would restrict
companies convicted of corruption from receiving government contracts. This has raised
concerns that ISDS clauses provide too much protection to foreign investors and effectively
curtail national sovereignty.
There is also evidence that US firms are better able to take advantage of ISDS provisions than
Canadian firms. The United States has won all of the 11 decided cases that it has initiated
under the ISDS, while Canada has won seven of its 13 decided cases (CCPA, 2015).
The scope of ISDS provisions is a key point of contention in the TTIP negotiations between
the United States and the EU. The EU has sufficient bargaining power to push back against
rules designed to advance the interests of US firms. It is unlikely that the UK alone would
have similar leverage.
Reducing trade barriers between the UK and the rest of the world is a laudable aim and would
be likely to increase trade and raise UK income. But it is not an adequate replacement for EU
membership. The best-known fact in international economics is that international trade and
investment fall substantially with distance (Head and Mayer, 2014). Doubling the distance
between two countries roughly halves the trade between them. The UK is much closer
9
geographically to the EU than to other large economies such as the United States or China
and, therefore, it is not surprising that roughly half of the UK’s trade is with the EU
(Ottaviano et al, 2014).
Put another way, it is geography rather than policy that makes the EU the UK’s most
important economic partner. Simply reorienting the focus of the UK’s trade policy away from
Europe will not change this underlying reality. Whatever agreements are reached with
countries outside Europe, the most important decision facing the government following
Brexit would still be the future of the UK’s relations with the EU.
Conclusions
A vote in favour of Brexit will fire the starting gun on a two-year renegotiation of the UK’s
place in Europe and the world. If the UK opts to cut ties with the rest of Europe, this
renegotiation could fundamentally change the political, economic and legal foundations of
UK life that have built up since the country joined the EU in 1973. Alternatively, if the UK
chooses to remain part of the EEA, the economic and legal changes would be much smaller.
During the renegotiation, the UK would face an unavoidable trade-off between economic
benefits and political sovereignty. The UK benefits from closer economic integration with the
EU, but the price for this integration is allowing the EU control over some areas of policy.
Leaving the EU will not free the UK from this fundamental trade-off.
At present, there is no consensus within the government or the public over what should
follow Brexit. This reflects the fact that all of the alternatives to EU membership have their
own drawbacks and would impose costs on the UK economy. To make an informed decision
about the merits of Brexit, voters need to know as much as possible about what Brexit would
mean.
To date, neither the Conservative government nor the Labour opposition have released
proposals for the UK’s future if the country votes to leave the EU. Just as the parties put
forward policy manifestos in the run-up to an election, they should publish their plans for a
post-Brexit world before the referendum. The alternative scenarios discussed in this report
embody very different visions of the UK’s future place in the world. The country’s voters
have the right to know what they are choosing between when they enter the polling booth.
February 2016
For further information, contact:
Swati Dhingra ([email protected]) or Thomas Sampson ([email protected]).
10
Further reading
Cameron, D. (2013) ‘EU Speech at Bloomberg’, 23 January 2013. Retrieved from:
https://www.gov.uk/government/speeches/eu-speech-at-bloomberg
Campos, N., F. Coricelli and L. Moretti. (2015) ‘Norwegian Rhapsody? The Political
Economy Benefits of Regional Integration’, CEPR Discussion Paper 10653.
CCPA (2015) ‘NAFTA Chapter 11 Investor-State Disputes’, 14 January 2015, Canadian
Centre for Policy Alternatives.
Head, K. and T. Mayer (2014) ‘Gravity Equations: Workhorse, Toolkit, Cookbook’,
Handbook of International Economics Vol. 4.
House of Commons (2013) ‘Leaving the EU’, Research Paper 13/42, 1 July 2013.
Koske, I., I. Wanner, R. Bitetti and O. Barbiero (2015) ‘The 2013 Update of the OECD
Product Market Regulation Indicators: Policy Insights for OECD and Non-OECD Countries’,
OECD Economics Department Working Papers, 1200/2015.
Ottaviano, G., J. P. Pessoa, T. Sampson and J. Van Reenen (2014) ‘The Costs and Benefits of
Leaving the EU’, Centre for Economic Performance Policy Analysis
http://cep.lse.ac.uk/pubs/download/pa016.pdf
Parker, G. (2015). ‘Tories Shun Brexit Contingency Plans’, Financial Times, 1 December
2015. Retrieved from: http://www.ft.com/cms/s/0/208fdf8c-9846-11e5-95c7-
d47aa298f769.html#axzz3xSEYNfkq
Philippidis, G. and L. J. Hubbard (2001) ‘The Economic Cost of the CAP Revisited’,
Agricultural Economics, 25, 375-385.
Poulsen, L., J. Bonnitcha and J. Yackee (2013) ‘Costs and Benefits of an EU-USA
Investment Protection Treaty’, LSE Enterprise Report to the BIS, April 2013.
Stewart-Brown, R. and F. Bungay (2012) ‘Rules of Origin in EU Free Trade Agreements’,
Trade Policy Centre Research Paper.
Ugwumadu, J. (2013) ‘Poland Takes Lion’s Share of EU Funds’, Public Finance
International, 28 November 2013. Retrieved from:
http://www.publicfinanceinternational.org/news/2013/11/poland-takes-lion%E2%80%99sshare-eu-funds
World Bank (2015) ‘World Development Indicators’.
11
The consequences of Brexit for
UK trade and living standards
Swati Dhingra, Gianmarco Ottaviano,
Thomas Sampson and John Van Reenen
#CEPBREXIT
BREXITPAPER
Introduction
The outcome of the UK’s referendum on membership of the European Union (EU) will shape
the future of the country’s relationship with its largest trade partner – the EU. Membership of
the EU has reduced trade costs between the UK and the rest of Europe. Most obviously, there
is a customs union between EU members, which means that all tariff barriers have been
removed within the EU, allowing for free trade in goods and services.
But equally important in reducing trade costs has been the reduction of non-tariff barriers
resulting from the EU’s continuing efforts to create a ‘single market’ within Europe.1 Nontariff barriers include a wide range of measures that raise the costs of trade such as border
controls, rules of origin checks, cross-country differences in regulations over things like
product standards and safety, and threats of anti-dumping.
Reductions in trade barriers have increased trade between the UK and the EU. Prior to the
UK joining the European Economic Community (EEC) in 1973, around one third of UK
trade was with the EEC. In 2014, the 27 other EU members accounted for 45% of the UK’s
exports and 53% of our imports (ONS, 2015). EU exports comprise 13% of UK national
income.
Higher trade benefits UK consumers through lower prices and access to better goods and
services. At the same time, the UK’s workers and businesses benefit from new export
opportunities that lead to higher sales and profits and allow the UK to specialise in industries
in which it has a comparative advantage. Through these channels, increased trade raises
output, incomes and living standards in the UK.
These standard ‘static’ effects of trade have been understood for many centuries since at least
the work of David Ricardo. But in recent decades, studies of trade have revealed very large
effects on wellbeing through other routes such as higher productivity and innovation.
How would Brexit affect the UK’s trade, and what impact would this have on incomes in the
UK? This briefing reports new estimates of how Brexit would affect UK living standards
through trade (updating our earlier analysis in Ottaviano et al, 2014). We report a range of
forecasts based on alternative estimation methods and different assumptions about how the
UK’s relationship with the EU would change following Brexit. We primarily focus on the
narrow, static trade consequences of Brexit rather than other channels through which Brexit
could affect the UK’s economy, such as investment or migration.
Although it is always hard to assess what the economic future may bring and there are many
uncertainties, we consistently find that by reducing trade, Brexit would lower UK living
standards. Importantly, the fall in income per capita resulting from lower trade more than
offsets any savings that the UK obtains from reduced fiscal contributions to the EU budget.
Our baseline estimates imply that, after accounting for fiscal savings, the effect of Brexit is
equivalent to a fall in UK income of between 1.3% and 2.6% – that is, a decline in average
annual household income of between £850 and £1,700 per year.
1 The single market is the name given to the integrated European economy created by removing economic
barriers between EU members.
13
Our baseline estimates come from a state-of-the-art static model of the global economy. We
also present estimates using empirical evidence on the links between EU membership, trade
and income. This ‘reduced-form’ approach captures the long-run effects of leaving the EU on
productivity growth and leads to much higher estimates. In this case, we calculate that Brexit
may reduce national income by between 6.3% and 9.5% – that is, about £4,200 to £6,400 per
household per year.
We abstract away from the cost of the policy uncertainty that will result from the negotiations
over Brexit. The impact of such uncertainty has been found to be important in much recent
research (Handley and Limão, 2015).
Estimating the effects of Brexit
To estimate the effect of Brexit on the UK’s trade and living standards, we use a modern
quantitative trade model of the global economy. Quantitative trade models incorporate the
channels through which trade affects consumers, firms and workers, and provide a mapping
from trade data to welfare. The model provides numbers for how much real incomes change
under different trade policies, using readily available data on trade volumes and potential
trade barriers. Our model uses the most recent data (WIOD) which divides the world into 35
sectors and 31 regions. It allows for trade in both intermediate inputs and final output in both
goods and services. The model takes into account the effects of Brexit on the UK’s trade with
the EU and the UK’s trade with the rest of the world.
To forecast the consequences of the UK leaving the EU, we must make assumptions about
how trade costs change following Brexit. It is not known exactly how the UK’s relations with
the EU would change following Brexit, which means that there is a lack of clarity over the
consequences of Brexit for trade costs between the UK and the EU.
To overcome this difficulty, we analyse two scenarios: an optimistic scenario in which the
increase in trade costs between the UK and the EU is small, and; a pessimistic scenario with a
larger rise in trade costs.
The optimistic scenario assumes that in a post-Brexit world, the UK’s trade relations with the
EU are similar to those currently enjoyed by Norway. As a member of the European
Economic Area (EEA), Norway has a free trade agreement with the EU, which means that
there are no tariffs on trade between Norway and the EU. Norway is also a member of the
European single market and adopts policies and regulations designed to reduce non-tariff
barriers within the single market.
But Norway is not a member of the EU’s customs union, so it faces some non-tariff barriers
that do not apply to EU members such as rules of origin requirements and anti-dumping
duties. Campos et al (2015) find that Norway’s productivity growth has been harmed by not
fully participating in the EU’s market integration programmes.
In the pessimistic scenario, we assume that the UK is not successful in negotiating a new
trade agreement with the EU and, therefore, that trade between the UK and the EU following
Brexit is governed by World Trade Organisation (WTO) rules. This implies larger increases
14
in trade costs than the optimistic scenario because most favoured nation (MFN) tariffs2 are
imposed on UK-EU trade and because the WTO has made less progress on reducing nontariff barriers than the EU.
Increases in trade costs between the UK and the EU following Brexit can be divided into
three parts: (i) higher tariffs on imports; (ii) higher non-tariff barriers to trade (arising from
different regulations, border controls, etc.); and (iii) the UK may not participate in future
steps that the EU takes towards deeper integration and the reduction of non-tariff barriers
within the EU.
In the optimistic scenario, we assume that the UK and the EU continue to enjoy a free trade
agreement and Brexit does not lead to any change in tariff barriers. In the pessimistic scenario
where trade is governed by WTO rules, we assume MFN tariffs are imposed on UK-EU
goods trade.
Regarding non-tariff barriers, in the optimistic scenario, we assume that UK-EU trade is
subject to one quarter of the reducible non-tariff barriers that are observed in trade between
the United States and the EU. In the pessimistic scenario, we assume a larger increase of
three quarters of reducible non-tariff barriers.3
Finally, trade costs between countries within the EU have been declining approximately 40%
faster than trade costs between other OECD countries (Méjean and Schwellnus, 2009). In the
event of Brexit, the UK would not benefit from any future reductions in intra-EU trade costs.
In the optimistic scenario, we assume that in the ten years following Brexit, intra-EU trade
costs fall 20% faster than in the rest of the world, while in the pessimistic scenario, we
assume intra-EU trade costs continue to fall 40% faster than in the rest of the world. This
implies that in the optimistic case, non-tariff barriers within the EU fall 5.7% over the next
decade, while in the pessimistic case they fall by 12.8%.4
Our estimates also account for fiscal transfers between the UK and the EU. Like all EU
members, the UK makes a contribution to the EU budget. The net fiscal contribution of the
UK to the EU budget has been estimated to be around 0.53% of national income (HM
Treasury, 2013). One benefit of Brexit for the UK would be a reduced contribution to the EU
budget.
But Brexit would not necessarily mean that the UK would make zero contribution to the EU
budget. In return for access to the single market, EEA members such as Norway make
substantial payments to the EU. On a per capita basis, Norway’s financial contribution to the
EU is 83% as large as the UK’s payment (House of Commons, 2013). Therefore, in the
optimistic case we assume that the UK’s contribution to the EU budget falls by 17% (that is,
0.09% of national income).
2
Under WTO rules, each member must grant the same ‘most favoured nation’ (MFN) market access, including
charging the same tariffs, to all other WTO members. The only exceptions to this principle are that countries can
choose to enter into free trade agreements such as the EU or the European Free Trade Association and can give
preferential market access to developing countries.
3 These assumptions imply a non-tariff barrier increase of 2.0% in the optimistic scenario and 6.0% in the
pessimistic scenario. Our data on non-tariff barriers between the United States and the EU are taken from
Berden et al (2009, 2013).
4 See Dhingra et al (2016) for a complete explanation of how these changes are calculated.
15
In the pessimistic case where the UK is outside the EEA, we assume that the UK saves more
of its current contribution. The 0.53% saving includes only the public finance components so
excludes all the transfers the EU makes directly to universities, firms and other nongovernmental bodies. Under the reasonable assumption that post-Brexit the UK government
does not cut this funding, the saving is 0.31% according to Eurostat
(http://ec.europa.eu/budget/figures/2007-2013/index_en.cfm).5 This cost essentially comes
from the agricultural subsidies in the Common Agricultural Policy.
Table 1 summarises the results of our analysis. For each case, we calculate the percentage
change in the level of income per capita that has the same effect on living standards in the
UK as Brexit.6 The numbers we report should be interpreted as permanent changes in average
income per capita in the UK that occur immediately following Brexit.
In the optimistic scenario, there is an overall fall in income of 1.28% that is largely driven by
current and future changes in non-tariff barriers. Non-tariff barriers play a particularly
important role in restricting trade in services, an area where the UK is a major exporter. In the
pessimistic scenario, the overall loss increases to 2.61%.
The costs of reduced trade far outweigh the fiscal savings in both scenarios. In cash terms, the
cost of Brexit to the average UK household is £850 per year in the optimistic scenario and
£1,700 per year in the pessimistic scenario.
Table 1: The effects of Brexit on UK living standards
Optimistic Pessimistic
Trade effects -1.37% -2.92%
Fiscal benefit 0.09% 0.31%
Total change in income per capita -1.28% -2.61%
Income change per household -£850 -£1,700
Source: CEP calculations (see Dhingra et al, 2016, for technical details).
Notes: Optimistic scenario: Increase in EU/UK Non-Tariff Barriers (+2%) + exclusion from future fall in NTB
within EU (-5.7%), saving of 17% of 0.53% lower fiscal transfer. Pessimistic scenario: MFN Tariff + increase
in EU/UK Non-Tariff Barriers (+6%) + exclusion from future fall in NTB within EU (-12.8%), saving of 0.31%
net fiscal transfer.
The effect of Brexit on other countries
Although we have focused on the UK, the fall in trade also affects other countries. Figure 1
shows the distribution of changes in income per capita across countries in the optimistic and
pessimistic scenarios. All EU members are worse off: Ireland suffers the largest proportional
losses from Brexit, alongside the Netherlands and Belgium. Countries that lose the most are
those currently trading the most with the UK. Some countries outside the EU, such as Russia
and Turkey, gain as trade is diverted towards them and away from the EU.
5 Note that we are overstating the benefits of Brexit in the optimistic scenario by using the higher 0.53%
number. But we do not have accurate calculations on the comparable fraction of the 0.31% net fiscal
contribution for Norway.
6 Formally, we calculate the permanent percentage change in income per capita that has the same present
discounted value effect on welfare in the UK as Brexit. We assume an annual discount rate of 4% and an
intertemporal elasticity of substitution equal to one.
16
Altogether the EU loses between -0.12% and -0.29% of its GDP which is offset by a 0.01%
to 0.02% gain for non-EU countries. These seem small percentages, but the rest of the
world’s GDP is, of course, much bigger than that of the UK. So whereas the UK loses
between £26 billion to £55 billion from Brexit the rest of the EU is collectively £12 billion to
£28 billion worse off.7 The ‘Brexit shock’ is almost half as big in the rest of the EU as it is in
the UK.
Figure 1: The effect of Brexit on living standards across countries
Source: CEP calculations (see Dhingra et al, 2016, for technical details).
Notes: Same assumptions as in notes to Table 1 except net fiscal savings not included (since we do not know
how Brexit would affect the budget contributions of other EU members).
A Swiss alternative?
Switzerland is not in the EEA but has many bilateral agreements with the EU, which gives it
some access to the single market. Like Norway, it has to adopt all the regulations covering
those parts of the single market in which it participates and it allows free movement of
labour. It does, however, benefit from a lower fiscal transfer to the EU (about 40% of the
UK’s contribution on a per capita basis). On the other hand, it does not have free trade in
services with the EU, which would be a disadvantage for an economy like the UK, which has
a comparative advantage in services.
We simulate the effects of Brexit using Switzerland as an alternative optimistic scenario. The
results are very similar: a loss of income of 1.30%. Although the fiscal transfers are lower
than for Norway, these are more than offset by higher costs of trade in services.
7 These calculation use IMF GDP estimates for 2014: world GDP $77.3tr; EU $18.5tr and UK $3tr.
https://en.wikipedia.org/wiki/List_of_countries_by_GDP_%28nominal%29.
17
Unilateral liberalisation after Brexit?
Following Brexit, the UK would no longer be bound by the EU’s common external tariff on
imports. Proponents of leaving the EU argue the UK could benefit from this change by
unilaterally removing all tariffs on imports into the UK in order to lower the cost of imported
goods. To analyse the consequences of this unilateral liberalisation policy, we re-run our
optimistic and pessimistic scenarios after including the additional assumption that the UK
removes all tariffs on imports from anywhere in the world.
Table 2 reports the results. We find that unilateral liberalisation reduces the costs of Brexit by
0.3 percentage points in both scenarios. But the overall effect of Brexit is still negative. The
reason that the benefits of such a radical move are small is simple. WTO tariffs are already
low, so further reductions do not make much difference. In today’s world, integration is not a
matter of lowering tariff rates. It requires policies, such as hammering out regulatory
differences in services provision that rely on international agreement and cannot be achieved
unilaterally.
Table 2: The effects of Brexit and unilateral trade liberalisation on UK living standards

Source: CEP calculations (see Dhingra et al, 2016, for technical details).
Notes: This includes simulating the unilateral removal of all tariffs on imports into the UK.
Long-run effects of Brexit
The estimates in Table 1 are based on a static trade model that does not account for the
dynamic effects of trade on productivity. Trade can have positive effects through increasing
competition, which reduces excess profits and promotes efficiency. Competition, access to
superior intermediate goods and a larger export market can also stimulate innovation. Recent
research finds that dynamic effects may double or triple the size of the static effects reported
in Table 1 (Bloom et al, 2014; Sampson, 2016).
An alternative way to evaluate the consequences of Brexit is to use the results of reducedform empirical studies of the effects of EU membership. Baier et al (2008) find that after
controlling for other determinants of bilateral trade, EU members trade substantially more
with other EU countries than they do with members of the EEA or EFTA. Their estimates
imply that, if the UK leaves the EU and joins EFTA, its trade with countries in the EU will
fall by about a quarter.
Combining this with estimates that a 1% decline in trade reduces income per capita by
between 0.5% and 0.75% (Feyrer, 2009) implies that leaving the EU and joining EEA would
reduce UK income per capita by between 6.3% and 9.5% (£4,200 to £6,400 per household
per year). These estimates are much higher than the costs obtained from the static trade
model, suggesting that the dynamic gains from trade may be important.
18
Interestingly, these larger long-run effects are in the same ballpark as the benefits that the UK
has gained since 1973 from being part of the EU. In a recent survey of the evidence of the
impact of EU membership, Crafts (2016) concludes that EU membership raised UK GDP per
capita by between 8.6% and 10.6%. Economists under-estimated the benefits from EU
membership because they focused on static trade models of the kind we have employed in
Table 1.
The bottom line is that the costs of Brexit could easily be about three times larger than those
in the static analysis shown in Table 1.
Future trade agreements
EU members have a common trade policy and are represented by the EU in all international
trade negotiations. After Brexit, the UK would become an independent player, free to seek its
own trade deals with the rest of the world. The UK could use this freedom to look for new
trade deals with countries such as China, India and the United States.
Our model shows that trade with such non-EU countries does indeed rise after Brexit. But the
magnitude of these increases is not enough to offset the decline in trade with the EU. Being
part of the EU does not restrict UK companies’ ability to trade with the rest of the world, but
as our nearest neighbour and the world’s largest market, the EU is the UK’s natural trade
partner.
When negotiating post-Brexit trade deals, the UK would not need to compromise with other
EU countries. On the other hand, the UK would have to take on the cost of hiring civil
servants to rebuild its capacity to undertake trade negotiations. More importantly, since the
UK is a smaller market than the EU, it would have less bargaining power in trade
negotiations than the EU does.
Has the UK benefited from past trade deals reached by the EU? CEP research by Breinlich et
al (2016) estimates that trade agreements negotiated by the EU over the past two decades
have reduced the quality-adjusted prices of imports into the UK by over one-third. Although
it is often argued that the EU does not pursue trade agreements that are beneficial to the UK,
these consumer benefits are twice as big as those enjoyed by the 12 other members that
joined before 1995.
The EU is currently negotiating a major new free trade agreement with the United States (the
Transatlantic Trade and Investment Partnership or TTIP) – as well as an ‘economic
partnership agreement’ with Japan. If the UK leaves the EU, it will not benefit from these.
Breinlich et al (2016) estimate that the US and Japanese agreements would lower prices by
0.4% and 0.2% respectively. The United States has stated that it would not do a trade deal
with the UK alone (Holehouse, 2015).
Other Brexit effects on migration, foreign investment and regulation
We have focused on the impact of Brexit on UK households through trade. Brexit could also
affect the UK economy through changes in investment, migration and regulation. We will
19
examine these channels more closely in future reports, but one way of interpreting our
findings is that for Brexit to have an economic benefit, these channels must have a
sufficiently large positive effect on the UK economy to outweigh the negative effects we
identify. This is extremely unlikely to be the case.
Brexit is likely to reduce foreign investment, which has been found to lead to higher
productivity – for example, Haskel et al (2002). Pain and Young (2004) estimate that EU
membership adds 2.25% to UK GDP via the channel of foreign direct investment. Similarly,
migration is found to aid growth and help to reduce the budget deficit without serious adverse
labour market effects (Wadsworth, 2015).
Eurosceptics often point to the promise of better and less regulation as a big benefit after
Brexit (for example, Minford, 2015). It is important to realise that regulation will not much
affect the optimistic scenario. This is because to access the single market, countries like
Norway or Switzerland must adopt the same regulations as the rest of the EU (without having
a vote on what these regulations are).
The UK could weaken social, employment and environmental regulation to some degree. But
even if this were politically possible, the UK already has one of the most flexible
employment and product market regulations in the world according to the OECD (second in
product regulation to the United States and third to the United States and Canada in labour
regulations). Even if the GDP impact of such regulations were large (a point on which there
is controversy), further weakening protection to say US levels would make little economic
difference.
If the UK were to accept higher trade costs by giving up high levels of access to EU markets
(the ‘pessimistic scenario’ above), there would be more scope for regulatory loosening. Booth
et al (2015) identify 56 regulations derived from EU legislation where the UK government’s
Impact Assessment finds that the costs outweigh the benefits. Crafts (2016) calculates the
cost of these regulations is 0.9% of the UK’s GDP.
But many of these regulations implement policies that the UK government is committed to
following inside or outside the EU. For example, half of the total cost comes from just two
policies: the Renewable Energy Strategy; and the Working Time Directive. Scrapping these
regulations would mean abandoning the UK’s renewable energy targets and removing rights
such as the entitlement to 20 days paid annual leave.
Even if the regulatory costs of EU membership were 0.9% of GDP, this figure is still less
than half as large as our estimates of the net cost of Brexit even in the purely static case, and
a lot less than the 6.3% to 9.5% costs under the dynamic case. There are many costs of
regulation in the UK, such as our inefficient planning system (as explained, for example, by
the LSE Growth Commission, 2013). But these problems are primarily home-grown, rather
than imports from Brussels.
Conclusions
The economic consequences of leaving the EU will depend on what policies the UK adopts
following Brexit. But lower trade due to reduced integration with EU countries is likely to
cost the UK economy far more than is gained from lower contributions to the EU budget.
20
Even setting aside foreign investment, migration and the dynamic consequences of reduced
trade, we estimate the effects of Brexit on trade and the UK’s contribution to the EU budget
would be equivalent to a fall in income of between 1.3% and 2.6% (£850 to £1,700 per
household per year). And once we include the long-run effects of Brexit on productivity, the
decline in income increases to between 6.3% and 9.5% (about £4,200 to £6,400 per
household per year).
March 2016
For further information, contact:
Swati Dhingra ([email protected]), Gianmarco Ottaviano ([email protected]),
Thomas Sampson ([email protected]), John Van Reenen ([email protected]) or
Romesh Vaitilingam on 07768-661095 ([email protected]).
21
Further reading
Baier, S. L., J. H. Bergstrand, P. Egger and P. A. McLaughlin (2008) ‘Do Economic
Integration Agreements Actually Work? Issues in Understanding the Causes and
Consequences of the Growth of Regionalism’, The World Economy 31(4): 461-97.
Berden, K., J. Francois, S. Tamminen, M. Thelle and P. Wymenga (2009) ‘Non-Tariff
Measures in EU-US Trade and Investment – An Economic Analysis’, Ecorys report prepared
for the European Commission, Reference OJ 2007/S180219493.
Berden, K., J. Francois, K. Tamminen, M. Thelle and P. Wymenga (2013) ‘Non-tariff
Barriers in EU-US Trade and Investment: An Economic Analysis’, Technical Report,
Institute for International and Development Economics.
Bloom, N., P. Romer, S. Terry and J. Van Reenen (2014) ‘A Trapped Factors Model of
Innovation’, Centre for Economic Performance Discussion Paper No. 1261
(http://cep.lse.ac.uk/pubs/download/dp1261.pdf).
Booth, S., C. Howarth, M. Persson, R. Ruparel and P. Swidlicki (2015) ‘What If..? The
Consequences, Challenges and Opportunities facing Britain outside the EU’, London: Open
Europe.
Breinlich, H., S. Dhingra and G. Ottaviano (2016) ‘The Impact of Trade Agreements on
Consumers’, Centre for Economic Performance mimeo.
Campos, N., F. Coricelli and L. Moretti. (2015) ‘Norwegian Rhapsody? The Political
Economy Benefits of Regional Integration’, CEPR Discussion Paper No. 10653.
Crafts, N. (2016) ‘The Growth Effects of EU Membership for the UK: A Review of the
Evidence’, University of Warwick mimeo.
Dhingra, S., H Huang, G Ottaviano, J Pessoa, T Sampson and J Van Reenen (2016) ‘The
Costs and Benefits of Leaving the EU’, Centre for Economic Performance Technical Report.
Feyrer, J. (2009) ‘Trade and Income – Exploiting Time Series in Geography’, NBER
Working Paper No. 14910.
Handley, K. and N. Limão (2015) ‘Trade and Investment under Policy Uncertainty: Theory
and Firm Evidence’, American Economic Journal: Economic Policy 7(4): 189-222.
Haskel, J., S. Pereira and M. Slaughter (2002) ‘Does Inward Foreign Direct Investment Boost
the Productivity of Domestic Firms?’, NBER Working Paper No. 8724.
HM Treasury (2013) ‘European Union Finances 2013’, 19th November.
Holehouse, M. (2015) ‘Major Blow for Brexit Campaign as US Rules Out UK-Only Trade
Deal’, The Telegraph, 29 October 2015
(retrieved from: http://www.telegraph.co.uk/news/worldnews/europe/eu/11962277/Majorblow-for-Brexit-campaign-as-US-rules-out-UK-only-trade-deal.html).
22
House of Commons (2013) ‘Leaving the EU’, Research Paper 13/42, 1st July.
LSE Growth Commission (2013) ‘Investing for Prosperity’
(http://www.lse.ac.uk/researchAndExpertise/units/growthCommission/documents/pdf/LSEG
C-Report.pdf).
Méjean, I. and S. Schwellnus (2009) ‘Price Convergence in the European Union: Within
Firms or Composition of Firms?’, Journal of International Economics 78(1): 1-10.
Minford, P. (2015) ‘Evaluating European Trading Arrangements’, Cardiff Economics
Working Paper No. E2015/17.
ONS (2015) ‘How Important is the European Union to UK Trade and Investment?’, Office
for National Statistics, 26 June.
Ottaviano, G., J. Pessoa, T. Sampson and J. Van Reenen (2014) ‘The Costs and Benefits of
Leaving the EU’, Centre for Economic Performance Policy Analysis.
Pain, N. and G. Young (2004) ‘The Macroeconomic Effect of UK Withdrawal from the EU’,
Economic Modelling 21: 387-408.
Sampson, T. (2016) ‘Dynamic Selection: An Idea Flows Theory of Entry, Trade and
Growth’, Quarterly Journal of Economics 131(1): 315-80, 131(1): 315-380.
Wadsworth, J. (2015) ‘Immigration and the UK Labour Market’, Centre for Economic
Performance Election Analysis No. 1 (http://cep.lse.ac.uk/pubs/download/EA019.pdf).
23
The impact of Brexit on foreign
investment in the UK
Swati Dhingra, Gianmarco
Ottaviano, Thomas Sampson and
John Van Reenen
#CEPBREXIT
PAPERBREXIT
Introduction
Foreign direct investment (FDI) comprises investments from outside a country to start up new
subsidiaries, to expand existing establishments or to acquire local companies. The UK is a
major recipient of FDI with an estimated stock value of over £1 trillion, about half of which is
from other members of the European Union (EU), according to UK Trade and Investment
(UKTI, 2015). Only the United States and China receive more FDI than the UK.
Countries generally welcome FDI as it tends to raise productivity, which increases output and
wages. FDI brings direct benefits as foreign firms are typically more productive and pay higher
wages than domestic firms. But FDI also brings indirect benefits as the new technological and
managerial know-how in foreign firms can be adopted by domestic firms, often through
multinationals’ supply chain (Harrison and Rodriguez-Clare, 2009). FDI can also increase
competitive pressure, which forces managers to improve their performance.
There are at least three reasons why FDI might fall if the UK left the EU:
• First, being fully in the Single Market makes the UK an attractive export platform for
multinationals as they do not bear potentially large costs from tariff and non-tariff
barriers when exporting to the rest of the EU.
• Second, multinationals have complex supply chains and many co-ordination costs
between their headquarters and local branches. These would become more difficult to
manage if the UK left the EU. For example, component parts would be subject to
different regulations and costs; and intra-firm staff transfers would become more
difficult with tougher migration controls.
• Third, uncertainty over the shape of the future trade arrangements between the UK and
the EU would also tend to dampen FDI.
This report analyses what could happen to UK FDI inflows after Brexit and what effect these
FDI changes could have on income levels in the UK.
The effect of EU membership on FDI
A number of factors determine where firms choose to locate and invest. Bigger and richer
markets tend to attract more firms, which want to be close to their customers. The UK has
strong rule of law, flexible labour markets and a highly educated workforce, all of which make
it an attractive FDI location whether or not it is in the EU. But since EU membership reduces
trade and investment costs, it is likely to have an impact even after controlling for these other
factors.
To estimate the size of the effect of being in the EU on FDI, we provide a new empirical
analysis – see Bruno et al (2016) in the Technical Appendix to this report. It is a statistical
model based on the bilateral FDI flows between 34 OECD countries from 1985 to 2013. The
model estimates why foreign investors choose to invest in the UK, as opposed to other countries
such as Germany, France or the United States. It is similar to the ‘gravity model’ that is the
standard way of estimating bilateral flows of exports and imports.
25
Bilateral FDI flows between any pair of countries depend on their respective market size
(measured by GDP), the geographical distance between them and other factors such as GDP
per capita. The model addresses the question of how much more FDI would flow between two
countries if the sender or the recipient joins the EU, once all these factors are taken into account.
Since many FDI determinants – such as geographical distance and culture – are broadly stable
over time, we can fully control for them by looking only at changes in FDI and its determinants.
The data show that there is always a statistically significant positive effect of being in the EU
on inward FDI. The magnitude ranges from a 14% to 38% increase in FDI depending on the
exact statistical method used with an average of 28% across the main three methods.
These estimates are also consistent with those in Campos and Coricelli (2015), who find an
impact of 25% to 30% on FDI flows from EU membership using an alternative method
comparing the evolution of UK FDI with a set of matched countries as a comparison group.
Similarly, Straathof et al (2008) find that EU membership increases inward FDI stocks by 14%
from non-EU countries and by 28% from other EU members (using a gravity model but with
earlier data).
Being a member of the European Free Trade Association (EFTA) like Switzerland does not
restore the FDI benefits of being in the EU. In fact, we find no statistical difference between
being in EFTA compared with being completely outside the EU like the United States or Japan.
So striking a comprehensive free trade deal after Brexit is not a good substitute for full EU
membership.
By comparison, Baier et al (2008) estimate that EU membership leads to trade with other EU
members increasing by a quarter or more (compared with EFTA membership). So the
magnitude of the FDI effect on Brexit is in the same ballpark as the effect on overall trade.
Since leaving the EU will likely have a smaller proportionate effect than joining, our Technical
Appendix concludes that Brexit is likely to reduce FDI inflows to the UK by about 22%.1
How do changes in FDI affect UK incomes?
There is much evidence that FDI brings benefits in terms of enhanced productivity. For
example, Bloom et al (2012) find that multinationals boost productivity in UK establishments
through enhanced technologies and management practices. On top of this direct effect, Haskel
et al (2007) find that there are foreign investment ‘spillovers’ to other, UK-owned firms in the
same industry.
But to get at the nation-wide impact of FDI on output, we need to factor in the many complex
ways in which FDI affects people and firms in multiple parts of the economy. This is a tricky
task, but fortunately we can draw on the work of Alfaro et al (2004), who estimate the effect
of changes in FDI on growth rates across 73 countries. They find that increases in FDI have a
large positive impact on GDP growth, especially for countries like the UK that have a highly
developed financial sector.
1 Using a baseline estimate of 0.28, we obtain 0.22 = 0.28/(1+0.28). This is very similar to PWC (2016), which
finds that UK FDI will be a quarter lower in 2020 because of Brexit.
26
To be very conservative, we assume a scenario where the Brexit-induced fall in FDI lasts only
for 10 years and then reverts to its current level. Using the average of the estimates in the
Technical Appendix combined with Alfaro et al’s estimates implies a fall in real income of
about 3.4% (see the Annex for more detail). Looking at the wider range, we obtain a fall in
income of between 1.8% and 4.3%.
The magnitude of our FDI effect on income, of 3.4%, is larger than our static estimates of the
losses from trade (between 1.3% and 2.6% in Dhingra et al, 2016). The effect of changes in
FDI is equivalent to a loss of GDP of around £2,200 per household.
Using earlier data, Pain and Young (2004) estimate that EU membership added 2.25% to UK
GDP via FDI. As FDI into the UK has grown over time, we find that this channel is becoming
more important for income.
Such macroeconomic analysis is useful for a bird’s-eye view of the impact of Brexit on national
income via lower FDI. Firm-level studies will tend to underestimate the positive impact of FDI
as they focus on the productivity of the foreign firm itself or can examine only a limited number
of mechanisms for the FDI spillovers (for example, firms who are in the same industry as the
multinational or are suppliers or customers). Nevertheless, identifying the causal effects of FDI
on economy-wide productivity is intrinsically very difficult and our estimates are subject to
considerably more uncertainty than the impact of Brexit on FDI (or trade) itself.
So, to obtain a more granular view of the way that key sectors may be affected, we analyse two
important UK industries in more detail: cars and financial services.
Concentrating on cars
Cars are a successful part of UK manufacturing. The UK is now the world’s fourth largest
producer and KPMG (2014) argues that ‘much of the recent investment by car manufacturers
is in new vehicles which will be predominantly for sale to the EU market.’ In 2014, the car
industry contributed around 5.1% to UK exports, and about 40% of its car exports were to the
EU.2 In a survey of its members in 2014, the Society for Motor Manufacturers and Traders
found that 70% of its members expect Brexit to have a negative medium to long-term impact
on their business.3
There are very rich data on the car industry, which enable us to extend the structural gravity
model of exports in Dhingra et al (2016) to the decisions of multinationals over where to base
their production. Head and Mayer (2015) use information on assembly and sales locations (IHS
Automotive data) on 1,775 models across 184 brands. These data include annual flows of each
model shipped from 49 assembly countries to 75 destination countries between 2000 and 2013.
They also contain information on the headquarters and assembly location of the car. The model
accounts for how the headquarters decide where to locate their production – for example, why
BMW chooses to produce Minis in the UK when selling to France.
2 http://researchbriefings.files.parliament.uk/documents/SN00611/SN00611.pdf;
http://researchbriefings.files.parliament.uk/documents/SN06091/SN06091.pdf.
3 http://www.smmt.co.uk/2014/04/uk-automotive-industry-europe.
27
Head and Mayer estimate the impact of Brexit on plant location as well as the levels of car
production and prices. In their work, Brexit has two main disadvantages:
• First, as trade costs rise (due to non-tariff and possibly tariff barriers), locating
production in the UK is less attractive because it becomes more costly to ship to the
rest of Europe.
• Second, there is an increase in the co-ordination costs between headquarters and the
local production plants. Transfers of key staff within the firm may be harder if migration
controls are put in place. Different regulatory standards can make engineering, R&D
and consultancy services trickier.
Generally, all the things that make trade more costly between firms in different countries will
also make trade within multinationals across countries more costly.4
Table 1 examines two scenarios for Brexit:
• First, row 1 considers both of the costs together – the increased trade costs of exporting
and the higher costs of headquarters co-ordination. Total UK car production is predicted
to fall by 12% or almost 180,000 cars per year. This is mainly because European car
manufacturers such as BMW move some production away from the UK. Prices faced
by UK consumers also rise by 2.55% as the cost of imported cars and their components
increase.
Table 1: The predicted impact of Brexit on UK car production and prices
Change in total
number of cars
produced
Percentage
change in cars
produced
Percentage
increase in car
prices
1. Increase in trade costs and
headquarters co-ordination
costs
-180,746 -12.0% 2.55%
2. Increase in headquarters coordination costs only
-35,728 -2.4% 0.003%
Source: Derived from Head and Mayer (2015).
• Row 2 takes a more optimistic approach and assumes that the UK faces no trade barriers
on cars and car components with the rest of the EU (for example, it joins EFTA). Hence,
the only increase in costs is due to increased headquarters co-ordination costs. Although
prices are stable in this rosier scenario, car production still falls by almost 36,000.5
In short, the detailed model in Head and Mayer confirms the macroeconomic and survey
evidence that the costs of Brexit for car production in the UK could be severe.
4 In theory, these intra-company transfers of services across country borders should be reflected in the trade
statistics, but we know that, in reality, these are not well captured. Because of transfer pricing, international flows
of technical and managerial know-how are hard to detect within multinationals.
5 We have abstracted away from a third channel quantified by Head and Mayer, which allows for the fact that the
UK brand would be less attractive to EU consumers after Brexit. This would further reduce welfare by increasing
prices by another 2%.
28
Focusing on financial services
Financial services have the largest stock of inward FDI in the UK (45%) and constitute 8% of
GDP and 12% of tax receipts (Tyler, 2015).
The Single Market allows a bank based in one member of the EU to set up a branch or provide
cross-border financial services in another, while being regulated by authorities in the home
country. This ‘single passport’ to conduct activities in EU member states is important for UK
exports of financial services. ‘Passporting’ means that a UK bank can provide services across
the EU from its UK home. It also means that a Swiss or an American bank can do the same
from a branch or subsidiary established in the UK.6
The UK might be able to negotiate some of these privileges after Brexit. Members of the
European Economic Area (EEA) outside the EU (for example, Norway) enjoy them, but they
also have to contribute substantially to the EU budget, to accept all EU regulations without a
vote on the rules and to allow free labour mobility with the EU. And still for these countries,
there seem to be greater difficulties in doing business with the EU (Souta, 2015; Bank of
England, 2015).
Switzerland is in EFTA (not the EEA) so it enjoys tariff-free access to the EU in goods. But it
has no passporting rights, so Swiss financial institutions mostly get access to the EU via special
bilateral treaties with the EU, which still require permissions to set up branches in EU
members.7 This is one of the reasons that Swiss banks often set up subsidiaries in the UK. The
EU’s new financial directives have set out further rules for authorisation of the EU operations
of Swiss firms, so the Swiss option is unlikely to ensure easy access to EU markets after Brexit
(City of London, 2013).
More generally, there are concerns that the EEA might not welcome the UK, and that the EU
may not grant the special bilateral terms it extended in the past to non-EU countries like
Switzerland8, since following Brexit, other cities like Frankfurt and Paris will be keen to grab
a larger share of the lucrative markets for financial services.9
Will Brexit relieve the UK financial services sector of onerous EU regulations? It is unlikely
to do so because UK-based financial firms would still need to comply with these regulations
for all their EU transactions.
Another question is whether EU regulations are imposing a big burden on UK firms in their
transactions outside the EU. As financial regulations are still evolving, it is difficult to put a
monetary value on the impact of Brexit on regulations. But it is unlikely that the regulations
put UK firms at a competitive disadvantage as the EU is the world’s largest exporter of financial
services, making up a quarter of world financial services exports. Half of the cross-border
lending in the world originates within the EU.
6 https://www.dlapiper.com/en/uk/insights/publications/2015/10/banking-disputes-quarterly/brexit/
7 http://www.bankofengland.co.uk/pra/Pages/authorisations/passporting/faqs.aspx.
8 http://www.publications.parliament.uk/pa/cm201213/cmselect/cmfaff/writev/futunion/m21.pdf;
https://www.cityoflondon.gov.uk/business/economic-research-and-information/researchpublications/Documents/research-2013/Switzerlands-approach-to-EU-engagement.pdf.
9 http://www.voxeurop.eu/en/content/article/5052502-big-opportunity-europe;
http://www.globalcapital.com/article/x1h6hjkpygxb/brexit-casts-long-shadow-over-nervous-uk-banking-sector.
29
The Balance of Competences Review questioned the City of London on the extent to which
the Single Market had raised the costs of transacting with countries outside the EU.10 The
consensus was that the City became a financial hub while being in the EU, so there was little
evidence that membership had seriously hindered the UK’s ability to trade with countries
outside the EU. City representatives said that access to the Single Market is one of the major
reasons for inward FDI in the UK.
Staying in the EU also gives the UK the ability to challenge new regulations at the European
Court of Justice, a right that it successfully exercised when the European Central Bank wanted
to limit clearing-house activities to the euro area. If the UK leaves the EU, it would lose its
leverage in negotiating and challenging future EU regulations.
Future trade agreements
After Brexit, would the UK strike great deals with non-EU countries that would reduce trade
costs and so actually boost FDI? It seems unlikely. Although the UK would no longer need to
compromise with other EU countries when negotiating, the UK is under a fifth of the economic
size of the EU’s Single Market. It would simply have much less bargaining clout than the EU
currently enjoys. Nor would it get automatic access to the new deals struck with the EU, such
as those currently being negotiated with Japan and the United States.
Conclusions
Overall, Brexit is likely to have a negative impact on inward FDI. Our new empirical analysis
implies that leaving the EU will reduce FDI inflows to the UK by around 22%.
Such losses of investment will damage UK productivity and could lower real incomes by 3.4%.
This is larger than our estimates of the static income losses from trade, which are 2.6% even
under our ‘pessimistic scenario’ (Dhingra et al, 2016).
Case studies of cars and finance also show that Brexit would lower EU-related output of goods
and services, and erode the UK’s ability to negotiate concessions from regulations on EUrelated transactions.
Of course, these costs may be a price that many people are willing to pay to leave the EU. But
they are not trivial costs.
April 2016
For further information, contact:
Swati Dhingra ([email protected]), Thomas Sampson ([email protected]), John
Van Reenen ([email protected]), Gianmarco Ottaviano ([email protected]) or
Romesh Vaitilingam ([email protected]).
10 https://www.gov.uk/guidance/review-of-the-balance-of-competences.
30
Further reading
Alfaro, L., A. Chanda, S. Kalemli-Ozcan and S. Sayek (2004) ‘FDI and Economic Growth:
The Role of Local Financial Markets’, Journal of International Economics 64(1): 89-112.
Baier, S. L., J. H. Bergstrand, P. Egger and P. A. McLaughlin (2008) ‘Do Economic Integration
Agreements Actually Work? Issues in Understanding the Causes and Consequences of the
Growth of Regionalism’, The World Economy 31(4): 461-97.
Bank of England (2015) ‘EU Membership and the Bank of England’
(http://www.bankofengland.co.uk/publications/Documents/speeches/2015/euboe211015.pdf).
Bloom, N., R. Sadun and J. Van Reenen (2012) ‘Americans Do IT Better: US Multinationals
and the Productivity Miracle’, American Economic Review 102(1): 167-201.
Bruno, R., N. Campos, S. Estrin and M. Tian (2016) ‘Gravitating towards Europe: An
Econometric Analysis of the FDI Effects of EU Membership’, Technical Appendix to this
report (http://cep.lse.ac.uk/pubs/download/brexit03_technical_paper.pdf).
Campos, N. and F. Coricelli (2015) ‘Some Unpleasant Brexit Econometrics’ VoxEU
(http://www.voxeu.org/article/some-unpleasant-brexit-econometrics).
City of London (2013) ‘Switzerland’s Approach to EU Engagement: A Financial Services
Perspective’, Centre for Swiss Politics, University of Kent.
Dhingra, S., G. Ottaviano, T. Sampson and J. Van Reenen (2016) ‘The Consequences of Brexit
for UK Trade and Living Standards’, CEP Brexit Analysis No. 2
(http://cep.lse.ac.uk/pubs/download/brexit02.pdf).
Harrison, A. and A. Rodriguez-Clare (2009) ‘Trade, Foreign Investment, and Industrial Policy
for Developing Countries’, NBER Working Paper No. 15621.
Haskel, J., S. Pereira and M. Slaughter (2007) ‘Does Inward Foreign Direct Investment Boost
the Productivity of Domestic Firms?’ Review of Economics and Statistics 89(3): 482-96.
Head, K. and T. Mayer (2015) ‘Brands in Motion: How Frictions Shape Multinational
Production’, UBC Working Paper.
Levine, R., N. Loayza, and T. Beck (2000) ‘Financial Intermediation and Growth: Causality
and Causes’, Journal of Monetary Economics 46(1): 31-77.
KPMG (2014) ‘The UK Automotive Industry and the EU’, report for the Society for Motor
Manufacturers and Traders.
Pain, N. and G. Young (2004) ‘The Macroeconomic Effect of UK Withdrawal from the EU’,
Economic Modelling 21: 387-408.
PWC (2016) ‘Leaving the EU: Implications for the UK Economy’
(http://www.pwc.co.uk/economic-services/assets/leaving-the-eu-implications-for-the-ukeconomy.pdf).
31
Souta, P. (2015) ‘What Could Brexit Mean for the Financial Services Sector?’
(http://www.britishinfluence.org/what_could_brexit_mean_for_the_financial_services_sector).
Straathof, S., G-J. Linders, A. Lejour and J. Mohlmann (2008) ‘The Internal Market and the
Dutch Economy: Implications for Trade and Economic Growth’, CPG Netherlands Document
No. 168.
Tyler, G. (2015) ‘Financial Services: Contribution to the UK Economy’, Economic Policy and
Statistics Section, House of Commons Library, February SN/EP/06193.
UKTI (2015) ‘UKTI Inward Investment Report 2014 to 2015’
(https://www.gov.uk/government/publications/ukti-inward-investment-report-2014-to-
2015/ukti-inward-investment-report-2014-to-2015-online-viewing).
32
Annex: Calculating the income effects of lower FDI
Alfaro et al (2004) estimate a cross-country growth regression with 73 countries to examine
the effects of FDI on economic growth. They allow the impact of FDI on GDP growth per
capita to vary with the level of development of domestic financial markets. They find that
countries with developed financial markets like the UK benefit significantly more from FDI
than those with less sophisticated financial systems (like many developing countries).
Alfaro et al (2004) present ‘instrumental variables’ (IV) regressions to deal with endogeneity
issues. For example, expected growth might affect the level of financial development. We use
these more rigorous IV estimates from their Table 7 (column 1), which is based on a regression
of average annual per capita growth rate of countries on FDI as a share of GDP, its interaction
with financial market development and various controls. The controls include initial GDP,
financial market development itself, schooling, population growth, government consumption
and a dummy for sub-Saharan Africa, the black market premium, inflation and trade volumes.
Since trade volumes are controlled for, the interpretation of the results is the impact of FDI on
growth over and above any influence of trade. The financial market development variable is
instrumented by dummies for English and Scandinavian legal origins. There is evidence that
these kind of legal origins enhanced the development of financial markets.
To calibrate the growth effect of FDI from this estimation, the parameter values we use are as
follows. The share of OECD FDI inflows in GDP for the UK is 2.4%, which is the average
from the data in Bruno et al (2016) from our Technical Appendix. The proxy for financial
market development in Alfaro et al (2004) is the share of private sector credit in GDP (in Table
7 column (1)). This takes a value of 0.463 (or 46.3% of GDP) in the UK in their data from
Levine et al (2000). We assume that the UK growth rate is 2% per year in the absence of Brexit,
which is taken from the Office for Budget Responsibility’s current projections of long-run UK
labor productivity growth.
Having calibrated the growth effect of FDI, we compute the extra income that would be needed
every year to ensure that a household gets the same discounted sum of log income with and
without Brexit (exactly like the analysis of the trade effects on income in Dhingra et al, 2016).
We use a discount rate of 0.96 for future incomes and set the intertemporal elasticity of
substitution equal to one.
When deriving the GDP per UK household, we use the current ONS estimates of 27 million
households and a GDP level of £1.8 trillion.
33
Brexit and the Impact of
Immigration on the UK
Jonathan Wadsworth,
Swati Dhingra, Gianmarco Ottaviano
and John Van Reenen
#CEPBREXIT
BREXITPAPER
Introduction
In the referendum debate about the UK’s membership of the European Union (EU), a major
argument of the Leave campaign is that Brexit would allow more control over the flow of
immigrants to the UK from the rest of the EU. Many people are concerned that high levels of
immigration may have hurt their jobs, wages and quality of life.
Immigration has grown a lot in the last 20 years and a significant fraction of this growth has
been from other EU countries, especially after 2004 and the accession of eight East European
countries (the ‘A8’). Between 1995 and 2015, the number of immigrants from other EU
countries living in the UK tripled from 0.9 million to 3.3 million. The share of EU nationals
grew from 1.5% to 5.3% of the total population and from 1.8% to 6.3% of the working age
population (adults aged 16-64).
Higher immigration has increased overall national income (more workers will generate more
GDP) and has benefited the immigrants who have come to the UK since, by and large, they are
better off than in their country of origin. But has it been economically harmful to people born
in the UK? In this Report, we present a new analysis of the most recent data to examine whether
EU immigration has affected the income prospects of the UK-born.
EU immigration to the UK
Figure 1: Net immigration to the UK, 1991-2015
Source: CEP analysis of ONS (2016) http://bit.ly/1Tz5WbW.
Net immigration is the difference between the number of people entering the UK and the
number of people leaving. Figure 1 shows how these have increased for EU and non-EU
immigrants. When the East European A8 countries1 joined the EU in 2004, immigration rose
1 The A8 countries are the Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Slovakia and Slovenia.
0 50 100 150 200 250
yearly flows (thousands)
1991 1995 1999 2003 2007 2011 2015
year
EU Non-EU
Net Immigration to UK
35
significantly, but it then fell back during the recession after 2007. In the last few years, net
inflows have again risen significantly as the economy has recovered. In the year to September
2015, net EU immigration was 172,000, comprising 257,000 EU nationals arriving and 85,000
leaving. This is only just below the figure of 191,000 net immigration for non-EU nationals.
The best source of data to explore the impact of immigration is the Labour Force Survey (LFS).
This is because it makes it possible to measure the economic circumstances of people born in
the UK and compare them with immigrants from the EU and other countries. The Annex gives
more details and compares the LFS with other data sources, such as National Insurance
numbers. The conclusions of this report are robust to using other data sources such as NI
numbers for migration and the Annual Survey of Hours and Earnings for wages.
In 2015, there were around 3.3 million EU immigrants living in the UK up from 0.9 million in
1995 – a rise to 5.3% of the population from 1.5%. Around 2.5 million of these migrants are
aged 16-64 and two million are in work.2 EU countries now account for 35% of all immigrants
living in the UK. While 29% of EU nationals are Polish and 12% are Irish, the nationalities of
other EU immigrants are quite evenly spread across the other 25 countries in the EU (Figure
2).
As with other immigrants, there is a greater concentration of EU nationals in London than in
the rest of the country (Figure 3). A third of EU nationals live in London compared with only
11% of UK nationals.
Figure 2: EU immigrants by nationality, 2015
Source: CEP analysis of Labour Force Survey.
2 The focus is on EU nationals (self-defined) rather than country of birth, since any decision to restrict entry
would presumably be based on nationality and not country of birth.
36
Figure 3: Proportion of UK and EU nationals in each region, 2015
Source: CEP analysis of Labour Force Survey.
Notes: London is marked in black. This is where immigrants are disproportionately concentrated.
EU immigrants are on average more educated than the UK-born (Table 1) – almost twice as
many of them have some form of higher education (43% compared with 23% UK-born). Only
15% of EU immigrants left school at 16 compared with 44% of the UK-born.
Table 1: Education and immigrant status (working age population) 2015
Age finished
education
UK-Born EU immigrants A8 immigrants All immigrants
High ( 21 or older) 23% 43% 36% 45%
Medium (17-20) 33% 42% 55% 36%
Low (16 or under) 44% 15% 9% 19%
All 100% 100% 100% 100%
Source: CEP analysis of Labour Force Survey.
Notes: The A8 countries are the Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Slovakia and
Slovenia, all of which joined the EU in 2004. Working age population is all individuals between the ages of 16
and 64.
Table 2 shows that EU immigrants are not only more educated, but they are also more likely
to be in work (78.2%) than UK-born individuals (72.5%) and less likely to be unemployed or
economically inactive. This is particularly true of A8 immigrants: almost 82% of them are in
work.
0 5 10 15 20 0 5 10 15 20
Northern Ireland
Rest of Scotland
Strathclyde
Wales
Rest of North West
Merseyside
Greater Manchester
Rest of West Midlands
West Midlands (met county)
South West
Rest of South East
Outer London
Inner London
East Anglia
East Midlands
Rest of Yorks & Humberside
West Yorkshire
South Yorkshire
Rest of Northern region
Tyne & Wear
Northern Ireland
Rest of Scotland
Strathclyde
Wales
Rest of North West
Merseyside
Greater Manchester
Rest of West Midlands
West Midlands (met county)
South West
Rest of South East
Outer London
Inner London
East Anglia
East Midlands
Rest of Yorks & Humberside
West Yorkshire
South Yorkshire
Rest of Northern region
Tyne & Wear
UK-Born EU Immigrants
Percent living in each region
37
Table 2: Employment, unemployment, students and economic inactivity by immigrant
status (working age population) 2015

Inactive 16.5% 11.6% 10.5% 18.3%
Source: CEP analysis of Labour Force Survey.
Notes: The A8 countries are the Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Slovakia and
Slovenia, all of which joined the EU in 2004. Working age population is all individuals between the ages of 16
and 64.
EU immigrants are typically younger. Among the working age population, the average age of
the UK-born is 40, the average western EU immigrant is 38 and the average A8 immigrant is
34.
Immigration, jobs and wages – national trends
About 70% of EU immigrants say they come to the UK because of work-related reasons, as
opposed to study or joining their families (ONS, 2016). Since immigration increases the total
number of people in work or looking for employment, does that mean that UK workers must
have been harmed by this increased competition for jobs?
The short answer is ‘no’. Believing otherwise is called the ‘lump of labour fallacy’. There
would be harm only if the total number of jobs is fixed and only where immigrants compete
for a particular job. But since immigrants also consume local services and goods, this increases
demand and so raises job prospects of those who produce those goods and services. Adding an
immigrant raises the population, just like a rise in the birth rate or a fall in the death rate. Over
the last 100 years, the UK population has grown by around 50% but the unemployment rate
has not trended inexorably upward.
But even if there is no reason to think that immigration should increase unemployment, is it
not obvious that an increase in the supply of workers must drive wages down? Again, it isn’t
necessarily so. Alongside the increased demand that a rising population brings, greater
movement of labour allows countries to specialise in what they are best at, just like increased
trade. Firms will change the mix of their products to account for the new skills available to
them. Immigrants, especially if they are more skilled, can boost productivity. All these effects
will tend to increase wages.
Consequently, the impact of immigration on UK-born workers is an empirical question and not
a foregone conclusion. We need to look at data for evidence.
There is a huge amount of research examining the effect of immigration on jobs and wages
(summarised in Wadsworth, 2015; Portes, 2016a; Centre for European Reform, 2016;
Dustmann et al, 2005, among others). The conclusion of this research is that the large increase
in immigration in the UK has not significantly harmed the job and wage prospects of UK-born
workers.
38
Most of this work, however, was conducted prior to the global financial crisis and the Eurozone
crisis. So it is reasonable to ask whether these findings have changed after the most severe
economic downturn for 80 years.
Figures 4 to 6 plot the unemployment, employment and wage trends for individuals born in the
UK alongside the trend in immigration from the EU. In Figure 4, at a time when EU
immigration has been rising sharply (after 2004), UK unemployment for those born in the UK
rose – but then fell back to a very low level, while EU immigration kept on rising. Indeed,
despite the global crash, the rise in unemployment for UK-born workers was much less than in
previous downturns when EU immigration was much lower.
Figure 4: Unemployment of UK-born and EU immigration, 1975-2015
Source: CEP analysis of Labour Force Survey.
Notes: % EU is the proportion of EU nationals in the working age population (16-64 for men, 16-59 for women).
A similar pattern can be seen using the fraction of the working population in jobs (the
employment rate) for UK-born workers (Figure 5).3 The employment rate of UK-born workers
goes up and down with the economic cycle, but has risen recently in the period when EU
immigration is also rising. It is almost back to levels seen at the peak of previous recoveries.
Median real wages for those born in the UK were growing from the late 1990s until the global
financial crisis. Since then, wages have fallen by about 10%. Such falls in real wages are
unprecedented in the post-war period. The story of the latest recession was not that many more
people lost their jobs, but that most people’s wages fell. Figure 6 confirms that this fall
happened while EU immigration was rising – but equally the big gains in real wages for UK
workers were experienced at a time when EU immigration was also rising. So the cause of the
fall of wages is the impact of the Great Recession – not immigration.
3 The employment count in Figure 5 excludes any students in work – but the trends are very similar if any
students in work are added back in to employment.
1 2 3 4 5 6
% EU nationals in working age population
4 6 8 10 12
UK-Born unemployment rate (%)
1975 1980 1985 1990 1995 2000 2005 2010 2015
Unemployment EU share in pop.
Onset of global
financial crisis
39
Figure 5: Employment rate of UK-born and EU immigration, 1975-2015
Source: CEP analysis of Labour Force Survey.
Notes: % EU is the proportion of EU nationals in the working age population (16-64 for men, 16-59 for women).
Figure 6: Median real hourly wages for the UK-born, 1995-2015
Source: CEP analysis of Labour Force Survey.
Notes: Median wage is deflated by the CPI.
1 2 3 4 5 6
% EU nationals in working age population
68 70 72 74
UK-Born employment rate (%)
1975 1980 1985 1990 1995 2000 2005 2010 2015
Employment EU share in pop.
2 3 4 5 6
% of EU nationals in working age pop
8 9 10 11 12
real hourly wages (£2015)
1995 1999 2003 2007 2011 2015
year
Real Wages EU share in pop.
Real Hourly Wages UK-Born Workers
40
Immigration, jobs and wages – local trends
Although there appears to be little correlation between EU immigration trends and the average
worker’s jobs or wages, what about an impact on certain types of workers? Even if no one loses
on average, could there be certain groups who do suffer badly?
The fact that EU immigrants are more educated would suggest that, if anything, they put
downward pressure on the wages of higher waged people, thus reducing inequality. No one
will shed many tears for bankers or university professors facing stiffer competition for their
services. But there is concern that less skilled workers are hurt if educated immigrants are
willing to accept low paying jobs (Migration Advisory Committee, 2014). For example,
according to 2015 data, a third of EU nationals are in the relatively low skilled ‘elementary and
processing occupations’ compared with 10% of UK nationals in work. Given that immigrants
are more highly educated, this may be because they are not using their skills fully. But it may
also reflect the fact that they are younger and so less likely to be in more senior managerial and
professional roles.
Figure 7: Wage inequality for UK-born workers and EU immigration 1995-2015
Source: CEP analysis of Labour Force Survey.
Notes: The Figure tracks growth in real wages (deflated by CPI) relative to level of wages in 1995 for the bottom
10th percentile, the median (50th) and top 10th (90th percentile) of UK-Born workers. For example a value of one
indicates the same level as 1995 and a value of 10 indicates real wages 10% above 1995 level. % EU is the
proportion of EU nationals in the working age population (16-64 for men, 16-59 for women).
If we track the wages of UK-born workers across the pay distribution over time (Figure 7), we
see that wages for those in the bottom tenth have, if anything, held up better than wages of UKborn workers elsewhere in the pay distribution. The sharp fall in real wages since the recession
has hit everyone. So most people are equally worse off. The introduction of the minimum wage
in 1999 gave a boost to pay growth at the bottom, which continued through to the late 2000s.
2 3 4 5 6
% of EU nationals in working age pop
1 5 15 25 35 45
% change on 1995 wage level (1995=1)
1995 1999 2003 2007 2011 2015
year

Real Hourly Wages Relative to 1995
41
Again, the rise in EU immigration happened in periods of relative wage growth at the bottom
wage levels and in periods of relative wage stability at the bottom. So it is hard to see evidence
of relative wage falls for low paid UK workers when EU immigration is rising.
On the face of it, it would seem that the recession coincides with much of the recent bad
experiences of UK-born workers with regard to jobs and pay rather than rising immigration.
Looking at economy-wide changes might disguise effects in local areas where immigration has
gone up by a lot. The most straightforward way to investigate this issue is to examine whether
areas of the UK that had larger influxes of EU immigrants also had worse job and wage
outcomes for the UK-born relative to other areas. Looking at the change over time controls for
lots of other features of the local labour market that could also explain unemployment and
wages in those areas.
Figure 8 considers changes in the unemployment rates of the UK-born across local authorities
in relation to changes in EU immigration between 2008 and 2015 (one dot for each of the 201
local authorities). The solid red line summarises the relationship between immigration and UKborn unemployment rates. If immigration increased unemployment, we would expect a strong
upward sloping line: more EU immigrants would mean more unemployment for local workers.
In fact, the line indicates that a 10 percentage point increase in the share of EU immigrants in
a local area is associated with a 0.4 percentage point reduction in the unemployment rate in
that area. But it is very clear from the graph that there is absolutely no statistically significant
relationship (negative or positive) of EU immigration on unemployment rates of those born in
the UK.
Figure 8: Unemployment rates of UK-born & EU immigration
Source: CEP analysis of Labour Force Survey.
Notes: Each dot represents a UK local authority. The solid line is the predicted ‘best fit’ from a regression of
changes in unemployment on the change in share of EU immigrants in each UK local authority. These are
weighted by the sample population in each area. Slope of this line is -0.04 with standard error of 0.05, statistically
insignificantly different from zero.

-8 -6 -4 -2 0 2 4 6 8
% point change in local area Unemployment Rate: 2008-2015
-2 0 2 4 6 8 10 12 14
% point change in local area EU Immigrant Share 2008-2015
local area values Fitted value
Change in UK-Born Unemployment Rate 2008-2015
42
So why does it feel like immigration hurts jobs? Think about two areas – dots A and B in Figure
8. Both have had increases in the EU immigrant share of over 8% – well above the national
average. In area A unemployment for the UK-born has risen by over 3%, which is also above
the national average. So in area A it feels like immigrants are bad for jobs. However, area B
has had a similar increase in immigration rates, while unemployment rates have fallen by 2%.
Therefore, just because immigration and unemployment both go up in an area does not mean
that this is true nationwide, since it is quite easy to find areas where the opposite has happened.
Something else must underlie the ill fortune of areas with rising unemployment.
Figure 9 provides the same analysis of the impact of EU immigration on pay. Again, there is
no apparent link between changes in the real wages of UK nationals and changes in EU
immigration. Wages of UK-born workers changed at much the same rate in areas with high EU
immigration as in areas where the change in EU immigration was low.
Figure 9: Wages of UK-born & EU immigration
Source: CEP analysis of Labour Force Survey.
Notes: Each dot represents a UK local authority. The solid line is the predicted ‘best fit’ from a regression of local
authority percentage change in wages on the local authority change in share of EU immigrants. These are weighted
by the sample population in each area. Slope of this line is -0.08 with standard error of 0.15, statistically
insignificantly different from zero.
To see if employment and wage prospects for less skilled UK nationals are associated with EU
immigration, Figure 10 looks at the change in the NEET rate (‘not in education, employment
or training’) for low skilled UK-born, defined as those who left school at the minimum leaving
age or younger. There is again no effect of EU immigration on their job prospects. If anything,
the relationship is negative – NEET rates fell furthest between 2008 and 2015 in areas where
EU immigration rose faster. But the estimate, like all others, is statistically insignificant.
-25 -20 -15 -10 -5 0 5 10 15 20
% Change in local area UK-Born Wages 2008-2015
-2 0 2 4 6 8 10 12 14
% point change in local area EU Immigrant Share 2008-2015
local area values Fitted value
% Changes in UK-Born Real Wages 2008-2015
43
Likewise, Figure 11 shows no obvious link between the pay of less skilled UK-born individuals
and changes in the local area population of EU nationals over this period.
The Technical Appendix presents a large number of variations of these graphs. For example,
we repeat the analysis for other periods such as starting in 2004 when the A8 joined the EU or
2011 when the Eurozone crisis really worsened. We drop London to make sure that the capital
is not biasing the relationships. Across all experiments, the results are essentially unchanged:
EU immigration does not seem to harm UK workers.
Figure 10: NEET (‘not in education, employment or training’) rates for less skilled UKborn & EU immigration, 2008-2015
Source: CEP analysis of Labour Force Survey.
Notes: Each dot represents a UK local authority. The solid line is the predicted ‘best fit’ from a regression of local
authority change in NEET rates for the less skilled on the local authority change in share of EU immigrants. These
are weighted by the sample population in each area. Slope of this line is -0.10 with standard error of 0.14,
statistically insignificantly different from zero. NEET stands for those ‘not in education, employment or training.’
Less skilled is defined by those who left school at 16 or earlier.
-20 -15 -10 -5 0 5 10 15
% Change in local area UK-born NEET Rate: 2008-2015
-2 0 2 4 6 8 10 12 14
% point change in local area EU Immigrant Share 2008-2015
local area values Fitted value
Change in UK-Born NEET Rate for Less Skilled 2008-2015
44
Figure 11: Wage rates for less skilled UK-born & EU immigration
Source: CEP analysis of Labour Force Survey.
Notes: Each dot represents a UK local authority. The solid line is the predicted ‘best fit’ from a regression of local
authority percentage changes in the wages of the less skilled on the change in share of EU immigrants. These are
weighted by the sample population in each area. Slope of this line is 0.02 with standard error of 0.21, statistically
insignificantly different from zero. Less skilled is defined by those who left school at 16 or earlier.
One group that does seem to lose out from new immigration is the stock of other recent arrivals
(Manacorda et al, 2011). So although there is no negative effect on UK-born workers, there
might be some depressing effects on other immigrants who settled in the UK a few years back.
Dustmann et al (2013) find small wage losses for those in the bottom 10% of the pay
distribution and larger wage gains for those in the middle of the pay ladder as a result of
immigration. Their estimates imply that the wave of EU immigration between 2004 and 2015
reduced wages in the bottom decile by 1.03% and raised wages for the median worker by
1.24%. But the areas where the bottom 10% had relatively slow wage growth were places like
London. London had big increases in immigration but also had the weakest bite of the rapidly
rising minimum wage. Consequently, it is likely that even these small negative effects are
overstated by not controlling for the minimum wage effects.
Nickell and Saleheen (2015) find small wage losses for occupations with fast increases in
immigration. Their results imply that all EU immigration since 2004 has reduced
semi/unskilled service sector wages by only about 0.7% (compared with a 4% increase in the
minimum wage over the same period, Centre for European Reform, 2016). But they do not
separate wages of the UK-born from immigrant wages as we do, so even their small effects of
EU immigration on wages may be coming from its effects on other immigrants. Furthermore,
the occupations that lose may be counterbalanced by the occupations that gain in a local area,
meaning that the overall effect on the area’s wages is zero, just as we find.
-30 -20 -10 0 10 20 30
% Change in local area UK-Born Wages: 2008-2015
-2 0 2 4 6 8 10 12 14
% point change in local area EU Immigrant Share 2008-2015
local area values Fitted value
% Changes in Real Wages of UK-Born Less Skilled 2008-2015
45
The impact of EU immigration on public finances and public services
What is the fiscal impact of immigration on public finances and public services?
Public expenditure will be lower on one point since UK taxpayers have not had to finance the
childhood schooling and healthcare costs of an immigrant adult as they would do for a UKborn adult. Second, we have documented above that EU immigrants are younger, more likely
to work and less likely to be on benefits. While immigrants, like UK nationals, would not be
eligible for contributory-related benefits until they have worked full-time for two years, they
could be eligible for means-tested benefits should they apply. HMRC estimates that around 6%
of tax credit claims are from households that include an EU national in line with the share of
EU nationals in the UK (House of Commons, 2014).
After trying to account for the many possible ways in which individuals pay taxes or draw
welfare, Dustmann and Frattini (2014) find that EU immigrants made a positive fiscal
contribution: they paid more in taxes than they received in welfare payments. For example, A8
immigrants paid in about £15 billion more than they took out in public spending in the decade
up to 2011. While this effect may seem small, the longer-run impact could be substantial. The
central estimate of the Office for Budget Responsibility (2013) is that the UK’s national debt
will be 40 percentage points higher in 2062 if net immigration is reduced to zero from 140,000
per year. By contrast, UK nationals, as a whole, received more in benefits than they paid in
taxes.
Given that EU immigrants are making net contributions, there is no reason to think that they
should crowd out any public services. In fact, they are bringing extra resources that could be
used to increase spending on local health and education for the UK-born. In other words,
reducing EU immigration would generate the need for greater austerity. This would magnify
the need for cutbacks caused by the slower growth of the economy due to reduced trade and
investment identified by Dhingra et al (2016a, 2016b).
Although the fact that the government has been cutting back on public services cannot therefore
be blamed on immigrants, it is still interesting to see whether immigration has led to worse
local services.
If immigrants cause social disruption, we would expect this to be reflected in crime rates. Bell
et al (2013) find no effect of the big 2004 increase in immigration from East European countries
on crime.
Geay et al (2013) find no effect of immigration on aspects of educational attainment and
actually some positive effect from Polish children on UK-born pupils. The disadvantage in
having English as a second language seems to be outweighed by a stronger immigrant push to
work hard at school.
For the NHS, Wadsworth (2013) finds no greater usage of doctors and hospitals by immigrants
relative to the UK-born; and Giuntella et al (2015) find little effect on NHS waiting times.
These studies do not distinguish between EU and non-EU immigrants, but since EU immigrants
are younger than non-EU immigrants, they are less likely to use health services, so the results
are likely to be stronger.
46
There is a general perception that immigrants are given better treatment when applying for
social housing. Battiston et al (2013) show that controlling for demographic, economic and
regional circumstances, immigrant households are less likely to be in social housing than their
UK-born counterparts. Lack of access to social housing has more to do with the falling supply
of social housing.
One area where we may be concerned is the effect of immigration on house prices. The UK’s
terrible track record of building insufficient houses does mean that the population increase
generated by immigrants adds to housing pressure. But the failure to create enough housing
supply would be a problem even in the absence of EU immigration. It is rooted in the failure
of the UK planning system to make appropriate infrastructure decisions more generally (LSE
Growth Commission, 2013; Hilber, 2015). Having said this, the empirical evidence does not
find positive effects of immigration on local house prices (Sa, 2015).
Refugees
The Syrian refugee crisis is not related to the UK’s continued membership of the EU. The
total immigration figures will not be much affected as the government will admit only around
20,000 adult refugees over the next five years. Refugees given the right to remain in Germany
or other EU countries have no right to live or work in the UK. It takes a number of years
(usually between five and eight) before refugees are even allowed to apply for citizenship.
Most of those who are settled are unlikely to seek work in the UK. The UK is not in the
Schengen passport-free travel agreement, so there are border checks preventing entry of
refugees. Stopping illegal entry to the UK would not be any easier after Brexit.
Could EU immigration really be restricted after Brexit?
At present, only work visas issued to non-EU nationals are restricted. If the UK were to leave
the EU but wanted to remain a member of the European Free Trade Area or the European
Economic Area, it may have to accept unrestricted EU immigration just as all other countries
like Norway or Switzerland do. Only a looser trading agreement with higher trade costs would
potentially enable the UK to restrict work-related EU immigration in much the same way as
non-EU immigration is restricted.
If EU immigration were cut following Brexit, then something like the current visa scheme that
applies to non-EU immigrants would have to be adopted to accommodate immigration from
the EU. Current rules effectively exclude non-EU immigration from all but graduate jobs and
limit numbers arriving on work visas each year to around 55,000, (5,000 in ‘Tier 1’ and 50,000
in ‘Tier 2’).4 This would mean decisions would have to be taken on whether to expand the
current quotas and which skill groups to allow. It is likely then that after Brexit, skilled EU
immigration would be cut and there is little realistic prospect of non-EU skilled immigration
being expanded.
4 See https://www.gov.uk/government/publications/immigration-statistics-october-to-december-2015/work.
In addition to the 55,000 work visas, there were an additional 38,000 dependents. The total for Tier 2 includes
a quota of 20,700 work visas with the rest made up of short-term Inter Company Transfer visas.
47
Immigration, welfare benefits and the National Living Wage
How much ‘benefit tourism’ from the EU is there? LFS data show that EU immigrants are less
likely than UK nationals to claim unemployment benefit, housing benefit or tax credits (Centre
for European Reform, 2016).
Another argument made in favour for Brexit is that the big increases in the minimum wage (the
National Living Wage) planned over the next four years will draw in many more EU
immigrants.5 It is unclear how big a draw this will be since it depends, in part, on what other
countries do to their own wages and on the relative cost of living in each country. Office for
Budget Responsibility, (2015), predicts an increase in unemployment of 60,000 which will also
be concentrated among the less skilled.
Productivity and immigration
A disadvantage of focusing on outcomes in local areas is that they may miss out on the
economy-wide effects of immigration. By enabling specialisation and raising productivity,
immigration could also raise national wages.
Migration acts much like trade, as people tend to move to countries where they can be more
productive and earn higher incomes. Migrants move from countries with lower productivity
(and hence lower wages) to countries with higher productivity: this increases welfare through
greater efficiency in labour allocation across the world. Immigrants also fill the gaps in the skill
composition of the national workforce. This fosters specialisation, increases productivity, and
raises the wages of national workers with complementary skills.
There is a consensus that there are positive effects of trade and foreign direct investment on
UK productivity. But there is somewhat less of a consensus on the effect of immigration on
productivity. There is strong evidence of positive effects for more educated immigrants (for
example, Ottaviano et al, 2016, for UK service productivity; Ortega and Peri, 2014). Indeed,
most studies show insignificant or positive effects of overall immigration.6 For example,
Felbermayr et al (2010) concludes that a 10% increase in the immigrant stock leads to a per
capita income gain of 2.2%.
Recent work by Boubtane et al (2015, Table 3) finds that a 50% decrease in the net immigration
rate would reduce UK productivity growth by 0.32% per annum. Since EU immigration is half
of the current UK total (see Figure 1), cutting EU immigrants to 80,000 per year is likely to
shave 0.16% off productivity growth. So about a decade after Brexit, UK GDP per capita will
be about 1.6% lower than it would have otherwise been.
Supporters of Brexit argue that economic benefits would flow from bringing EU immigration
under the same rules as non-EU immigrants. Boubtane et al (2015) also look at how improving
the average skill level of immigrants could increase productivity. To offset the productivity
loss from halving EU net immigration, the UK immigration system would have to improve the
5 See http://www.dailymail.co.uk/wires/pa/article-3554751/Brexit-camp-backlash-Obama-queue-warning.html.
6 For example, http://data.parliament.uk/writtenevidence/committeeevidence.svc/evidencedocument/homeaffairs-committee/immigration-skill-shortages/written/23066.pdf.
48
relative education levels of EU immigrants by about a quarter. Since EU immigrants are already
significantly better educated than the UK-born, this may be hard to engineer.
Conclusion
We cannot be precise about the size of the losses from restricting immigration following a
Brexit. But we can confidently say that the empirical evidence shows that EU immigration has
not had significantly negative effects on average employment, wages, inequality or public
services at the local level for the UK-born. Nor, it should be said, are there large positive
effects. Any adverse experiences of UK-born workers with regard to jobs and wages are more
closely associated with the biggest economic crash for more than 80 years.
At the national level, falls in EU immigration are likely to lead to lower living standards for
the UK-born. This is partly because immigrants help to reduce the deficit: they are more likely
to work and pay tax and less likely to use public services as they are younger and better
educated than the UK-born. It is also partly due to the positive effects of EU immigrants on
productivity.
Our earlier reports reflect a wide consensus that trade and foreign investment will also fall after
Brexit, both of which would reduce UK incomes. Lower immigration is a third channel that
will push UK living standards even lower.
May 2016
For further information, contact:
Jonathan Wadsworth ([email protected]), Swati Dhingra ([email protected]),
Gianmarco Ottaviano ([email protected]), John Van Reenen ([email protected])
or Romesh Vaitilingam on 07768-661095 ([email protected]). Jonathan Wadsworth
was a member of the Home Office’s Migration Advisory Committee 2007 to 2015.
49
Further reading
Battiston, D., R. Dickens, A. Manning and J. Wadsworth (2013) ‘Immigration and the Access
to Social Housing in the UK’, CEP Discussion Paper No. 1264
(http://cep.lse.ac.uk/pubs/download/dp1264.pdf).
Bell B., F. Fasani and S. Machin (2013) ‘Crime and Immigration: Evidence from Large
Immigrant Waves’, Review of Economics and Statistics 95(4): 1278-90.
Boubtane, E., J-C. Dumont and C. Rault (2015) ‘Immigration and Economic Growth in the
OECD Countries 1986-2006’ CESIFO Working Paper 5392
(http://papers.ssrn.com/sol3/Delivery.cfm/SSRN_ID2622005_code459177.pdf?abstractid=26
22005&mirid=3).
Centre for European Reform (2016) The Economic Consequences of Leaving the EU
(https://www.cer.org.uk/publications/archive/report/2016/economic-consequences-leavingeu-final-report-cer-commission-brexit).
Dhingra, S., G. Ottaviano, T. Sampson and J. Van Reenen (2016a) ‘The Consequences of
Brexit for UK Trade and Living Standards’ CEP Brexit Analysis No. 2
(http://cep.lse.ac.uk/pubs/download/brexit02.pdf).
Dhingra, S., G. Ottaviano, T. Sampson and J. Van Reenen (2016b) ‘The Impact of Brexit on
Foreign Investment in the UK’, CEP Brexit Analysis No. 3
(http://cep.lse.ac.uk/pubs/download/brexit03.pdf).
Dustmann, C., F. Fabbri and I. Preston (2005) ‘The Impact of Immigration on the UK Labour
Market’, Economic Journal 115: F324-41.
Dustmann, C., T. Frattini and I. Preston (2013) ‘The Effect of Immigration along the
Distribution of Wages’, Review of Economic Studies 80: 145-73.
Dustmann, C. and T. Frattini (2014) ‘The Fiscal Effects of Immigration to the UK.’
Economic Journal 124: F593-643.
Geay, C., S. McNally and S. Telhaj (2013) ‘Non-Native Speakers in the Classroom: What are
the Effects on Pupil Performance?’ Economic Journal 123: F281-307.
Felbermayr, G., S. Hiller and D. Sala (2010) ‘Does Immigration Boost per Capita Income?’
Economic Letters 107(2): 177-79.
Giuntella, O., C. Nicodemo and C. Vargas Silva (2015) ‘The Effects of Immigration on NHS
Waiting Times’, University of Oxford Working Paper.
Hilber, C (2015) ‘UK Housing and Planning Policies’ CEP Election Analysis No. 3
(http://cep.lse.ac.uk/pubs/download/EA033.pdf).
House of Commons (2014) ‘Statistics on migrants and benefits’, House of Commons Note
Standard (http://researchbriefings.parliament.uk/ResearchBriefing/Summary/SN06955).
50
LSE Growth Commission (2013) Investing for Prosperity,
(http://www.lse.ac.uk/researchAndExpertise/units/growthCommission/documents/pdf/LSEG
C-Report.pdf).
Manacorda, M., A. Manning and J. Wadsworth (2011) ‘The Impact of Immigration on the
Structure of Male Wages: Theory and Evidence from Britain’, Journal of the European
Economic Association 10: 120-51.
Migration Advisory Committee (2014), ‘Migrants in Low Skilled Work’
(https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/333083/MAC
-Migrants_in_low-skilled_work__Full_report_2014.pdf).
Nickell, S. and J. Saleheen (2015) ‘The Impact of Immigration on Occupational Wages:
Evidence from Britain’, Bank of England Staff Working Paper No. 574.
Office for Budget Responsibility (2013) ‘Fiscal Sustainability Report’, July.
Office for Budget Responsibility (2015) ‘Economic and Fiscal Outlook’, July.
ONS (2016) Migration Statistics Quarterly Review
(https://www.ons.gov.uk/peoplepopulationandcommunity/populationandmigration/internatio
nalmigration/bulletins/migrationstatisticsquarterlyreport/february2016).
Ortega F. and G. Peri (2014) ‘Migration, Trade and Income’, Journal of International
Economics 92: 231-51.
Ottaviano, G., G. Peri and G. Wright (2016) ‘Immigration, Trade and productivity in
services’, CEP Discussion Paper 1353 (http://cep.lse.ac.uk/pubs/download/dp1353.pdf).
Portes, J. (2016a) ‘Immigration, Free Movement and the UK Referendum’, NIESR Review
236.
Portes, J. (2016b) ‘EU Immigration and the UK Labour Market’, NIESR mimeo.
Sa, F., (2015) ‘Immigration and House Prices in the UK’, Economic Journal 125: 1393-1424.
Wadsworth, J. (2013) ‘Mustn’t Grumble: Immigration, Health and Health Service Use in the
UK and Germany’, Fiscal Studies 34(1): 55-82.
Wadsworth, J. (2015) ‘Immigration and the UK Labour Market’, CEP Election Analysis
No. 1 (http://cep.lse.ac.uk/pubs/download/EA019.pdf).
51
Annex: Data Sources and Definitions
Labour Force Survey (LFS)
Most of this report is based on CEP analysis of the latest individual-level data from Labour
Force Survey (LFS). The LFS is the best data source to use covering a representative sample
of individuals living in the UK. For example, in 2015, it includes about 90,000 people. The
analysis in Figures 8-11 uses the Annual Population Survey (APS) which is based on pooling
the LFS quarterly panel over about a year. It has about 350,000 observations a year for the
working age population. In the regression lines we weight by the sample population to correct
the standard errors for small areas.
With the LFS, it is possible to separate out who is a UK-born individual from those who are
EU nationals. This enables us to examine not just the reported trends in the labour market but
also to break this down into the UK-born and immigrants.
National Insurance numbers
There is some discrepancy between the number of National Insurance (NI) numbers issued to
EU nationals and survey estimates of the EU immigrant population and inflows based on the
LFS and International Passenger Survey (IPS) respectively. Anyone should claim a NI number
if they are in work, looking to work or wish to be eligible for tax credits or benefits. The NI
count is based on these inflows. It has the advantage over the LFS that it is administrative data
on the population of NI numbers and therefore not a sample. But compared with the LFS, it
will underestimate the flows for EU immigrants who are working but do not have NI numbers
or those who are not seeking to work.
A big disadvantage of NI numbers is that individuals keep their NI number when they leave
the country. Consequently, we do not know the immigrant outflows using NI numbers. Thus,
it is not possible to make a reliable calculation of the net immigration numbers using NI counts.
According to the IPS, about 90,000 EU nationals leave the UK each year, taking their NI
numbers with them. Another disadvantage of the NI numbers is that an immigrant may have
one even if they are only in the UK for a short space of time. It is not simply the immigrants
who are in the UK for at least a year.
Consequently, the LFS is a sample of all individuals living in the UK at any one time, so it is
a better snapshot of the current immigration position in our view.
Definitions of immigrant status
The LFS asks people whether they were born in the UK and (except where noted otherwise)
this forms the basis of our outcomes for the UK-born. For EU immigrants, we use the
information on whether someone responded in the LFS that that they were a (non-UK) citizen
of the EU. We use EU ‘nationals’ rather than EU-born because any post-Brexit policy would
be to restrict people from entering the UK based on their citizenship rather than where they
were born. But the results are similar using whether individuals were born in the EU as an
52
immigrant measure rather than an ‘EU national’, so nothing much hinges on this.
Definitions of working age population
In most of the text, we define the working age population as all those aged between 16 and 64.
In the figures, we use a shorter age range for women of 16-59 since women retired at 60 for
much of the period used in the graphs. Nothing materially changes if we use a higher age for
women.
Alternative sources of data on labour market outcomes
As an alternative method to our local analysis, Portes (2016b) correlates constituency-level
changes in labour market outcomes with changes in immigration rates based on National
Insurance numbers. He uses NOMIS administrative data and ASHE to calculate numbers on
Job Seekers Allowance, unemployment and employment rates and hourly wages. Like us, he
finds no evidence of significantly negative effects of EU immigration on jobs or wages.
These alternative measures have the advantage over LFS of being administrative data rather
than a survey. But these data cannot be used to distinguish the labour market outcomes of the
UK-born from immigrants, which is an advantage of our study.
53
Who Bears the Pain?: How
the costs of Brexit would be
distributed across income
groups
Holger Breinlich, Swati Dhingra,
Thomas Sampson and John Van
Reenen
#CEPBREXIT
PAPERBREXIT
Introduction
There now is a consensus among economists that a UK referendum decision to leave the
European Union (EU) would lead to significant losses in average national income (for
example, HM Treasury, 2016; IFS, 2016; OECD, 2016; NIESR, 2016). But supporters of
Brexit argue that the costs would only be borne by the elite and that most of the population,
especially those on lower incomes, would be better off.
This report analyses the distributional effects of Brexit across different income groups and
types of households. We look at how prices would change after Brexit due to changing trade
patterns. We conclude that the pain of Brexit would be widely shared, with the middle classes
being slightly harder hit than the richest and poorest.
We focus on the consequences of changes in trade barriers, which are the best-understood
mechanism through which Brexit would affect the UK economy. But we also discuss
potential impacts through reduced EU immigration and changes in skill demands.
Measuring the economic impact of Brexit
Brexit would lead to lower trade with the EU, resulting in reductions in average income. But
not all households would be affected equally. If Brexit led to an increase in the trade barriers
imposed on goods and services that are predominantly consumed by richer households, for
example, such households would see their incomes decline more than others.
We use the same ‘computable general equilibrium’ model of the global economy developed
by Dhingra et al (2016a) to study the consequences of Brexit, but focus on the impact across
different income groups. We have 31 industries in the model, so we can track how the
changes in trade after Brexit would affect prices across these sectors. Since people spend
their money on different things, they would be differentially affected by the price changes.
As before, we consider an ‘optimistic scenario’, in which the UK remains a member of the
European Economic Area (EEA) similar to Norway, and a ‘pessimistic scenario’, in which
the UK is simply a member of the World Trade Organization (WTO). Currently, the Leave
campaign appears to be rejecting the EEA model, so the pessimistic outcome is more likely.
Trade costs may increase after Brexit due to (i) higher tariff barriers between the UK and the
EU; (ii) higher non-tariff barriers to trade (for example, arising from border controls, etc.);
and (iii) non-participation in future steps the EU takes towards the reduction of non-tariff
barriers.
Distributional consequences of Brexit
Since our 31 sectors include business-to-business sales (intermediate inputs), which
consumers do not directly purchase, we focus on the final goods and service price changes.
Note that price changes for intermediates would also affect consumers indirectly. For
example, in our pessimistic scenario, tariffs between the EU and the UK are 7.3% for
transport equipment such as cars and 1.8% for basic metals such as steel. Higher tariffs on
55
transport equipment directly affect consumers because they raise the prices of imported cars.
But higher tariffs on steel also raise the price of cars indirectly because they increase the
production costs for domestic producers that import steel.
Figure 1 shows the overall predicted price changes for 13 broad groups of goods and services
consumed by households. Groups that have a substantial share of tradeable products are
predicted to see the largest price increases. Prices would rise most for transport (4% to 7.5%),
alcohol (4% to 7%), food (3% to 5%) and clothing (2% to 4%). By contrast, service sectors
such as education or hotels and restaurants would be less affected because they rely more on
non-tradeable local inputs.
Figure 1: Price increases of goods and services after Brexit
0.09
0.08
0.07
0.06
0.05
0.04
0.03
0.02
0.01
0

Notes: Predicted price increases are based on the model by Dhingra et al (2016a). See Table A1 in the Annex for
the exact percentage changes for each product group.
We look at 10 income groups, from the poorest 10% to the richest 10% of household income
in Figure 2. There are substantial differences in how they choose to spend their money as
indicated by their expenditure shares across product groups. For example, the poorest 10% of
households spend 16% of their income on ‘Food and non-alcoholic drinks’, whereas the
richest 10% of households only spend around 8% on this category. This reflects the wellknown fact that poorer consumers need to spend a larger proportion of their income on
essentials.
By contrast, low-income households spend only 7% on ‘Transport’, which includes the
purchase of vehicles as well as transport services such as rail and air travel; the richest 10%
of households spend 16% of their income on transport.
56
Figure 2: Spending shares on goods and services by income deciles in the UK
Lowest Second Third Fourth Fifth Sixth Seventh Eighth Ninth Highest
0.25
0.2
0.15
0.1
0.05
0
Notes: Data from ONS (2012). Income deciles are based on gross household income.
Figure 3 summarises the effect of the price changes following Brexit on the real incomes of
the different household groups. In the optimistic scenario, the average household sees real
income falls of 1.8%, whereas in the pessimistic scenario, this loss increases to 4%.1 Looking
across income groups, it seems that the drops are reasonably even. It is certainly not the
richest 10% who do a lot worse. Households in the middle income groups are hit slightly
harder, with income drops of up to 4.2%.
Another way of looking at the distributional effects of Brexit is to look across different types
of households, such as pensioners, single households or households with children. The
consumption expenditures of these different groups are unsurprising: pensioners spend more
on ‘Health’ and families with children more on ‘Education’.
Overall, however, spending on the most important expenditure categories such as housing,
transport and recreation is not too dissimilar across household groups. This explains why the
effects of Brexit are similar across different types of households (Figure 4), with average
losses ranging from 1.8% in the optimistic scenario to 4% in the pessimistic scenario. The
only exception is single households, which lose slightly less (1.7% and 3.6%) because they
spend relatively less on ‘Food and non-alcoholic drinks’, a product group that we predict to
see large price increases after Brexit.
1 Note that these changes take into account the net savings from lower EU contributions (see Dhingra et al, 2016b)
for details. We assume that these are evenly spread across the population in proportion to gross income.
57
Figure 3: Real income losses by household income decile (%)
Notes: Predicted real income losses based on the model by Dhingra et al (2016a). See Table A2 in the Annex for
the exact percentage changes for each income decile.
Figure 4: Real income losses by demographic groups (%)
Notes: Predicted real income losses based on the model by Dhingra et al (2016a). See Table A3 in the Annex for
the exact percentage changes for each household group.
Pensioners Single person
Average
Single parent Fam. with children Household
0.0%
-0.5%
-1.0%
-1.5%
-2.0%
-2.5%
-3.0%
-3.5%
-4.0%
-4.5%
Optimistic Scenario Pessimistic Scenario
lowest second third fourth fifth sixth seventh eighth ninth highest average
0.0%
-0.5%
-1.0%
-1.5%
-2.0%
-2.5%
-3.0%
-3.5%
-4.0%
-4.5%
Optimistic Scenario Pessimistic Scenario
58
The effects analysed so far are ‘static’, so they ignore the effects of trade on productivity as
well as the impact of Brexit on foreign direct investment (FDI), which is substantial.
Incorporating these effects more than triples the magnitude of the income losses, as
demonstrated in Dhingra et al (2016a). We illustrate this in Figure 5. We summarise the
equivalent figures in cash (2016 prices) as well as percentage terms in Table 1 with more
details in the Annex (Table A2). Note that these are all in terms of household gross income
(not GDP).
Figure 5: Long-run real income losses by household income decile (%)
Notes: Predicted real income losses based on the model by Dhingra et al (2016a). See Table A2 in the Annex for
the exact percentage changes for each income decile.
Looking solely at the static impact of trade, the income of the average UK household drops
by between 1.8% (£754) per year and 4% (£1,637) per year. If we take into account the
longer-run dynamic effects of Brexit on productivity, average households lose between 6.1%
and 13.5% per year of their real incomes per year (£2,519 to £5,573). For the poorest tenth of
households (bottom decile), real income losses are 1.7% to 3.6% in the short run and 5.7% to
12.5% in the long run. For the richest households, the short-run losses are 1.8% to 3.9% and
the long-run losses are 6% to 13.4%.
lowest second third fourth fifth sixth seventh eighth ninth highest average
0.0%
-2.0%
-4.0%
-6.0%
-8.0%
-10.0%
-12.0%
-14.0%
-16.0%
Optimistic Scenario Pessimistic Scenario
59
Table 1: Summary of distributional effects of Brexit in cash terms

Notes: 2015-16 gross incomes figures projected from ONS (2012, 2014). Optimistic is the EEA model and
Pessimistic is the WTO model. Dynamic model includes productivity effects of trade and FDI, static model does
not. Summarised version of Table A2 in Annex.
Incorporating other distributional effects of Brexit
The calculations so far focus on price changes and therefore the effects on real income,
implicitly assuming that nominal wage changes are proportional across income groups (that
is, the percentage changes are the same). This seems to be a reasonable assumption. The
changes in prices across sectors predicted by our model are not significantly correlated with
average skills (as measured by earnings) across sectors (Annex).
Economists for Brexit (2016) recommend that the UK unilaterally abolishes all trade barriers
after leaving the EU. They claim that this will increase UK trade and incomes. Their model
predicts extremely large increases in wage inequality, which would mean that lower income
groups would lose out much more from Brexit than we find. Dhingra et al (2016c) show that
the modelling approach used by Economists for Brexit is unreliable as it is inconsistent with
the most basic empirical evidence on how countries trade (for example, their model implies
that the UK could cease exporting to the EU entirely at no cost). When we examine the
impact of trading under WTO rules and moving unilaterally to free trade, the losses in real
income would still be substantial.
We do not model adjustment costs arising from the need to move workers across different
sectors because it is difficult to know what the exact effects would be. As there are likely to
be such adjustment costs from the Brexit shock, this would magnify the income losses
discussed here.
There is a view that EU immigration has had a negative effect on the labour market and that
Brexit would help this by reducing migration. But as Wadsworth et al (2016) show, EU
immigration has not increased unemployment or reduced the wages of the UK-born living in
the areas where immigration has risen the most. Nor is there any robust evidence that the
low-waged were hurt by much more than those at the top of the income ladder (Centre for
European Reform, 2016). In fact, the most likely impact of EU immigration on wages is a
60
positive one due to improved productivity. So it is unlikely that changes to immigration
patterns would materially alter the pattern of income losses reported here.
Conclusions
Economists consistently find that Brexit would lower real incomes in the UK. The cost of
lower trade and foreign investment would not be outweighed by a reduction in the net fiscal
transfer to other parts of the EU.
This report shows that the economic cost of Brexit would not be just born by the rich. The
pain would be evenly shared across the income distribution – every group would lose by
broadly similar proportions. Those in the middle would lose slightly more than others, with
average losses of between 1.8% (£754 per year) and 13.5% (£5,573 per year).
When it comes to the pain from leaving the EU, it appears that no one would be spared.
May 2016
For further information
Contact: Holger Breinlich ([email protected]), Swati Dhingra
([email protected]) Thomas Sampson ([email protected]), John Van Reenen
([email protected]) or Romesh Vaitilingam ([email protected])
61
Further reading
Centre for European Reform (2016) The Economic Consequences of Leaving the EU
(https://www.cer.org.uk/publications/archive/report/2016/economic-consequences-leavingeu-final-report-cer-commission-brexit).
Dhingra, S., H. Huang, G. Ottaviano, J.P. Pessoa, T. Sampson and J. Van Reenen (2016a)
‘The Costs and Benefits of Leaving the EU: Trade Effects’, CEP
(http://cep.lse.ac.uk/pubs/download/brexit02_technical_paper.pdf).
Dhingra, S., G. Ottaviano, T. Sampson and J. Van Reenen (2016b) ‘The Consequences of
Brexit for UK Trade and Living Standards’, CEP Brexit Analysis No. 2
(http://cep.lse.ac.uk/pubs/download/brexit02.pdf).
Dhingra, S., G. Ottaviano, T. Sampson and J. Van Reenen (2016c) ‘Economists for Brexit: A
Critique’, CEP Brexit Analysis No. 6 (http://cep.lse.ac.uk/pubs/download/brexit06.pdf).
Economists for Brexit (2016a) ‘The Economy After Brexit’
(http://issuu.com/efbkl/docs/economists_for_brexit_-
_the_economy/1?e=24629146/35248609).
HM Treasury (2016) ‘The Long-term Economic Impact of EU Membership and the
Alternatives’, London: HMSO.
IFS (2016) ‘Brexit and the UK’s Public Finances’, Institute for Fiscal Studies.
NIESR (2016) ‘The Economic Consequences of Leaving the EU’, May Special Issue,
National Institute of Economic and Social Research.
OECD (2001) ‘COICOP-CPC and CPC-COICOP Correspondence Tables’, Paris.
OECD (2016) ‘The Economic Consequences of Brexit: A Taxing Decision’ Paris: OECD
http://www.oecd.org/economy/the-economic-consequences-of-brexit-a-taxing-decision.htm
ONS (2012), ‘Living Costs and Food Survey 2011,’ Office for National Statistics.
ONS (2014), ‘The Effects of Taxes and Benefits on Household Incomes 2013/2014,’ Office
for National Statistics.
ONS (2016), ‘Average weekly earnings by industry (EARN03),’ Office for National
Statistics.
Wadsworth, J. Dhingra, S., G. Ottaviano, and J. Van Reenen (2016) ‘Brexit and the Impact of
Immigration on the UK’, CEP Brexit Analysis No. 5
(http://cep.lse.ac.uk/pubs/download/brexit05.pdf).
62
Annex: More details of the calculations
To calculate the implied effects of Brexit across the income distribution and different
demographic groups, we start with the estimated effects on industry prices from our structural
computable general equilibrium model in Dhingra et al (2016a). For different assumptions
over post-Brexit trading relationships, we estimate what effect this will have on industry
prices and average nominal wages. In our model, Brexit-induced price changes vary across
sectors for three reasons:
 First, trade barriers increase by different amounts across sectors.

The end result of our model simulations is a list of price changes for each of the 31 sectors
from the World Input Output Data (WIOD), shown in Table A1 below.
Data for household expenditure on goods and services by income and demographic groups
are available from the 2012 edition of the Office for National Statistics (ONS) Living Costs
and Food Survey (Tables A5, A6 and A17). Figure 1 shows expenditures of each income
decile on 13 broad groups of final goods and services. We map the 31 WIOD sectors onto
these 13 expenditure groups using correspondences provided by OECD (2001). As explained
in the main text, households do not directly consume some of the intermediate input sectors
in WIOD (for example, basic metals such as steel) but are still affected by the corresponding
price changes as it feeds into the prices of final goods that they do buy (for example, cars).
Figure A2 shows the correlation between average weekly earnings by sector and the price
changes from Table A1. Average weekly earnings are from ONS (2016). The fitted line in
Figure A2 is based on an OLS regression of price changes on average weekly earnings. The
slope of the fitted line is 0.0009 and the corresponding robust standard error is 0.0013.
Because the slope coefficient is not statistically significantly different from zero, we assume
in our analysis that there is no differential impact of Brexit on nominal wages.
Table A2 gives the details on the exact proportionate losses by income groups and their cash
equivalents shown in the Figures and Table 1. We update the level of income from ONS
(2014), which has 2013-14 data to 2015-16 values using the growth in income of 5.2% over
the two years from ONS. Note that although the numbers are similar to those in Dhingra et al
(2016a) they are not identical for two main reasons.
First, we are looking at the effects on real incomes of final consumers. The average price
increase in Figure 1 is larger than in Table A1 because the effects of Brexit are greater on
prices of final goods and services that households directly spend money on (like food) than
they are for intermediate goods (like non-metallic minerals).
Second, we are looking at household income and spending rather than GDP (average GDP
per household is higher than average income per household). For example, households report
63
spending only small amounts on health and education as these are provided as free public
services by the state. Since the prices of health and education do not rise much, there appears
to be little negative effect on these items. This is why looking at overall GDP, which includes
government spending (as we did in Dhingra et al, 2016a), is in some respects preferable to
looking at just household spending and income patterns.
The long-run calculations are based on calculations in Dhingra et al (2016a). There we find a
long-run reduced form impact reduction of real GDP per capita of between 6.3% and 9.5%
compared with 1.3% (optimistic) and 2.6% (pessimistic) in the short-run static model. Using
the midpoint of the static effects (1.95%) and a conservative lower bound of 6.3% for the
dynamic effects, this implies a ratio of 3.23 (=6.3/1.95) between static and dynamic. We prorata up the static effects by this ratio to calculate the dynamic effects in Figure 5.
Table A1: Price changes by product groups shown in Figure 1
Product group Optimistic scenario Pessimistic scenario
Food and non-alcoholic
drinks 3.1% 5.3%
Alcoholic drinks and tobacco 3.9% 7.2%
Clothing and footwear 2.1% 3.9%
Housing, fuel and power 0.8% 1.2%
Household goods and
services 1.6% 2.8%
Health 2.1% 3.8%
Transport 4.0% 7.7%
Communication 1.4% 2.6%
Recreation and culture 1.2% 2.1%
Education 0.0% 0.1%
Restaurants and hotels 0.7% 1.2%
Miscellaneous goods and
services 0.6% 1.1%
Other expenditure items 0.4% 0.8%
Notes: Table shows the exact price changes underlying Figure 1.
64
Table A2: Change in household income by decile group

Notes: ‘%’is the implied percentage change in gross income from the static structural model in columns (1) and (3) and ‘£’ the cash equivalent of that change
(2016 prices). ‘Dynamic’ takes into account long-run productivity effects. 2013/14 Gross income figures from ONS (2014) projected forward by 5.2% for
2015/16 to account for income growth.
65
Table A3: Change in household income by household groups (Figure 4)
Household group Pensioner Single person Single parent Families with
children
Optimistic scenario -1.8% -1.7% -1.8% -1.8%
Pessimistic scenario -3.9% -3.6% -3.9% -3.9%
Notes: Table shows the exact real income changes underlying Figure 4.
66
Table A4: Price changes by WIOD sector
WIOD31 Sectors
Price
Change:
Optimistic
Price
Change:
Pessimistic
1 Agriculture, Hunting, Forestry and Fishing 1.60% 2.00%
2 Mining and Quarrying 2.10% 3.20%
3 Food, Beverages and Tobacco 4.20% 7.70%
4 Textiles and Textile Products; Leather, Leather and
Footwear 3.50% 5.10%
5 Wood and Products of Wood and Cork 1.60% 2.50%
6 Pulp, Paper, Paper , Printing and Publishing 1.60% 3.00%
7 Coke, Refined Petroleum and Nuclear Fuel 3.30% 3.80%
8 Chemicals and Chemical Products 6.30% 12.30%
9 Rubber and Plastics 2.50% 4.80%
10 Other Non-Metallic Mineral 1.40% 2.60%
11 Basic Metals and Fabricated Metal 3.70% 6.50%
12 Machinery, NEC (Not Elsewhere Classified) 4.20% 8.50%
13 Electrical and Optical Equipment 5.70% 10.10%
14 Transport Equipment 6.60% 13.90%
15 Manufacturing, NEC; Recycling 2.80% 5.20%
16 Electricity, Gas and Water Supply 0.10% 0.30%
17 construction 0.10% 0.20%
18 Retail Sale of Fuel; Wholesale Trade, Commission
Trade, including Motor Vehicles and Motorcycles 0.30% 0.60%
19 Retail Trade, Except for Motor Vehicles and
Motorcycles; Repair of Household Goods 0.10% 0.20%
20 Hotels and Restaurants 0.70% 1.20%
21 Inland Transport 0.30% 0.40%
22 Water Transport 1.70% 2.80%
23 Air Transport 2.50% 4.90%
24 Other Supporting and Auxiliary Transport Activities;
Activities of Travel Agencies 0.30% 0.50%
25 Post and Telecommunications 0.40% 0.70%
26 Financial Intermediation 0.50% 0.90%
27 Real Estate Activities 0.00% 0.10%
28 Other Business Activities and renting of equipment 1.50% 2.70%
29 Education 0.00% 0.10%
30 Health and Social Work 0.10% 0.10%
31 Public Admin, Defence, Social Security and other public
service 0.50% 1.00%
67

.2
8 6 4 2
.1
8 6 4 2 0
p t
Families with children

Construction
600
200 400 800 1000 1200
Average Weekly Earning (£)
Figure A1: Spending shares on goods and services by demographic groups in the UK
0
0.1
0.1
0.1
0.1
0
0.0
0.0
0.0
0.0
Notes: Data from ONS (2012). Income deciles are based on gross household income.
Figure A2: Average sectoral earnings and Brexit-induced price changes
Notes: Average weekly earnings from ONS (2016), predicted price changes based on the model by
Dhingra et al (2016a).
0 2 4 6
68
The UK Treasury analysis of
‘The long-term economic
impact of EU membership and
the alternatives’:
CEP Commentary
#CEPBREXIT
Swati Dhingra, Gianmarco Ottaviano,
Thomas Sampson and
John Van Reenen
BREXITPAPER
The UK Treasury analysis of ‘The long-term economic
impact of EU membership and the alternatives’:
CEP Commentary
This week the UK Treasury released a much-discussed Report on the economic impact of the
UK leaving the EU. Their central estimate was a 6.2% fall in GDP (£4,300 per household).
We explain and analyse the Report and conclude it is a serious contribution to the Brexit
debate. Our major criticism is that the Treasury have been overly cautious in their assumptions
for their central case (a negotiated bilateral agreement, like the Canadian trade deal) and the
true long-run costs of Brexit are likely to be higher than they estimate.
The UK Treasury recently released an analysis of the long-run consequences of leaving the
European Union (EU) for the UK economy https://www.gov.uk/government/publications/hmtreasury-analysis-the-long-term-economic-impact-of-eu-membership-and-the-alternatives.
The Report analyses the consequences of three different post-Brexit scenarios:

The Treasury’s analysis suggests that Brexit would lead to a long-run fall in UK GDP of 3.8%
under the Norway option, 6.2% under the Canada option and 7.5% under the WTO option.
But how did the Whitehall mandarins calculate these numbers? And should we believe them?
Roadmap to the Treasury analysis
The Treasury studies the long-run effects of Brexit using a three-step process:

to predict how Brexit would affect overall UK national income.
We consider each of these steps in turn.
Step 1: How would Brexit affect trade and FDI?
A ‘gravity model’ is used to estimate how leaving the EU would affect the UK’s trade (separate
estimates are calculated for goods trade and services trade) and FDI. This gravity model
approach follows best practice in the academic literature and the Treasury’s estimates are
consistent with a large body of existing work. It is an empirical model relating data on the flows
of trade between all pairs of countries to the characteristics of these countries (for example,
their respective size as measured by GDP). For example, the bilateral goods data covers over
200 countries for the period 1948-2013 leading to the 390,521 observations in Table A.1. The
bilateral flows of service trade and FDI works in basically the same way.
70
The models focus only on changes over time, so lots of things like culture and distance between
countries are controlled for. Variables indicating whether or not two countries are in the EU
(or other FTAs or the EEA) are included in these models to estimate how much more EU
members trade with each other, after controlling for all the other determinants of trade or FDI.
In the ‘Canada’ case, the Treasury estimates that Brexit would reduce UK trade by up to 19%
(Table 3.A) and FDI inflows would fall by up to 20% (Table 3.B). This is how much trade
would fall by if, instead of being in the EU, the UK simply had a ‘deep’ FTA like the imminent
EU-Canada deal.
Step 2: How would Brexit affect productivity?
The effects of Brexit on UK productivity are obtained by combining the changes in trade and
FDI from Step 1 with estimates of how these changes affect productivity. Step 2 captures the
dynamic effects of Brexit on productivity. Estimating the exact relationship between trade or
FDI and productivity is tricky. For trade, the Treasury survey a range of estimates and settle
on those from Feyrer (2011), who finds elasticities of 0.2 to 0.3 – that is, a 10% fall in trade
reduces productivity by 2% to 3%.
For the impact of FDI on productivity, the literature is thinner than for trade so the Treasury
conduct original empirical analysis. They relate productivity at the broad industry level to the
level of the FDI stock. They find an elasticity of 0.04 – that is, a doubling of the FDI stock
raises productivity by 4%.
Step 3: How do these changes affect total GDP?
The changes in trade, FDI and productivity from Steps 1 and 2 are fed into a macroeconomic
model (NIESR’s NIGEM) to forecast the overall effects of Brexit on UK GDP. This final step
also incorporates several additional effects of Brexit.
First, the short-run shock of Brexit through uncertainty and other channels is likely to cause
some permanent loss of capacity. The Treasury assume this causes a 1% decline in GDP.
Although the idea that uncertainty and other negative shocks can have large and long-run
effects is reasonable (for example, Bloom, 2014), the precise magnitude of the persistence of
the shock is hard to judge.1
Second, the fall in productivity growth will have an impact on the aggregate capital stock,
which will also reduce GDP, and the macroeconomic model helps to pin this down. Third, the
macroeconomic model allows for complex interactions between trade, investment, prices, etc.
Step 3 is the least transparent part of the analysis as the macroeconomic model has a lot of
moving parts. It is difficult to incorporate all the elements in Steps 1-3. However, given the
reasonable desire to take a comprehensive view of many elements of the economic effects of
Brexit and allow for their interactions it seems a fair attempt.
Fortunately, it is not too hard to see how the various steps contribute to the overall decline in
GDP. It seems that most of the quantitative action comes from the effect of reduced trade on
1 The Treasury has said that a future report will provide greater detail on the short-run disruption resulting from
Brexit.
71
GDP, with a smaller contribution from FDI and the persistence of the shock. The
macroeconomic model itself in Step 3 plays a more minor role.
An illustration of the Treasury calculations
To illustrate how the numbers add up we give an example in Table 1 that shows how the
different steps contribute to the overall decline in GDP. We do this for the ‘upper end of the
negotiated bilateral agreement’ (for example, Canada) but the same logic applies to the other
cases considered.
The first row of Table 1 looks at trade effects and the second row looks at FDI effects. The
trade effects come from the gravity model comparing the coefficients on EU membership to
FTA membership (a weighted average over the goods and service regressions). This is a 19%
fall for trade (column 1). Column (2) is the effect of trade on GDP (0.3). Column (3) is simply
the first two columns multiplied together, which generates the overall fall in GDP because of
a Brexit induced trade loss of 5.7% (= 19% x 0.3).
Row 2 repeats the same analysis for FDI. The gravity model predicts a fall in FDI of 20%
(column (1)), which combined with the Treasury’s estimate of the FDI effect on productivity
of 0.04 (column (2)) leads to a 0.8% (= 20% x 0.04) fall in productivity in column (3).2
If we add the effects of trade (last column of row 1) and FDI (row 2) to the 1% fall of GDP due
to the persistence of the ‘Brexit shock’ (row 3), we get in row 4 a total of 7.5% fall of GDP (=
5.7% + 0.8% + 1.0%).
The Treasury Report calculates that the overall GDP loss after feeding all the effects into their
macroeconomic model is 7.8%. The additional 0.3% (=7.8% – 7.5%) loss of GDP in getting
from row 4 to row 5 is from the general equilibrium effects introduced by the macroeconomic
model. Exactly how their additional effects work through is not entirely clear. Part of it comes
from the endogenous response of the capital stock to lower productivity. Other parts may come
from interactions between various parts of the economy. Annex A goes into some more detail
on this.
2 The reason why the sum of the effects of trade and FDI is 6.5% = (5.7% + 0.8%) instead of the 6% given in the
middle column of the Treasury Report Table 3.C is basically to avoid double counting the effects of trade in the
macro-model. Feyrer’s (2011) estimates are trade effects on GDP so include a pure productivity effect plus an
effect on the capital stock. We explain this in Annex A.
72
Table 1:
How the Treasury arrives at their effects of Brexit on GDP- the example of the
‘upper end’ of the negotiated bilateral arrangement (‘Canadian model’)
(1) (2) (3)
Row 1: Trade Step 1: Brexit effect
on trade, Tables 3.A &
A.7
Effect of trade on GDP
(elasticity), Feyrer
(2011)
Step 2: Brexit

-19% 0.30 -5.7%
Row 2: FDI Step 1: Brexit effect
on FDI, Table 3.B
Effect of FDI on
productivity
(elasticity), Table A.14
Step 2: Brexit effect on
productivity via FDI
-20% 0.04 -0.8%
Row 3: Persistent effect of Brexit
disruption shock (1% of GDP)
-1%
Row 4: Sum of effects of trade, FDI &
persistence shock
-7.5%
Row 5: Overall Effect on GDP (Step
3), Table 3.D
-7.8%
Implied additional effects of
macroeconomic model
-0.3%
Notes: CEP analysis of how to think about the Treasury’s results.
73
The Treasury’s bottom line
An important point in all of this is that the main parts of the quantitative effects are not driven
by the less transparent macroeconomic model which only accounts for a small fraction of the
overall effect (0.3% of the 7.8% total effect). The majority of the impact is through easily
understood mechanisms of trade, FDI and the permanent effect on output from a big short-run
Brexit shock.
The second important point is that most of the magnitude of the effect is coming from the
impact of lower trade on productivity: (5.7% of the 7.8%).
Similar calculations can be made for the ‘lower end of the range’ negotiated bilateral
arrangement. In this case, the fall in trade is only 14%3 (instead of 19%), the fall in FDI only
15%4 (instead of 20%) and the effect of trade on GDP reduced to 0.2 (instead of 0.3). Putting
all these together with the transition shock leads to a total of 4.4% (= 2.8% + 0.6% + 1%). If
compared to the Treasury Report’s final number of 4.6% in Table 3.D, the implication is that
the macroeconomic model adds another 0.2% of GDP to the cost. Again, Step 3 is
quantitatively small and trade effects are the largest component of the overall effect.
The headline number of a loss of 6.2% of GDP is the simple average of the lower end and
upper end estimates (i.e. = (7.8% + 4.6%)/2).
The main problem of the Treasury Report: too cautious by half?
The Treasury analysis has been criticised by some for being deliberately designed to generate
very large negative effects of Brexit. This criticism is off the mark. The assumptions and
analysis are all explicitly stated and economically reasonable. Different people will have
different views of which of the three trading arrangements is most likely after negotiations with
the EU are complete, but the analysis lays out the economic consequences of the possible
options. The 15-year horizon is sensible since such negotiations are usually extremely
protracted.
In fact, our view is that the Treasury have been too conservative in many of their assumptions
and should have generated larger effects. The following are some examples of their cautious
approach.
The impact of Brexit on the loss of trade and FDI
The Treasury’s ‘lower end’ estimates assume some of the benefits of EU membership are
‘banked’ and will remain, even 15 years after the UK leaves the EU. For example, in our Table
1, their lower end estimates are of a case where trade and FDI fall substantially less than their
empirical estimates suggest (they use 14% instead of 19% for trade and 15% instead of 20%
for FDI). There is no obvious reason to do this, apart from wanting to be very cautious.
3 Middle column of Table 3.A.
4 Middle column of Table 3.B.
74
The impact of trade on productivity
The Feyrer (2011) elasticity of 0.2 to 0.3 is conservative. This paper uses the fall in trade costs
due to the closure of the Suez Canal as a natural experiment. But Feyrer (2011) only looks at
the effects over an eight year horizon, whereas the long-run horizon the Treasury is considering
is 15 years, leading to larger dynamic effects.
A better study to use would have been Feyrer (2009) that looks over a longer period using
multiple natural experiments of falls in transport costs. This paper finds elasticities of 0.5 to
0.75 – over twice as large as those used by the Treasury. One reason for the Treasury’s more
cautious approach is that these elasticities might include some FDI component that they want
to model separately. But this seems excessively cautious as FDI flows were not so important
during the time period for which the trade elasticity is estimated (Feyrer, 2011, ends in 1975).
The impact of FDI on productivity
The approach used by the Treasury may also underestimate the benefits from FDI. First, the
industry level data used to estimate the FDI elasticity may miss out on some of the crossindustry productivity effects that the literature has stressed (e.g. across supply chains and from
learning). Second, the use of UK only data with time trends may reduce a lot of the useful
variation in FDI, making – and so underestimating – the positive effects on productivity. This
is suggested by the fact that dropping the industry trends doubles the impact of FDI on
productivity (first column of Table A.14).
An alternative, using cross country data over time, is Alfaro et al (2004) whose model captures
more of the cross industry benefits of FDI. We use this in Dhingra et al (2016b) which implies
much larger impacts of FDI on GDP, as does the earlier work of Pain and Young (2004).
Continuing reductions in trade costs within the EU
The Treasury Report assumes that there is no further integration within the EU to reduce trade
costs. But one section of the Report does look at this and suggests that it would add up to
another 4% of GDP (paragraph 3.89). It seems highly likely that there will be further reductions
in non-tariff barriers within te EU, especially in services, which is critical to the UK economy.
Over the last 50 years there has been an on-going process of reductions in within EU trade
costs, so assuming this suddenly comes to a juddering halt is unduly pessimistic.
In Dhingra et al (2016a), we model this by considering whether falls in trade costs within the
EU continue along their historical trend (business as usual) or slow down to half the historical
rates. This seems a more likely scenario than the baseline considered by the Treasury.
Immigration
The Report uses ONS projections of migration. This does not affect the results in any material
way as they are on a per capita. The Treasury are essentially assuming no impact of migration
on productivity. This is again rather cautious as most economists find positive effects of
immigration on aggregate income per head, especially since EU migrants are more educated
and younger than the UK-born (http://www.niesr.ac.uk/blog/treasurys-brexit-analysis-andimmigration#.VxiKaE32aUn).
75
Other criticisms of the Treasury Report
Is forecasting so far out impossible?
Many ‘Leave’ commentators have attacked the Report by saying it is impossible to forecast
events so far in the future. Although the Report is pitched as what will be GDP in 15 years’
time, it is important to realise that this is an analysis rather than a forecast. The Report is in
essence saying how much poorer the UK is likely to be because of Brexit. Current forecasts of
the global economy could be too pessimistic as technological change may pick up. Or they
might be too optimistic if China has a debt meltdown. But whatever happens, we can consider
what will be the state of the UK economy relative to these outcomes. And the Report is
essentially saying that whatever unexpected events happen in the world, the UK is likely be
considerably poorer than it would have been if it remained in the EU.
Fiscal transfer
The Report does not focus on net fiscal transfers to the EU. As we argue in Dhingra et al
(2016a), these are not likely to be large enough to outweigh the economic losses as the transfer
is about 0.31% of GDP. In the case of the EEA, there is likely to be a continued large transfer
(Norway pays in 88% of what the UK does on a per capita basis) and similarly large transfers
are made even in looser arrangements (for example, 40% in the Swiss case).
Benefits of cutting ‘red tape’
The Report argues there is unlikely to be much of an effect from changes in regulation
following Brexit, as the UK is already one of the most lightly regulated countries in the
developed world. This is the same conclusion that we reach in Dhingra et al (2016a).
Although the UK Government’s Impact Assessment finds that the benefits of most regulations
derived from EU laws have outweighed their costs, Booth et al (2015) point to 56 regulations
where costs outweigh the benefits. Crafts (2016) estimates that the cost of these regulations is
0.9% of the UK’s GDP. But many of these regulations implement policies that the UK
government is committed to following inside or outside the EU. For example, half of the total
cost comes from just two policies: the Renewable Energy Strategy; and the Working Time
Directive. Scrapping these regulations would mean abandoning the UK’s renewable energy
targets and removing rights such as the entitlement to 20 days paid annual leave. And even if
the regulatory costs of EU membership were 0.9% of GDP and there was the political will to
abolish these regulations, this figure is substantially smaller than the GDP decline forecast in
the Report.
‘Look globally and strike great trade deals’
Another criticism is that the document ignores the possibility that after Brexit, the UK will reorient its trade towards faster-growing economies such as China, India and the United States.
The UK actively promote trade with these countries now, but it cannot strike trade deals with
them alone.
When negotiating post-Brexit trade deals, the UK would not need to compromise with other
EU countries. On the other hand, the UK would have to take on the cost of hiring civil servants
to rebuild its capacity to undertake trade negotiations which can take decades. More
76
importantly, since the UK is under a fifth of the economic size of the EU Single Market it
would have less bargaining power in trade negotiations than the EU does. And being outside
the EU would mean it no longer automatically accessed the benefits from the EU’s trade deals,
such as the current ones being negotiated with the US and Japan worth around 0.6% of GDP.
It beggars belief that these putative trade deals would be on so much better terms than the
existing and future deals that the EU has negotiated, that they would outweigh the larger loss
of trade and investment identified in the Report.
‘Leaving the EU will shield us from its decline’
Some critics believe that the EU is about to collapse and the UK will be shielded from the fall
out by jumping from the sinking ship. It is true that the EU (like the UK) has become a smaller
part of the world economy over the last two decades. But this is not because the EU has not
grown – it has simply not grown as fast as the rising economies such as China or India. Given
the geographical fact that we are a neighbour to Europe, developments in the EU block will
always have a major effect on the UK. Throwing up masses of trade barriers with the EU (which
Brexit supporters claim they will not do) will ‘shield’ us only in the sense it would decouple
us from our major economic partners.
Conclusions
Forecasting the economic consequences of Brexit is a difficult challenge and all estimates will
be subject to a degree of uncertainty. But our overall assessment is that the Treasury Report is
a credible analysis, which, for the most part, uses the best available estimation methods. Our
main criticism is that they have taken overly cautious assumptions which will tend to
underestimate the economic costs of Brexit.
Their headline forecast that Brexit would reduce long-run UK GDP by 6.2% in the baseline
case is broadly consistent with our previous work and many other independent estimates. For
example, in Dhingra et al (2016a), our dynamic estimates of the cost of Brexit indicate a GDP
loss of 6.3% to 9.5% in the case of moving from the EU to EFTA (see Annex B). Treasury
estimates are at the lower end of this range.
The Treasury’s findings reinforce the academic and business consensus that Brexit would make
the UK significantly poorer. The Report is a serious contribution to the debate.
77
Further Reading
Alfaro, L., A. Chanda, S. Kalemli-Ozcan and S. Sayek (2004) ‘FDI and Economic Growth:
The Role of Local Financial Markets’, Journal of International Economics 64(1): 89-112.
Bloom, N. (2014) ‘Fluctuations in Uncertainty’, Journal of Economic Perspectives 28(2) 153-
76.
Booth, S., C. Howarth, M. Persson, R. Ruparel and P. Swidlicki (2015) ‘What If..? The
Consequences, Challenges and Opportunities facing Britain outside the EU’, London: Open
Europe.
Crafts, N. (2016) ‘The Growth Effects of EU Membership for the UK: A Review of the
Evidence’, University of Warwick mimeo.
Dhingra, S., G. Ottaviano, T. Sampson and J. Van Reenen (2016a) ‘The Consequences of
Brexit for UK Trade and Living Standards’, CEP Brexit Analysis No. 2
(http://cep.lse.ac.uk/pubs/download/brexit02.pdf).
Dhingra, S., G. Ottaviano, T. Sampson and J. Van Reenen (2016b) ‘The Impact of Brexit on
Foreign Investment in the UK’, CEP Brexit Analysis No. 3
(http://cep.lse.ac.uk/pubs/download/brexit03.pdf).
Feyrer, J. (2009) ‘Trade and Income – Exploiting Time Series in Geography’, NBER Working
Paper No. 14910.
Feyrer, J. (2011) ‘Distance, Trade and Income – The 1967 to 1975 Closing of the Suez Canal
as a Natural Experiment’, NBER Working Paper No. 15557.
Pain, N. and G. Young (2004) ‘The Macroeconomic Effect of UK Withdrawal from the EU’,
Economic Modelling 21: 387-408.
78
Annex A: How the macroeconomic model affects the results
In NIESR’s NIGEM macroeconomic model the Treasury simulate for each of their scenarios:

The macroeconomic model then calculates the long-run changes in GDP, prices, etc. These are
compared to the baseline case without these shocks (i.e. the case if the UK stayed in the EU).
One tricky issue is the size of the technology shock. Consider our example in Table 1 of the
upper end of the case of the negotiated bilateral agreement (Canada). An obvious strategy
would be simply to use the implied productivity effect of -5.7% in Table 1 (column (3) of row
1). However, this would imply some ‘double counting’, as Feyrer’s (2011) estimates of the
impact of trade changes on income already include productivity increases due to endogenous
increases in the capital stock that trade induces.
The Treasury’s solution is to figure out what is the size of the productivity/technology shock
necessary to generate a change of GDP equal to 5.7%. They do this by using the NIGEM model
and the assumed change in exports (or its equivalent in a shrinkage of the exporter market size).
It turns out that the implied magnitude of the technology shock is 5.2%.5 They then add the
productivity impact of FDI to get to a total technology shock of 6% (5.2% + 0.8%) They then
add this shock of 6% into the NIGEM model alongside the trade shock to simulate the new
level of GDP and other macro-economic outcomes.
5 We can tell this because Table 3.C indicates that a 6% technology shock has been fed into the NIGEM model,
and the impact of FDI on productivity is 0.8% in our Table 1 (see column (3) of row 2). So 5.2% = 6.0% – 0.8%.
79
Annex B: How the Treasury Report relates to our own estimates
The closest calculation that is comparable is the dynamic reduced form approach in Dhingra et
al (2016a). There we find a 12.5% fall in trade after Brexit from a free trade agreement (like
the Canadian case or EFTA). So this is similar to the Treasury’s ‘lower end’ estimate of 14%.
We use a higher GDP-trade elasticity from Feyrer (2009) of 0.5 to 0.75, which gives us a 6.3%
to 9.5% GDP loss. These also probably include some FDI response.
The Treasury’s baseline 6.2% GDP loss is at the lower end of these range of estimates.
The focus of Dhingra et al (2016a) is on static estimates from the costs of trade using a
structural model. This finds smaller losses of GDP of between 1.3% and 2.6%. This is a very
different methodology to the Treasury approach. In particular, it does not consider the dynamic
effects of trade on productivity. In addition it does not consider the effects of FDI or
immigration. In Dhingra et al (2016b) we take a simpler approach to looking at the effects of
FDI and find that Brexit would cause a 3.4% loss of GDP if we moved into EFTA.
80
Economists for Brexit:
A Critique
Thomas Sampson, Swati
Dhingra, Gianmarco Ottaviano
and John Van Reenen
#CEPBREXIT
BREXITPAPER
Introduction
Much publicity recently surrounded ‘Economists for Brexit’ (2016a, 2016b). Since the
economic case for leaving the European Union (EU) has been largely missing in action, it is
refreshing to obtain some clarity over the Leave campaign’s vision of the UK’s post-Brexit
economic arrangements.
Professor Patrick Minford of Cardiff University is the only one of the group who has provided
some economic modelling. He predicts that there would be a welfare gain of 4% of GDP by
2020 if the UK were to leave the EU. This prediction is surprising because just about every
other piece of economic analysis finds negative economic effects from the UK leaving the EU
(for example, Dhingra et al, 2016a; HM Treasury, 2016; NIESR, 2016; OECD, 2016; PWC,
2016).
Such studies simply remind people that the EU has been good for trade and trade is good for
welfare. It follows that leaving the EU will reduce trade and so have an economic cost. The
main question is not so much the direction of the effect, but rather the magnitude of the hit to
living standards. There may be offsetting or reinforcing factors from other things – such as
regulation, foreign investment, immigration, lower fiscal transfers to Brussels, uncertainty and
so on – but distancing ourselves from our closest trading partner could not be beneficial for
trade.
Yet Economists for Brexit make just such a claim, so we were curious to understand where
Minford’s positive effects come from. We summarise the main points of our analysis here and
leave more technical details to the Annex.
‘Britain Alone’ – unilateral free trade
One feature of Minford’s approach is that after leaving the EU, the UK is assumed to trade
simply under World Trade Organization (WTO) rules, without seeking a new trade agreement
with the EU or other trading partners like the United States. The UK would simply pay the
external tariffs. This is usually the worst case scenario that other economists have modelled.
HM Treasury (2016), for example, finds a GDP drop of 7.5% under this scenario.
A second feature of the Minford argument is the assumption that the UK will unilaterally drop
all its trade protection against imports from everywhere else in the world after Brexit. One
reason why most economists have not focused on this scenario is that it seems politically
unlikely. As far as we know, no developed country has ever unilaterally removed all
manufacturing tariffs against all other countries – Minford’s ‘Britain Alone’ scenario.
In fact, one can easily imagine the UK establishing greater trade protection after Brexit. For
example, the recent furore over Port Talbot’s steelworks suggests that domestic political
pressures may have pushed the government to increase tariffs on Chinese steel. Indeed,
Minford describes in his model that an implication of the ‘Britain Alone’ policy (Minford et
al, 2016, p. 74 Table 4.3) is that it will ‘effectively eliminate manufacturing’ in the UK. Another
implication of his preferred policy is a dramatic increase in wage inequality: skilled workers’
nominal wages increase by around 11%, but unskilled workers’ wages fall by 14%.1 These
changes are unlikely to be an easy sell politically, to say the least.
1 These inequality changes will not be offset by reductions in EU immigration as the impact of immigration on
inequality is close to zero (Wadsworth et al, 2016).
82
Nonetheless, standard economics does suggest that there will be some benefits from ‘unilateral
trade disarmament’. Indeed, in the work we published in March (Dhingra et al, 2016a, Table
2) we look at what would happen if the UK eliminated all tariffs after Brexit.2 We find that if
the UK trades under WTO rules following Brexit, but maintains import tariffs, then UK income
per capita falls by 2.6%. Under Minford’s ‘Britain Alone’ scenario of unilateral liberalisation
after Brexit only, UK real incomes still fall by 2.3%. In other words, there is a gain of only 0.3
percentage points from eliminating tariffs compared to just trading under WTO rules – this is
completely insufficient to offset the other trade costs of Brexit.
So the real question is not whether moving to unilateral free trade can have some benefits in
economic models, but rather:

The answers to these questions require an understanding of how thinking about the economics
of trade has developed in the last five decades, and how these features are overlooked in the
Minford approach.
The basic idea
There are basically two steps in Minford’s analysis. First, he assumes that because of EU tariff
and non-tariff barriers, prices paid by UK consumers for manufacturing and agricultural goods
would fall by 10% under his ‘Britain Alone’ policy recommendation. Second, he feeds this
10% tariff equivalent fall in trade costs into his ‘Liverpool model’ to come up with a GDP
increase of 4% (roughly speaking, the increase in GDP is much less than 10% because people
consume a lot of services, which are not directly affected by Brexit under Minford’s
assumptions).
The 10% fall in trade costs
How on earth can trade costs fall by 10% when the UK’s average tariff is currently around 3%?
The answer is that the 10% number does not come from looking at the actual level of tariffs. It
comes from looking at the differences in price levels between the UK and some other countries
and arguing that these higher prices are due to protectionism caused mainly by EU regulations
(non-tariff barriers).
We go through this in detail in the Annex, but there are several very basic problems. First, the
estimates he makes are from data in 2002 – 14 years out of date.
Second, it seems extraordinarily unlikely that all the cross-country price differences are really
from trade protection rather than a multitude of other factors, such as quality differences,
variation in producer mark-ups or measurement error in estimates of distribution margins. For
example, say Europeans put a higher premium on high-quality clothing compared with
Americans. It will look like Europeans are paying more for their clothes, but in reality, the
higher average prices simply reflect a different mix of purchases – we are comparing apples
2 Mr. Minford is under the misapprehension that we did not look at his ‘Britain Alone’ recommendation, but this
is because he only refers to the work from two years ago (Ottaviano et al, 2014) and not the recent work (for
example, Dhingra et al, 2016a).
83
with oranges across countries (Deaton, 2015)4 Minford attributes these price differences to
nefarious EU regulation excluding cheaper clothing, whereas in fact it could reflect different
tastes for quality.
It is true that regulations could mean that prices are higher in the EU as there are stricter quality
controls than in other countries. The EU has tougher regulations over children’s milk and toys
than China does, so sub-standard products cannot be sold. This does create a trade barrier with
China and in Minford’s data a children’s toy will appear as identical, but more expensive in
the EU. But this reflects a quality difference. It is true that if the UK left the EU and relaxed
the safety standards down to China’s level prices would fall. But quality-adjusted prices would
not, and this is what is relevant for consumer welfare.
Third, Minford misunderstands the nature of regulations and product standards. The idea of the
Single Market is to have common rules so that a product sold in one EU country can also be
sold in any other. If there are 28 different sets of rules governing the sale of a product, it will
be harder to sell this product across all EU countries. The basic misconception in Minford’s
world is that the harmonisation of regulations between EU countries to reduce trade barriers is
simply a pernicious plot by vested interests to raise prices. In fact, playing by a common set of
rules is what helps increase trade and competition in a modern economy. Modern trade
agreements are hard because countries are trying to agree on common standards and to
harmonise rules that are different.
Minford overestimates the scope for reducing trade costs through unilateral liberalisation. In
our analysis of unilateral liberalisation, we focus on the removal of import tariffs because tariffs
are measurable and, in the event of Brexit, could be removed at the stroke of a pen.
One way to align standards is simply by co-ordinating on one rule or another; there is no better
or worse, weaker or stronger. But it takes two to tango. There simply is no way of unilaterally
aligning these type of standards. If the UK simply goes its own way on its own regulatory
standards, then this will increase the costs of trading with European countries and reduce the
amount of trade.
Other forms of harmonising rules require explicit agreement on how ‘tough’ a product standard
must be. Consider safety standards for children’s toys. Some countries may have very relaxed
standards over toy safety, but others may have very high standards. Let’s say the EU settles on
a standard for toys that is tougher than the UK would unilaterally choose, but weaker than
Germany would like.3 The single standard means that all manufacturers know that so long as
they meet the safety requirements, they can sell toys anywhere in the EU.
The high product standard is annoying for UK toy manufacturers, some of whom will now
have to comply with the EU Toy Safety Directive, even when they do not export to the EU.
They will complain that it’s only exporters that should have to comply with the higher EU
safety standards, as most Britons don’t care. But if the UK gets an ‘opt out’ to produce low
safety toys for domestic consumption, it means that there isn’t a level playing field – every
country will want an opt-out to decrease or increase the standard.
3 There are many other examples of such regulations – see Springford (2016). Examples include powdered milk:
http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=OJ:L:2007:258:0027:0028:EN:PDF; levels of arsenic
found in rice products for children:
https://www.food.gov.uk/enforcement/regulation/europeleg/euupdates/january-update-maximum-limits-forarsenic; and of course the classic case on lawnmower noise: https://next.ft.com/content/ac04efc8-34c8-11e3-
a13a-00144feab7de.
84
In our example, the UK can sell high quality safe toys to Germany, but German toy
manufacturers can’t sell lower quality products to the UK as they are banned from producing
them. This is not just a political problem. UK consumers are worse off because locking out
foreign competition means that they face higher prices and less innovation.
Another practical problem with multiple standards is that with complex supply chains,
countries may not want to import from others with lax standards solely for domestic
consumption as it might contaminate the entire batch. This is why the EU and even the United
States want to have a global standard for toy safety (http://uk.reuters.com/article/health-toyssafety-rules-dc-idUKL0889219620071108?pageNumber=1).
In this context, Minford’s ‘Britain Alone’ proposal would be that we leave the EU and lower
product standards. It is certainly possible for the UK to adopt the lowest standards unilaterally.
There would then be lower average prices and quality for children’s toys in the UK. But even
if this was what the British people wanted, the rest of the EU would not continue to grant the
same access to the Single Market as EU toy manufacturers would be excluded from part of the
UK market because of higher EU standards. This is why the EU insists that countries play by
the same rules if they want to be in the Single Market.
How Minford defies the laws of gravity
The gravity equation is the most reliable empirical relationship in international economics.
First estimated by Nobel laureate Jan Tinbergen, it shows that the trade flow between any pair
of countries increases as the economic size of the countries grows, and decreases with rising
costs of trade between them caused by import tariffs, transport costs and other trade barriers.
Geography matters – the further apart countries are, the less they trade. There are literally
hundreds of data-based studies showing the robustness of this relationship across many
countries, industries, time periods and multiple specifications (see Head and Mayer, 2014, for
a survey).
Today, the gravity equation is central to how economists understand international trade. It is a
key economic relationship, which performs extremely well in predicting actual trade flows. To
evaluate the effects of changes in trade policy, new methods have been developed that both
explain why trade follows a gravity equation and takes into account all the general equilibrium
effects of changes in trade policy on prices, wages and output in a multi-country, multi-industry
world with trade in both final goods and intermediate inputs.4
This matters because the analysis undertaken by Minford uses an old trade model in which all
firms in an industry everywhere in the world produce the same goods and competition is perfect
so that trade does not follow the gravity equation. This choice is largely responsible for why
Minford’s findings contradict the results of numerous other studies that conclude Brexit would
lower UK GDP.
How does his analysis work? With perfect competition and homogeneous products, the EU’s
tariffs and other regulations raise the price of imports and all other goods sold in the EU above
the free trade price. Therefore, if the UK leaves the EU and simultaneously removes all tariffs
and non-tariff barriers, prices fall, making the average UK consumer better off.
4 We use such a model to analyse the consequences of Brexit in Dhingra et al (2016a), which builds on Ottaviano
et al (2014). Minford (2016) is mistaken in thinking that general equilibrium effects are missing from this analysis.
They are not missing; they are just incorporated into a richer (but more transparent) model than the one he uses.
85
The problems with this analysis stem from the assumption that all firms in an industry produce
the same product. There are two main limitations with Minford’s model:
1. Exporters sell all their output at world prices. In reality, exporters sell their output in many
countries and face different trade barriers in each market. In Minford’s model, following
Brexit exporters sell all their output in a fictional world market. Consequently, the level of
trade barriers with the EU after Brexit does not matter to exporters as they do not care
whether trade goes to the EU or elsewhere. This feature of the model gets rid of the costs
of Brexit from reduced access to the EU market. In reality, as our geographically closest
neighbour, we will continue to trade with the EU. Brexit increases trade costs with the EU
and this causes us to trade less with them. We cannot just sell everything to the rest of the
world at the same price to make up for this loss. This is the primary cost of Brexit, but it is
absent from Minford’s model.
2. Both imports and domestic output have the same price. Therefore, the decline in import
prices when the UK removes import tariffs leads to an equal fall in domestic prices. In the
real world, domestic and foreign firms produce differentiated products, so a fall in import
prices will reduce domestic prices to a smaller extent and the benefits from unilateral trade
liberalisation are much smaller.
Comparing Minford’s approach with modern trade models
As noted in Dhingra et al (2016a, Table 2), we analyse the consequences of unilateral trade
liberalisation following Brexit in a modern general equilibrium trade model that is consistent
with the gravity equation. In this experiment, we continue to assume perfect competition, but
allow for product differentiation. This means that there is not perfect substitution between the
products of any given industry and thus the cheapest source country of a product is not
necessarily the best source country of all products in the industry. Our model also allows for
varieties of the same product to be sourced from different countries as consistent with different
consumer tastes.
Under Minford’s assumption that the UK trades simply under WTO rules and unilaterally
removes all import tariffs, we find that the Brexit effect is equivalent to a 2.3% decline in UK
income per capita. We conclude that Brexit would reduce UK living standards even if the UK
unilaterally sets all import tariffs to zero.
We also consider what happens if the UK falls back on general WTO rules and imposes the
EU’s current ‘most-favoured nation’ tariffs following Brexit. In this case, Brexit is equivalent
to a 2.6% decline in UK income per capita. Thus, Minford is right that there are benefits from
removing import tariffs, but these benefits are around 0.3 percentage points – much smaller
than the costs of Brexit resulting from increased trade costs with the EU.
In addition to satisfying the gravity equation, our model predicts that EU membership is tradecreating, which means that it increases the UK’s trade. By contrast, Minford’s model is hardwired to predict that EU membership is only trade-diverting and Brexit would lead to higher
trade.
Our work and that of the other economic studies relies on data that show what has actually
happened to trade after joining the EU, rather than just asserting what should happen in a
theoretically dubious model.
The empirical research literature supports the conclusion that EU membership is trade-creating.
For example, Baier et al (2008) find that goods trade between EU members is 62% higher than
86
trade between otherwise comparable countries that have no trade agreement between them.5
Using more recent data, HM Treasury (2016) finds that EU membership raises intra-EU goods
trade by an even larger 115% relative to WTO membership.
There is little evidence that regional trade agreements lead to substantial trade diversion – see
the recent reviews by Bagwell et al (2014) and Limão (2016). For example, Magee (2008,
2016) finds no evidence of trade diversion from economic integration agreements. Consistent
with this evidence, HM Treasury (2016) finds no significant evidence of trade diversion
because of the EU.
Services
Services exports to the EU accounted for 16% of all UK exports in 2014 (ONS, 2015). UK
services exporters benefit from lower trade barriers with the EU resulting from the Single
Market. In particular, financial services firms can undertake business throughout the EU under
the EU’s ‘passporting’ rules. These rights would be lost if the UK left the Single Market.
Minford does not take this into account.
Foreign investment
Membership of the EU increases foreign direct investment (FDI) in the UK, which raises
productivity and output (Dhingra et al, 2016b). Minford argues that there are no benefits from
FDI, whereas the evidence points in the opposite direction. His views seem to be based on the
fact that the empirical estimates are ‘insecure’ (Minford, 2016) without saying why.
The role of empirical evidence in Minford’s world
Minford’s style of work was popular in some quarters in the 1970s. In those days, economics
did not need to be well-grounded in facts and data, and could rely on highly simplified theories.
The revolution over the last 40 years has been the explosion of data and empirical techniques
for its analysis. Good theory has evolved in tandem with this new evidence.
Theoretical foundations, ranging from Ricardian comparative advantage to modern product
differentiation models of imperfect competition, explain why the gravity model describes
international trade flows (for example, Head and Mayer, 2014). The approach that we use in
Dhingra (2016a) is to employ a model consistent with the basic facts of trade. It is a computable
general equilibrium model and well-grounded in theory. The difference between our approach
and Minford’s is that the theories we use are based on the facts of life in trade, such as the
gravity relationship, whereas his theory is unhinged from the most basic features of trade
reality.
Minford’s attitude seems to be that if empirical work is imperfect, it should be ignored. The
voluminous evidence on the positive impact of the EU on trade is dismissed because of
statistical concerns (Economists for Brexit, 2016b, p.20). Of course there are issues with all
empirical work. Some of these problems might mean we over-estimate the EU’s effect on trade
and FDI; some might mean we under-estimate it.
5 This estimate comes from Table 6, column 1 of Baier et al (2008) where 62% = e0.48 – 1.
87
But to take the position that since no econometric work can be perfect, all inconvenient facts
should be ignored is poor scholarship and bad science.
Conclusions
Alternative economic models have different advantages and drawbacks and are suited for
different purposes. Unfortunately, Minford’s model is inconsistent with two basic facts about
international trade; first that trade satisfies the gravity equation; and second, that the EU has
been trade-creating, not simply a tool for trade diversion.
Consequently, Minford’s model is not the right tool to use for predicting the consequences of
Brexit for trade and living standards. When we analyse the same scenario considered by
Minford using modern economics that incorporate advances in our understanding of
international trade data since the 1960s and a more realistic assessment of how UK ‘unilateral
trade liberalisation’ could actually work, we find (alongside just about everyone else) that
Brexit leads to a decline in UK living standards.
88
Further reading
Baier, S.L., J.H. Bergstrand, P. Egger and P.A. McLaughlin (2008) ‘Do Economic Integration
Agreements Actually Work? Issues in Understanding the Causes and Consequences of the
Growth of Regionalism’, The World Economy 31(4): 461-97.
Bagwell, K., C. Brown and R. Staiger (2014) ‘Is the WTO Passé?’ Stanford mimeo
(http://www.dartmouth.edu/~rstaiger/JEL_WTO_BBS_Draft_111014.pdf).
Bradford, S. (2003) ‘Paying the Price: Final Goods Protection in OECD countries’, Review of
Economics and Statistics 85(1): 24-37.
Deaton, A. (2014) ‘Getting Prices Right: The Mysteries of the Index’, LSE Lionel Robbins
Memorial Lectures
(http://www.lse.ac.uk/publicEvents/events/2014/12/20141210t1830vOT.aspx).
Dhingra, S., G. Ottaviano, T. Sampson and J. Van Reenen (2016a) ‘The Consequences of
Brexit for UK Trade and Living Standards’, CEP Brexit Analysis No. 2
(http://cep.lse.ac.uk/pubs/download/brexit02.pdf).
Dhingra, S., G. Ottaviano, T. Sampson and J. Van Reenen (2016b) ‘The Impact of Brexit on
Foreign Investment in the UK’, CEP Brexit Analysis No. 3
(http://cep.lse.ac.uk/pubs/download/brexit03.pdf).
Economists for Brexit (2016a) ‘The Economy After Brexit’
(http://issuu.com/efbkl/docs/economists_for_brexit_-
_the_economy/1?e=24629146/35248609).
Economists for Brexit (2016b) ‘The Treasury Report on Brexit: A Critique’
(https://issuu.com/efbkl/docs/economists_for_brexit_-_the_treasur).
Head, K. and T. Mayer (2014) ‘Gravity Equations: Workhorse, Toolkit, and Cookbook’,
Chapter 3 in Gopinath G., E. Helpman and K. Rogoff (eds) Handbook of International
Economics Vol. 4: 131-95, Elsevier.
HM Treasury (2016) ‘The Long-term Economic Impact of EU Membership and the
Alternatives’, London: HMSO
Limão, N. (2016) ‘Preferential Trade Agreements’, NBER Working Paper No. 22138.
Magee, C. (2008) ‘New Measures of Trade Creation and Trade Divergence’ Journal of
International Economics 75(2), 349-62.
Magee, C. (2016). ‘Trade Creation, Trade Diversion, and the General Equilibrium Effects of
Regional Trade Agreements: A Study of the European Community-Turkey customs union’,
Review of World Economics 152(2): 383-99.
Minford, P. (2015) ‘Evaluating European Trading Arrangements’, Cardiff Economics Working
Paper No. E2015/17 (http://patrickminford.net/wp/E2015_17.pdf).
89
Minford, P. (2016) ‘Understanding UK Trade Agreements with the EU and Other Countries’,
Cardiff Economics Working Paper No. E2016/1 (http://patrickminford.net/wp/E2016_1.pdf).
Minford, P., S. Gupta, V. Le, V. Mahambare and Y. Xu (2016) Should Britain Leave the EU?
An Economic Analysis of a Troubled Relationship, Second Edition, IEA.
NIESR (2016) ‘The Economic Consequences of Leaving the EU’, May Special Issue.
ONS (2015) ‘National Accounts: Balance of Payments’ HMSO: London
https://www.ons.gov.uk/economy/nationalaccounts/balanceofpayments/datasets/pinkbook
OECD (2016) ‘The Economic Consequences of Brexit: A Taxing Decision’, OECD
(http://www.oecd.org/economy/the-economic-consequences-of-brexit-a-taxing-decision.htm).
Ottaviano, G., J. Pessoa, T. Sampson and J. Van Reenen (2014) ‘The Costs and Benefits of
Leaving the EU’, CEP Policy Analysis (http://cep.lse.ac.uk/pubs/download/pa016_tech.pdf).
PWC (2016) ‘Leaving the EU: Implications for the UK Economy’
(http://www.pwc.co.uk/economic-services/assets/leaving-the-eu-implications-for-the-ukeconomy.pdf).
Springford, J. (2016) ‘Brexit and EU Regulation: A Bonfire of the Vanities’, Centre for
European Reform (http://www.cer.org.uk/publications/archive/policy-brief/2016/brexit-andeu-regulation-bonfire-vanities).
Wadsworth, J., G. Ottaviano, T. Sampson and J. Van Reenen (2016a) ‘Brexit and the Impact
of Immigration on the UK’ http://cep.lse.ac.uk/pubs/download/brexit05.pdf
90
ANNEX: Some other limitations of Minford’s analysis
How is the 10% higher prices in EU calculated?
Minford claims that UK goods and food prices will fall by 10% after Brexit. Where does this
figure come from? He draws on the methods of a paper by Bradford (2003), which looks at
prices of around 3,000 goods sold in several OECD countries from 1993. Minford and his coauthors have tried to update this using data up to 2002 (Minford et al, 2016). The average of
the EU (Belgium, Germany, Italy, Netherlands and the UK) is estimated to be 21% above the
lowest cost OECD supplier, half the level of the early 1990s. They extrapolate their 2002
numbers forward by another two decades to 2020, claiming protection levels now raise prices
by 10%.
Before getting into the details of the methods, it is worth noting that:

that by 2020 price levels in the EU should be 0% above world prices not 10%.
Producer prices are not directly observed. So Minford’s approach is to start from consumer
prices and deduct a distribution margin and a trade costs adjustment (at a higher level of
aggregation than the disaggregated product prices) to estimate ‘producer prices’.
The weighted average EU tariff on goods is only about 3%, so the additional 7% comes from
the assumption that there are various non-tariff barriers holding up the prices of EU goods.
Minford gives no direct evidence of this, but offers various vague appeals to threats of antidumping actions, which would mysteriously end completely if the UK were to leave.
As we discuss above in the main text of this analysis, the regulations of the Single Market are
in large part designed to harmonise standards across EU members, with the goal of reducing
trade costs between them. Empirical evidence shows that this has been successful.
It is extraordinarily unlikely that all the cross-country price differences are really from trade
protection rather than a multitude of other factors, such as quality differences, variation in
producer mark-ups or measurement error in estimates of distribution margins.
The entire exercise is deeply flawed empirically and conceptually.
Gravity
Fifty years ago economists could not explain the success of the gravity equation. Traditional
trade models assume a perfectly competitive economy where all firms in an industry produce
the same good. Such models do not predict that trade flows will satisfy the gravity equation.
Instead, they predict that consumers in the UK purchase each good from whichever country is
the cheapest supplier. For example, all cars purchased in the UK would come from whichever
country could supply cars to the UK at the lowest cost. In reality, the UK imports cars from
many different countries, not just the cheapest, because cars are of different qualities are styles.
91
Since the late 1970s, a series of breakthroughs in international economics have helped to
explain why the gravity equation holds. The answer is product differentiation. Products
produced by different countries or different firms are not the same. German cars are not perfect
substitutes for Japanese cars; and consumers care whether their car is a BMW or a Toyota.
Allowing for imperfect substitution in trade models (often also accompanied by imperfect
competition) has enabled economists to explain why trade flows satisfy the gravity equation.
Contrast with CEP modelling
Minford argues for the superiority of his approach over the modern gravity-based trade models
(Minford, 2015).
He argues in favour of his theory-based computable general equilibrium (CGE) model rather
than our gravity model. In fact, our model is a CGE model as well. It is also a theory-based
general equilibrium model, but it is a more general set-up than Minford’s. We have imperfect
substitution of products within industries and across different countries, so that the Minford
approach is a special case of our model. The idea that the gravity approach is not theoretically
grounded is absurd. Minford may not like the theory, but it is a theory consistent with the basic
facts of trade, which his model is not.
Minford argues that a problem with our approach is that we have to make assumptions over the
responsiveness of trade flows to trade costs and that these trade elasticities may be wrong. He
is correct in that there is uncertainty over the exact magnitude of the elasticities (they are laid
out in Table A4 on p. 31 of the Technical Appendix in Dhingra et al, 2016a). We take these
industry-specific elasticities from the best current estimates from peer-reviewed journals.
Minford is arguing that rather than take data-determined estimates, it is better to use his theory,
which assumes that the elasticity for every industry is equal to infinity! In his world, a one
penny difference in price induces everyone to buy every car from the other producer. This is
an absurd position.
The data that we use for our model are much richer. His model considers four regions whereas
ours considers 35 regions, which allows us to look at the effects in many other countries. We
use trade flow data at a disaggregated level of 31 industries across all pairs of countries, so that
we can accurately model changes in trade flows across industries and countries. Minford uses
only three aggregate sectors: agriculture, manufacturing and services.
One thing his model allows for that ours does not is to solve for the market equilibrium in four
factor inputs – capital, land, skilled labour and unskilled labour. Capital is assumed to be
determined on the world market, but the other factors are in fixed supply on the home market.
Hence, he models how factor prices change in response to tariff changes. We abstract away
from these distributional effects in our modelling as we are focused on the overall welfare
effects of Brexit. It is unclear what this more complex modelling structure buys him in terms
of the effects of Brexit (except perhaps to highlight the increase in wage inequality that would
occur).
History repeating itself?
Building a model that does such violence to basic facts of economic life is why the ‘Liverpool
model’ has such a poor record of accurately analysing major policy changes. Minford was
predicting huge job losses from the introduction of the National Minimum Wage in 1999. In
the event, multiple studies have shown there was effectively no increase in unemployment.
92
Minford’s claims were based on models of homogenous workers and perfect competition. CEP
analysis (http://cep.lse.ac.uk/pubs/download/cp290.pdf), by contrast, respected the data and
developed models that allowed for labour market imperfections. We showed that minimum
wages sensibly introduced could reduce wage inequality without increasing unemployment
93
BREXIT 2016
#CEPBREXIT
PAPERBREXIT
Technical Papers
#CEPBREXIT
PAPERBREXIT
TECHNICAL PAPER
The costs and benefits
of leaving the EU: Trade effects
The Costs and Benefits of Leaving the EU: Trade Effects
Swati Dhingra
LSE/CEP
Hanwei Huang
LSE/CEP
Gianmarco Ottaviano
LSE/CEP
Jo~ao Paulo Pessoa
FGV-EESP/CEP
Thomas Sampson
LSE/CEP
John Van Reenen
LSE/CEP
March 17, 2016
Abstract
What would be the economic consequences of leaving the European Union for living standards in the UK? We estimate the welfare effects of changes in trade and fiscal transfers following
Brexit. We use a standard quantitative general equilibrium trade model with multiple sectors,
countries and intermediates, as in Costinot and Rodriguez-Clare (2013). Static losses range
between 1.28% and 2.61%, depending on the assumptions used in our counterfactual scenarios.
The finding that Brexit reduces UK living standards is robust to a wide range of alternative
assumptions about what follows Brexit. A more reduced form approach that includes dynamic
effects would triple such losses to between 6.3% and 9.5%.
Keywords: Trade, European Union, welfare
JEL Classification:
Acknowledgements: We would like to thanks the ESRC for financial support through the
Centre for Economic Performance. We would also like to thank Arnaud Costinot for sharing his
programmes and Robert Feenstra for helpful comments.
96
1 Introduction
In January 2013 Britain’s Prime Minister David Cameron committed to holding a referendum on
EU membership which will now take place on June 23rd 2016. Supporters of Brexit” focus on
the democratic benefits of repatriating powers back from Brussels. Supporters of the EU argue
that the EU reduces the risk of conflict and makes Britain stronger on the world stage. These are
important issues, but this technical report focuses purely on the economic costs and benefits, in
particular from changes in trade.
Eurosceptics (Morris, 2013) believe that trade with EU countries would not be much affected
by an exit because UK’s trade deficit with the EU provides enough bargaining power to allow the
negotiation of a free trade agreement between the two parties, similar to that enjoyed by Norway
or Switzerland. Secondly, Britain would be able to expand its trade with non-EU countries through
the negotiations of new trade agreements that would not be subjected to constraints imposed by
other EU members. Thirdly, the UK country would be free from the regulatory burden and the
costs associated with the EU membership.
Europhiles (Springford and Tilford, 2014) argue that it is unrealistic to expect the same trade
terms as smaller countries like Norway or Switzerland, that the UK’s ability to strike trade deals
with other countries will be weakened, not strengthened outside the EU and that the costs of
regulation solely due to the EU are vastly exaggerated. Furthermore, there are many other aspects
that need to be taken into consideration (Harari and Thompson, 2013).
In this paper we focus on the welfare gains arising from trade openness with EU countries to
quantify some of the effects associated with an eventual withdrawal from the EU. Our methodology
is based on Costinot and Rodriguez-Clare (2013). We set up a general equilibrium trade model
which covers 31 sectors and aggregates the world into 35 regions. We define distinct scenarios after
Brexit and calculate the changes in welfare as measured in real consumption. The welfare loss from
Brexit is obtained by comparing the welfare one in which UK remains to be an EU member and
one in which the UK does not. We find that increases in bilateral tariffs and non-tariff barriers
between the UK and the EU and exclusion of the UK from future integration of the EU lead to a
drop in UK welfare even after accounting for lower fiscal transfers to the EU. The welfare changes
range from -1.28% in an optimistic scenario, to -2.61% in a pessimistic one. We carry out various
checks to test the robustness of our results.
The welfare loss is not limited to the UK. EU countries that trade intensively with the UK tend
97
to lose more. For example, Ireland suffers the largest losses from Brexit. For countries outside the
EU, they tend to have a small welfare gain, mostly due to a trade diversion effect. As a whole,
however, the world beyond the UK’s shores is poorer after Brexit.
In our basic setup, trade liberalization tends to increase welfare due to the specialization of
countries in their areas of comparative advantage 1 and the availability of cheaper goods and services
and/or cheaper inputs (Eaton and Kortum, 2002). Our baseline calculations, however, leave out
many factors that could lead to further losses following an exit from the EU. For example, the
reduction in the variety of goods and services available for consumption (Krugman, 1980) and the
fall in productivity due to weaker competition from abroad (Melitz, 2003), and the presence of
vertical production chains in the UK (Melitz and Redding, 2014) will most likely increase such
losses. Furthermore, contrary to popular belief, ceasing migration flows between UK and other EU
countries, one of the EU most basic principles, will also tend to decrease welfare not only in the
source region but also in the destination one (di Giovanni, Levchenko, and Ortega, 2015).
Our main analysis is also static in nature and gains from trade can be considerably larger when
we factor in dynamic effects. Trade openness can increase growth rates due to a rise in investment
in capital (Wacziarg, 1998), increases in technology diffusion (Sampson, 2016; Wacziarg, 1998),
export learning effects (Albornoz, Calvo Pardo, Corcos, and Ornelas, 2012; Egger, Larch, Staub,
and Winkelmann, 2011) and greater investment in R&D (Bloom, Draca, and Van Reenen, 2015;
Keller, 1999, 2002).
An alternative way to evaluate the impact of Brexit and take into account part of these dynamic
effects is to use the results of simple, less theory-based empirical studies of the effects of EU
membership. Baier, Bergstrand, Egger, and McLaughlin (2008) find that, after controlling for other
determinants of bilateral trade, EU members trade substantially more with other EU countries than
they do with members of the European Free Trade Association (EFTA). Their estimates imply that,
if the UK leaves the EU and joins EFTA, its trade with countries in the EU will fall by about a
quarter. Combining this with the estimates from Feyrer (2009) implies that leaving the EU (and
joining EFTA) will reduce UK income by between 6.3% and 9.5%. These estimates are much
higher than the costs obtained from the static analysis, implying that dynamic effects from trade
are important.
The structure of the paper is as follows. We lay out the conceptual framework which captures
the welfare effect of Brexit in section 2. Then we present the data and our counterfactual analysis
1More technically, an expansion in the set of feasible allocations leads to Pareto superior outcomes.
98
in Section 3. In Section 4 we do various robustness check of our empirical results. Section 5 uses
alternative non-structural approaches to calculate welfare losses. We offer concluding comments in
Section 6.
2 Conceptual Framework
Formally, to quantify the trade-related welfare effects of Brexit we rely on structural estimation
based on a quantitative trade models whose calibration and simulation have been increasingly used
to investigate ex ante the implications of trade policies in counterfactual scenarios for which data
are necessarily unavailable.2
In particular, we build on (Arkolakis, Costinot, and Rodriguez-Clare, 2012) and (Costinot and
Rodriguez-Clare, 2013), who show that some of the most popular models used by trade economists
fall in a specific class share the same predicted ‘gains from trade’ (defined as welfare with trade relative to welfare with autarky), conditional on the changes in two aggregate statistics: the observed
share of domestic expenditure and an estimate of the trade elasticity.3
We use some simple relationships from this class of models to calculate what happens to income
(and therefore consumption and welfare) when trade costs change. Essentially, we use information
we know on current trade patterns and feed in different counterfactual scenarios about changes in
trade costs after Brexit. Taking the estimates of the trade elasticity from the literature we can then
figures out how trade patterns and income will change, depending on the degree to which trade
costs rise.
These models have four primitive assumptions in common: (a) Dixit-Stiglitz preferences; (b)
one factor of production; (c) linear cost functions; (d) perfect or monopolistic competition. They
also share three common macro-level restrictions: (A) trade is balanced; (B) aggregate profits are a
constant share of aggregate revenues; (C) the import demand system exhibits constant elasticity of
substitution (CES). As this set of assumptions is extremely restrictive, one would be forgiven for –
thinking they have limited practical relevance. What makes, instead, those restrictive assumptions
relevant is that some of the most popular trade models do satisfy them, from the workhorse CGE
model by (Armington, 1969) to the hallmark ‘new trade theory’ model by (Krugman, 1980), to the
Ricardian model by (Eaton and Kortum, 2002) and several variations of the heterogeneous model
2This section is based on (Ottaviano, 2014).
3See (Head and Mayer, 2014) as well as (Simonovska and Waugh, 2014) for recent discussions of methodological
issues related to the estimation of the trade elasticity.
99
by (Melitz, 2003).
The idea of using mathematical or statistical models to simulate the effects of counterfactual
scenarios has a long tradition (Baldwin and Venables, 1995). In particular, ‘Computable general
equilibrium’ (CGE) models remain a cornerstone of trade policy evaluation (Piermartini and Teh,
2005), having also contributed to the design of advanced softwares for their numerical solution such
as GAMS or GEMPACK. To this tradition the class of models we rely on contribute a tighter
connection between theory and data thanks to more appealing micro-theoretical foundations and
careful estimation of the structural parameters necessary for counterfactual analysis (Costinot and
Rodriguez-Clare, 2013).
We first explain the basic logic of this approach to calibration and simulation through a simplified model. We then describe the elements of the model we actually use.
2.1 The Armington Model
Following (Costinot and Rodriguez-Clare, 2013), the basic logic of our approach can be usefully
illustrated through a simple Armington model. The economy consists of n countries, indexed
i = 1; :::; n, with each country supplying its own distinct good. There are thus n goods, also
indexed i = 1; :::; n, with country i being the only supplier of good i in fixed quantity Qi, which
corresponds to the country’s endowment of the good.
Preferences in country j are captured by a representative household with Dixit-Stiglitz utility
function:
Cj = “Xi=1 n Cij ij σ–σ 1#σ–σ 1 (1)
where Cij is country j’s consumption of the good supplied by country i, ij > 0 is an inverse measure
of the appeal of this good for country j, and σ > 1 is the constant elasticity of substitution (CES)
between goods supplied by different countries. According to (1), utility can be interpreted as the
level of consumption of an aggregate composite (‘quantity index’) of the various goods whose ‘price
index’ is

( ijPij)1–σ#
1
1–σ
where Pij is the price of good i in country j. Denoting aggregate expenditure by Ej, the price and
quantity indices satisfy PjCj = Pn i=1 PijCij = Ej, which is the representative household’s budget
100
constraint. Utility (1) can then be equivalently rewritten as
Cj = Ej
Pj
(3)
which identifies real expenditure as a measure of country j’s welfare.
External trade between countries is subject to trade costs, consisting of frictional and tariff
barriers. Frictions are of the iceberg type: country i has to ship τij ≥ 1 units of its good for one
unit to reach country j. Tariff barriers are of the ad-valorem type with tij ≥ 0 denoting the tariff
imposed by country j on imports from country i. There are, instead, no trade costs for internal
trade: τjj = τjj 0 = 1 and tjj = t0 jj = 0.
Markets are perfectly competitive and perfect arbitrage implies that the price of a good at destination equals its price at the origin once trade costs are taken into account: Pij = (1 + tij) τijPii.
This in turn implies that a country’s income equals the country’s good endowment times its domestic price: Yi = PiiQi. Hence, the price at destination satisfies
Pij = φijYi
Qi (4)
where φij ≡ (1 + tij)τij denotes the trade costs from country i to country j.
Given (1), utility maximization under the representative household’s budget constraint determines the value of country j’s imports from country i inclusive of the associated tariff revenue
Xij =  ij PPj ij 1–σ Ej (5)
with Ej = Pn i=1 Xij. By (2) and (5), the share of expenditure of country j on imports from country
i evaluates to
λij = Xij
Ej
=  ij PPj ij 1–σ = Pn i(=1 φij(Yφiij)Y–i”)(–Q” i(=Qiij =)“ij)” (6)
where ” ≡ @ ln (Xij=Xjj) =@ ln τij = σ – 1 denotes the ‘trade elasticity’: the elasticity of imports
relative to domestic demand Xij=Xjj with respect to bilateral trade costs φij holding income levels
constant. Given (6), equation (5) can be then restated as a standard ‘gravity equation’
Xij = λijEj = Pn i(=1 φij(Yφiij)Y–i”)(–Q” i(=Qiij =)“ij)” Ej (7)
which expresses the bilateral trade flow from i to j as a function of characteristics of the country
101
of origin (Yi and Qi), characteristics of the country of destination (Ej), and bilateral obstacles (φij
and ij).
In equilibrium expenditure equals income plus tariff revenue
Ej = Yj + Tj (8)
with
Tj =
nX i
=1
tij
1 + tij
Xij (9)
and
Yi =
nX j
=1
1
1 + tij
Xij (10)
where Xij=(1 + tij) is the tax base. By (6) the share of tariff revenue in country j’s expenditure
can be expressed as
πj =
Tj
Ej
=
nX i
=1
tij
1 + tij
λij (11)
which allows one to use (8) to state country j’s total expenditure as a function of its income
Ej =
Yj
1 – πj
(12)
Plugged together with (7) into (10), (12) implies that good i’s market clears as long as
Yi =
nX j
=1
1
1 + tij
(φijYi)–” (Qi= ij)”
Pn i=1 (φijYi)–” (Qi= ij)”
Yj
1 – πj
(13)
holds. After using (11) and (6) to substitute πj with an expression in which income levels are the
only endogenous variables, for i = 1; :::; n (13) generates a system of n equations in n unknowns
that can be solved for the equilibrium income levels Y = fYig. However, as by Walras’ Law, one
of those equations is redundant, income levels can be determined only up to a constant pinned
down by the choice of the numeraire good. Having determined the equilibrium income levels,
the corresponding bilateral prices and price indices P = fPijg can be recovered from (4) and
(2) respectively. With the price information at hand, trade flows X = fXijg and expenditures
E = fEig can then be obtained from (5) and Ej = Pn i=1 Xij. This provides also information
required to compute expenditure shares λ = fλijg from (6) and tax revenue shares π = fπig from
(11). Finally, knowing prices and expenditures, welfare C = fCig can be measured from (3). This
102
concludes the description of the model and its equilibrium solution.
2.2 Welfare Effects of Changes in Trade Costs
How do changes in trade costs affect national welfare? To answer this question one has to assess
what happens to C when trade costs change from actual levels φ = fφijg to counterfactual levels
φ0 = nφ0 ijo. The main insights of Arkolakis, Costinot and Rodriguez-Clare (2012) is that changes
in the real expenditure of a country j can be readily computed using only few statistics: the trade
elasticity (“) and the changes in the country’s shares of expenditure across goods (from λ = fλijg
to λ0 = nλ0 ijo).
To see this, one needs first to derive three preliminary results on the effects of an infinitesimal
change in trade costs. First, given (2), partially differentiating Pj with respect to Pij yields
@Pj
@Pij = “Xi=1 n ( ijPij)1–σ#
σ
1–σ
( ij)1–σ (Pij)–σ =  ij PPj ij 1–σ PPij j
which, by (5), can be rewritten as
@Pj
@Pij =
Xij
Ej
Pj
Pij
implying the total differential
d ln Pj =
nX i
=0
λijd ln Pij (14)
This change in country j’s price index can be further broken down into changes of domestic and
import prices as
d ln Pj = λjjd ln Pjj + (1 – λjj) d ln PjM (15)
where
P M
j =
24
X i6=j
( ijPij)1–σ3 5
1
1–σ
is the component of Pj associated with imports, and
d ln P M
j =
1
1 – λjj
X i6=j
λijd ln Pij
103
is its variation. Second, (6) and (5) imply

1–σ =  jj ij 1–σ PPjjj M !1–σ
which can be totally differentiated to obtain
d ln P M
j = d ln Pjj + 1 –1 σ [d ln (1 – λjj) – d ln λjj] (16)
Third, the fact that expenditure shares sum up to one requires
λjj + (1 – λjj) = 1
the total differentiation of which leads to
(1 – λjj) d ln(1 – λjj) = –λjjd ln λjj (17)
Then, plugging (16) and (17) into (15) gives
d ln Pj = d ln Pjj – 1
1 – σ
d ln λjj (18)
so that the change in country j’s real expenditure Cj = Ej=Pj can be written as
d ln Cj = d ln Ej – d ln Pj = d ln Ej – d ln Pjj – 1
1 – σ
d ln λjj (19)
This expression can be further simplified recalling that there are no internal trade costs (τjj =
τ0
jj = 1 and tjj = t0 jj = 0) and trade must balance (Yj = (1 – πj)Ej). Under these conditions, (4)
implies PjjQj = Yj = (1 – πj)Ej and thus d ln Ej – d ln Pjj = –d ln(1 – πj) since Qj is a fixed
endowment. Given ” = σ – 1, (19) finally becomes
d ln Cj = –d ln(1 – πj) – 1
”
d ln λjj (20)
which shows that the welfare change d ln Cj is driven by the changes in the expenditure share of
tariff revenue πj and in the expenditure share on the domestic good λjj.
Expression (20) holds only for infinitesimal changes in trade costs, which tend to be of little
104
practical relevance. Nevertheless, it can be readily integrated to characterize the welfare effects of
discrete changes. This yields
bCj
=
1 – πj
1 – π0
j
λbjj–
1”
(21)
where the share of tariff revenues in the actual and counterfactual equilibria are given by
πj =
nX i
=1
tij

j0 nX t0
Hence, the welfare consequences of any arbitrary change in trade costs can indeed be computed
based only on few sufficient statistics: the trade elasticity and the change in the shares of expenditure across goods.
However, knowing that only few sufficient statistics are needed to compute the welfare effects of
trade integration would be of little use unless we had a consistent way of identifying the values of
those statistics in the counterfactual scenario. This is clearly not much of a problem for the trade
elasticity “, which, given utility (1), is constant by assumption. It may look more of a problem for
the counterfactual expenditure shares λ0 = nλ0 ijo. Luckily the structure of the model lends a hand.
Consider (6). As ij is constant, taking log changes gives
d ln λij = d ln (Pij)1–σ – d ln (Pj)1–σ
which, by (14), can be rewritten as
d ln λij = d ln (Pij)1–σ –
nX i
=0
λijd ln (Pij)1–σ (22)
As Qi is also constant, (4) implies
d ln (Pij)1–σ = d ln (φijYi)1–σ
which allows one to restate (22) as
d ln λij = d ln (φijYi)1–σ –
nX i
=0
λijd ln (φijYi)1–σ
105
for infinitesimal changes, or, by integration, as
λbij =
φbijYbi–”
Pn l=0 λlj φbijYbl–”
(23)
for discrete changes given ” = σ – 1.
In the counterfactual equilibrium, (6), (12) and (10) further imply
Y 0
j =
nX i
=1
1
1 + t0
ij
λ0
ij
Y 0
i
1 – π0
i
which can be rewritten as

i
=1
0
i
so that using (23) to substitute for λbij yields
YbjYj =
nX i
=1
1
1 + t0
ij
λij φbijYbi–”
Pn l=0 λlj φbljYbl–”
YbiYi
1 – π0
i
(24)
The share of tariff revenues in the counterfactual equilibrium is itself given by
πi0 =
nX i
=1
t0
ij
1 + t0
ij
λ0
ij =
nX i
=1
t0
ij
1 + t0
ij
λbijλij
which, by (23), becomes
πi0 =
nX i
=1
t0
ij
1 + t0
ij
λij φbijYbi–”
Pn l=0 λlj φbijYbl–”
(25)
After using (25) to substitute for πi0, (24) generates a system of n equations in n unknown
income changes that can be solved for the counterfactual Yb = nYbio (up to a normalization due the
choice of the numeraire good). As the system does not depend directly on the utility parameters
= f ijg and the endowments Q = fQig, changes in factor income levels Yb = nYbio can be
determined using only the initial expenditure shares λ = fλijg, the initial income levels Y = fYig,
and the trade elasticity “. Once the changes in income Yb have been solved for, the changes in
expenditure shares λb = nλbijo and the counterfactual tax revenues π0 = fπi0g can be obtained
106
from (23) and (25) respectively. Plugging them into (21) finally determines the welfare change Cbj
in the counterfactual scenario. Hence, the welfare effects of trade cost changes can be evaluated
estimating only the trade elasticity and not all the structural parameters of the model.
2.3 Intuition
Although the structure can appear complex at first sight, it is in fact very simple. Consider (24) as
the central relationship we can exploit to figure out the implications of Brexit. For each country we
want to measure the income changes Yb as trade costs rise after Brexit. We have different scenarios
(optimistic or pessimistic) associated with different trade costs changes φbij. We also have data
on the initial income Y and expenditure shares λ of each country, and take an estimate of the
trade elasticity ” from the literature. So basically we find the pattern of income changes that are
consistent with the new set of bilateral trade costs given the initial level of trade and how sensitive
these patterns are to price changes.
Think of this from a single country’s perspective. When trade costs rise, revenues from exports
fall as other countries buy less exports. To maintain trade balance, imports will also have to fall.
Both of these will decrease income (and this will have knock-on effects to other countries even if
trade costs have not changed for these countries). In equilibrium trade must balance so all of the
trade and income changes must be consistent with each other for every country. This is what (24)
describes.
2.4 Brexit and National Welfare
In the case of Brexit, we want to quantify not only the instantaneous welfare effects as done so far
but also its cumulative welfare effects in the future. This forward-looking perspective introduces
two additional layers of complexity. First of all, we need to evaluate the present value of future
utility flows. To do so, we assume that our representative household in country j has an infinite
life horizon so that its intertemporal utility can be expressed as
Uj;0 =
1X t
=0
βt ln Cj;t (26)
where real consumption Cj;t is defined by (1) after making time dependence explicit and t is a
time index starting from the current period t = 0.4 Analogously, in a counterfactual scenario
4We follow (Caliendo, Dvorkin, and Parro, 2015) in adopting log-preferences. These imply constant unit elasticity
of intertemporal substitution. Alternative assumptions on this elasticity are incompatible with the exact hat algebra”
107
intertemporal utility can be expressed as
U0
j;0 =
1X t
=0
βt ln Cj;t 0 :
The second layer of complexity comes from the fact that the future welfare effects of the UK
leaving the EU should be calculated relative not only to today’s welfare but also to the evolution
of future welfare if the UK stays in the EU. This implies that we have to compare the present value
of future utility between two counterfactuals: if the UK stays in the EU
UIn
j;0 =
1X t
=0
βt ln Cj;t In
and if the UK leaves
UOut
j;0 =
1X t
=0
βt ln Cj;t Out:
Following (Sampson, 2016), we can then measure the welfare effects of Brexit by using the
consumption adjustment that makes the representative household indifferent between staying or
leaving the EU over its entire life span. This is measured by the value of the parameter δt such
that
UIn
j;0(δj) =
1X t
=0
βt ln δj + ln Cj;t In
=
1X t
=0
βt ln Cj;t Out = Uj; Out 0
Solving this equation for ln δj gives the Brexit consumption equivalent change for country j
ln δBrexit
j = (1 – β)
1X t
=0
βt ln Cj;t Out – ln Cj;t In
(27)
Accordingly, the welfare effects of Brexit can be quantified by evaluating (27). Note that, after
defining
ln δIn
j = (1 – β)
1X t
=0
βt ln Cbj;t In
and
ln δOut
j = (1 – β)
1X t
=0
βt ln Cbj;t Out;
we apply throughout in the wake of (Costinot and Rodriguez-Clare, 2013).
108
expression (27) can be equivalently rewritten as
ln δBrexit
j = ln δjOut – ln δjIn = (1 – β)
1X t
=0
βt ln Cbj;t Out – ln Cbj;t In (28)
where CbIn
j;t = Cj;t In=Cj;t and Cbj;t Out = Cj;t Out=Cj;t are the changes in real consumption in period t
relative to the status quo in period 0 if the UK stays and if the the UK leaves the EU respectively.
The idea is that, if the UK stays, it will enjoy any further fall in trade costs among EU members
relative to the status quo. The corresponding welfare effects are captured by the consumption
equivalent change δjIn. On the other hand, if the UK leaves the EU, it will miss not only any
further fall in trade costs among EU members but might also face higher trade costs when trading
with EU members relative to the status quo. The corresponding welfare effects are captured by

2.5 Our Model
The Armington model has helped us to explain the basic logic of our approach, and in detail the
mechanics of its calibration and simulation procedure. It is, however, too stylized for our purposes.
Fortunately, (Costinot and Rodriguez-Clare, 2013) have shown that the gravity equation (7), which
is the basis for counterfactual analysis in the Armington model, holds under various assumptions
about technology and market structure. In what follows, we make the realistic choice of allowing
for multiple sectors and tradable intermediate inputs as well as the conservative choice of focusing
on the case of perfect competition as this has been shown to provide a lower bound to the welfare
effects of changes in trade costs.
Specifically, as in (Costinot and Rodriguez-Clare, 2013), we extend the model to multiple sectors,
indexed s = 1; :::; N, by assuming that the good consumed by the representative household in (26)
is a Cobb-Douglas basket of the goods supplied by the different sectors
Cj = YS s=1 Cj;s βj;s
where, leaving time dependence t implicit for ease of notation, Cj;s is real consumption of the good
supplied by sector s and βj;s 2 (0; 1) is its share of household expenditures with PS s=0 βj;s = 1.
By analogy with the single sector expression (21), the change in real consumption generated by a
109
counterfactual change in trade costs equals
bCj
=
1 – πj
1 – π0
j
YS s=1 λbjj;s–
βj;s
“s (29)
where the πj and πj0 are again the shares of tariff revenue in country j’s expenditure in the current
and counterfactual scenarios respectively. Clearly, (29) boils down to (21) when we only have one
sector. To add also intermediates, we assume that each sector output is used not only in final
consumption but also, together with primary factors, as input for its own and any other sector’s
production. If we use αj;sk 2 [0; 1] to denote the share of sector k’s output in sector s’s expenditure
on intermediate inputs, we get
bCj
=
1 – πj
1 – π0
j
YS s;k=1 λbjj;k–
βj;saej;sk
“k (30)
where eaj;sk is the elasticity of the price index in sector s with respect to changes in the price of
sector k. These price elasticities are given by the elements of the S × S Leontief inverse matrix
(Id – Aj)–1 where Aj is the matrix with typical element αj;sk. Expression (30) is what we use to
evaluate CbIn
j;t and Cbj;t Out for the quantification of the the welfare effects of Brexit through (28).
3 Empirical Analysis
In this section, we specify the time path of changes in the iceberg trade costs and tariffs if the UK
remains a member of the EU and if it does not under different scenarios. Then we quantify the
effects of such changes in tariff and/or trade costs on welfare and trade using the structural model
specified in previous section.
3.1 Data
We use the latest World Input-Output Database (WIOD) for year 2011.5. This database aggregates
the world into 40 countries and covers 35 sectors which we further aggregate into 35 regions and
31 sectors as in Costinot and Rodriguez-Clare (2013) We also collect information on the applied
most favoured nation (MFN) tariff by the EU from the World Trade Organization (WTO) website6,
which provides information on tariffs at the product level (HS classification) for all tradable goods.
5The data could be found at http://www.wiod.org/new_site/home.htm. For more details on how this database
is constructed, see Dietzenbacher, Los, Stehrer, Timmer, and de Vries (2013).
6http://tariffdata.wto.org/
110
We also use the United Nations (UN) Comtrade bilateral database at the product level. These two
datasets permit us to calculate an average MFN tariff at the WIOD sector level for UK imports
(exports), from (to) the EU by using import (export) value at the product level as weights. 7 The
resulting average MFN tariffs for imports and exports from/to the EU can be seen in Table A.3,
which summarizes the UK trade and MFN tariff information at the sector level. The table splits
the sectors between ’Goods’ and ’Services’.
The most intensively traded good in the UK/EU bilateral relationship is ’Transport Equipment’,
that includes auto-mobiles, amounting to 95.7 billion of US dollars in 2011. This sector also
possesses one of the highest average tariffs: 8.09% for imports from the EU and 7.22% for exports
to the EU. Note that most part of this trade is composed by imports (60.4 billion, or 63%). On the
other hand, the trade champion among services, the sector ’Renting of Machinery and Equip. and
Other Business Services’, is more intensively exported (USD 53 billions) than imported (USD 28
billion) by the UK. Financial services also are responsible for a significant trade share. Together,
the two former sectors are responsible for more than two thirds of the flows of services between the
UK and the EU. In general, we can see that the UK holds a deficit among goods and a surplus
among services, with reasonable variability within the two groups.
3.2 Counterfactuals
In this section we present counterfactual exercises associated with the UK leaving the EU. We
aim to quantify changes in welfare (real UK consumption) coming from three distinct sources: i)
immediate changes in goods tariffs, ii) immediate changes in non-tariff barriers, and iii) exclusion
from future market integration in the EU.
We consider two different scenarios. In the pessimistic case we assume that the UK is no longer
part of the single market and will trade with the EU under the regulations of the WTO. The UK
will apply the MFN tariffs seen in column (4) of Table A.3 on goods imported from the EU, while
the EU will apply the tariffs observed in column (7) on goods originating from the UK. This seems
reasonable just after withdrawal, but the hope is that the UK will eventually be able to negotiate a
better deal such as enjoyed by Norway and Iceland (in the European Economic Area) or Switzerland
(which has a series of bilateral deals). Hence, in our optimistic scenario we consider that tariffs on
goods continue to be zero between the UK and the EU.
Another important source of trade costs around the world is due to non-tariff barriers. Non-
7We aggregate HS 6-digit industries into 2-digit WIOD industries using a concordance between HS and ISIC Rev3.
111
tariff barriers are related to costs of shipment, differences in product regulations, legal barriers,
whether countries share a border, a common currency or language, search and other transaction
costs for both goods and services, etc (Anderson and van Wincoop, 2004; Head and Mayer, 2013).
Many authors point out that such costs are higher than formal tariffs (Anderson and van Wincoop,
2004; Novy, 2013; LooiKee, Nicita, and Olarreaga, 2009). In fact, most part of the negotiations
regarding the Transatlantic Trade and Investment Partnership between the EU and the USA aim
to diminish non-tariff barriers.
To incorporate non-tariff barriers we use information provided by Berden, Francois, Tamminen,
Thelle, and Wymenga (2009, 2013). The authors calculate detailed tariff equivalents of non-tariff
barriers between the USA and the EU, using econometric techniques and business surveys. They
also calculate the fraction of these non-tariff barriers that is reducible for each sector, i.e. the fraction of the trade cost that could in principle be eliminated by policy action. We collect information
on sectors that can be easily matched to our classification shown in Table A.3. The sectors used,
their non-tariff costs (in tariff equivalent terms) and the share of the costs that can be reduced are
shown in Table A.1 in the Appendix.
As it is rather unlikely that the UK would face the same costs as the US in a case of withdrawal,
in our optimistic scenario we assume that the UK would face one quarter (1/4) of the reducible
cost faced by the USA, while in our pessimistic scenario we assume that they would face three
quarters (3/4). We calculate the weighted average of these cost shares, using total EU/UK trade
in each sector as weights and the subset of sectors shown in the Table A.1, which include several of
the relevant sectors in the EU/UK relationship. This calculation leads to an increase in non-tariff
costs of 2.01% and 6.04% in our optimistic and pessimistic scenarios, respectively. We then apply
such costs to all sectors in our economy.
We also consider that the intra-EU trade costs are falling over time (Ilzkovitz, Dierx, Kovacs,
and Sousa, 2007), and this rate is approximately 40% faster than in other OECD countries according
to M´ejean and Schwellnus (2009), which uses panel data on French firms to study price convergence
in different markets between 1995 and 2004.8 We consider the scenario that 10 years from now
non-tariff barriers inside the EU would keep falling faster and the UK would not benefit from this
evolution. In our pessimistic scenario we assume that intra-EU non-tariff costs continue to fall 40%
faster than in the rest of the world. This may not necessarily be true since the OECD does not
include countries like China, which has seen a rapid decrease in trade costs with other countries.
8They find that the rate of price convergence is -0.412 for OECD countries -0.593 for EU countries.
112
Hence, in our optimistic scenario we assume that intra-EU barriers fall only 20% faster than in the
rest of the world.
To calculate this last counterfactual we need a measure of price differences across the EU. We
use a rough measure from Eaton and Kortum (2002) of 49%9, meaning that if the UK imported
(exported) all goods from (to) other European countries prices would be 49% higher. Naturally,
part of this price difference may not be reducible. We assume that the reducible proportion is 55%,
which is the same share of non-tariff barriers that are actionable in the EU-USA trade case. To
be conservative, in our pessimistic case we further assume that three quarters of the potentially
reducible share will actually diminish throughout the years, while in the optimistic case we assume
that such share is only one half. And to be even more conservative, we assume that the faster
market integration will peter out in 10 years after Brexit. And then, again using the estimates of
reducible price gaps from M´ejean and Schwellnus (2009), we calculate the future falls of non-tariff
barriers within EU, which lead to a fall of 12.77% and 6.04% in our pessimistic and optimistic
scenarios respectively at year 10 after Brexit. 10
Armed with these numbers, we simulate the model by feeding in the sequence of shocks in trade
costs and tariffs for different scenarios of Brexit. The model then generates a sequences of changes
in real consumption. This allows us to compute the welfare change due to Brexit using equation
(28), assuming that the discount rate of future consumption is β = 0:96, which is a common value
in the macro literature.
Our results are shown in Table A.5. Panel A shows the result of the optimistic scenario. We
find that the welfare loss of the UK via the trade effect is 1.37%. How large are these numbers
when compared to the costs generated by the EU membership? HM Treasury (2013) estimates
that the net fiscal contribution of the UK to the EU is around 0.53% (or $ 8.6 billion) of the
UK GDP (2013). We assume that the UK would keep contributing 83% of the current per capita
contribution as Norway does in order to remain in the single market (House of Commons, 2013).
This leads to a fiscal saving of about 0.09%. Taking this benefit into account, the UK would still
lose a total of 1.28% in the case of an exit from the EU. We also calculate the implied loss per
household. In 2015 the UK had a population of about 65m with 27m households and a GDP of
$1.8 trillion. 1.28% of 1800/27 is $853 per household, which we round to $850 in the Table.
Panel B of Table A.5 shows the result of the pessimistic scenario. We see that the cost of a
9Table II, UK row average of the trade cost values.
10Please refer Appendix B for the details of the computation.
113
withdrawal doubles. The UK loses 2.92% via trade due to higher tariff, non-tariff trade barriers
and exclusion from future further integration of the EU. Discounting the fiscal benefits still implies
a total welfare loss of 2.61%. This is equivalent to $1,700 per household.
We have shown the welfare losses suffered by the UK. We also estimate the effect of Brexit
on welfare for other countries again using equation (28). The results are shown in Figure C.1.
Two groups of countries have relatively larger welfare losses. First, countries for which UK is an
important trade partner, such as Ireland, Netherlands, Belgium, Denmark, Sweden, and Germany.
Figure C.2 shows the average expenditure share of intermediates sourcing from the UK across
sectors. These countries source relatively more intermediates from the UK, especially Ireland which
sources about 12% of intermediate inputs from the UK. A second group of countries that lose out
are those that do not trade much with the UK but exhibit a negative cross-sectoral correlation
between the expenditure share on intermediates sourcing from the UK and the trade elasticity.
Figure C.3 shows this correlation across countries. Countries such as Hungary, Czech Republic,
and Slovakia tend to trade more with the UK in sectors with relatively low trade elasticity. In
other words, if trade costs rise with the UK, they cannot easily substitute towards goods from
other countries. Thus they will have a relatively larger welfare loss as the prices they pay will rise
even if they trade relatively less with the UK.
Finally, countries outside EU tend to gain from Brexit, such as Russia, Turkey and China,
although the numbers are very close to zero. This is because of a trade diversion effect due to the
fact that the UK partially switches from trading with the EU to trading with non-EU countries
(which in turn benefit from more trade with UK). This can be illustrated by Table A.6. As we can
see, total British trade falls less than trade with the EU after Brexit.
4 Extensions and Robustness Checks
To check the robustness of our results to alternative assumptions, we first simulate the model for
two different scenarios. The results is shown in Panel A of Table A.8. The first scenario that
we simulate is the “Swiss Alternative”. Switzerland is not in the EEA but has many bilateral
agreements with the EU, which give it some access to the single market. Like Norway, it has
to adopt all the regulations covering those parts of the single market it participates in and also
allows free movement of labour. It does, however, benefit from a lower fiscal transfer (about 40%
of the UK’s contribution on a per capita basis). On the other hand, it does not have free trade
114
in services with the EU, which would be a disadvantage for an economy like the UK, which has
a comparative advantage in services. We simulate the effects of Brexit using Switzerland as an
alternative optimistic scenario.
The result is very similar to the benchmark optimistic scenario – a loss of income of 1.28%.
Although the fiscal transfers are lower than for Norway (40% of 0.53% = 0.31% ) , these are more
than offset by higher costs of trade in services (a total welfare loss from lower trade of 1.6% vs.
1.37%).
Another scenario that we consider is what we call the Big Bang scenario. Under this scenario,
we assume that, if Brexit happens, the UK and the EU would impose MFN tariffs on each other
and the non-tariff trade barriers between the UK and the EU would rise to the reducible level
between the USA and the EU (+8.06%). Integration in the EU would continue to be 40% faster
than in the rest of world and 100% of the reducible price gaps could be reduced. However, the UK
would not benefit from such further integration. We assume that further integration would happen
in the year following Brexit, which implies that we are simulating the upper bound of welfare loss
for UK in our model. In this scenario we find that the UK welfare loss is about 3.5%.
Brexit campaigners have argued that the UK could neutralize the trade effect by unilaterally
liberalizing with all other countries. We check whether this is the case by removing all UK import
tariffs. We measure these import tariffs by constructing the sectoral MFN tariff as the weighted
average of HS 6-digit level UK imports from non-EU countries. The results are shown in Table A.2.
The overall weighted average UK MFN import tariff is around 3%. Feeding these tariffs into our
model for both the optimistic and pessimistic scenario, we find the effect of unilateral liberalization
is very limited as shown in Panel B of Table A.8. The welfare gain from removing the MFN import
tariff of the UK is just around 0.3%, far from neutralizing the adverse trade effect of Brexit. In the
optimistic case the income loss is 1% instead of 1.3% and in the pessimistic case the loss if 2.3%
instead of 2.6%. This is not surprising given that UK’s import tariffs are already very low.
In Panel C, we simulate the welfare loss of UK using alternative values of discount factor β
and fiscal benefits. So far, we have used a real interest rate of 4% which is standard in the macro
literature, but currently real interest rates are much lower than this, near zero in many cases. Using
a lower interest rate increases the costs of Brexit, because it gives larger weights to future losses
of income. For example, using a real interest rate of 1% leads to a welfare loss of 2.68% in the
optimistic case. Hence, given the current low interest rate, the results that we present in Table A.5
might actually understate the real loss.
115
Finally, in the second column of Panel C we show that varying the fiscal benefit from 0.31% to
0.53% (to account for different available estimates) makes little difference to our results.
4.1 Summary
In this section, we vary parameters within plausible ranges to test the robustness of our findings.
Although the exact magnitude of the welfare loss changes in each experiment, it is consistently
negative with a loss of income ranging between 1% and 4%. The qualitative finding that British
households will be poorer after Brexit is robust, the only question is exactly how much poorer they
will be.
5 Reduced Form Estimates
In the previous section we attempted to quantify the welfare effects of the UK leaving the EU using
a quantitative model of international trade. An alternative approach is to use existing empirical
estimates of the effects of EU membership to infer the impact of leaving the EU on UK income.
In particular, we can decompose the question into two parts. First, what effect will leaving the
EU have on the UK’s trade with the rest of the world? Second, what is the effect of changes in
trade levels on income? There exist substantial literatures addressing both the effect of joining an
economic integration agreement (EIA), such as the EU, on trade and the effect of trade on income.
Suppose that if the UK leaves the EU it will become a member of the European Free Trade
Association (EFTA). Does EU membership cause a country to trade more with other EU members
than EFTA membership? Baier, Bergstrand, Egger, and McLaughlin (2008) address exactly this
question using a gravity model of bilateral trade augmented with dummy variables for which EIAs
the exporter and importer belong to. In particular, they include dummy variables for both countries
being in the EU, both countries being in EFTA, one country being in the EU and the other in
EFTA and for both countries belonging to any other EIA. Importantly, they control for endogeneity
of selection into the formation of EIAs using country-pair fixed effects with panel data. They find
robust evidence that being a member of the EU leads a country to trade significantly more with
other members of the EU than if it were only a member of EFTA. Quantitatively, their estimates
imply that leaving the EU and joining EFTA would reduce the UK’s trade with EU members by
25%.11
11This figure is calculated using the estimates in Table 6, column 1. Both countries being in the EU increases trade
by e0:48 – 1 = 62%, while one country being in the EU and the other in EFTA increases trade by e0:19 – 1 = 21%.
116
To predict the change in the UK’s overall trade we also need to know how leaving the EU
would affect the UK’s trade with non-EU members. Baier, Bergstrand, Egger, and McLaughlin
(2008) estimates suggest that whether a country is a member of the EU or EFTA does not have a
significant effect on its trade with EFTA members. However, their estimates do not address how
EU membership affects trade with countries outside of both the EU and EFTA. Structural gravity
models such as that developed by Egger, Larch, Staub, and Winkelmann (2011) can be used to
infer the general equilibrium effects of EIAs on trade between all country-pairs, but we are not
aware of any work that applies the structural gravity methodology to estimate the effects of EU
membership. Instead, we will rely on reduced form gravity model estimates of the trade diversion
effects of EIAs. Studies of trade diversion typically find little evidence that joining an EIA leads
to a reduction in trade with countries outside of the EIA. For example, Magee (2008) fails to find
robust evidence of significant trade diversion effects from EIAs. Therefore, we will assume that
leaving the EU will not affect the UK’s trade with the rest of the world.
To quantify the effect of trade on income we will use the estimates of Feyrer (2009). Using data
on the air and sea distances between countries, Feyrer (2009) uses changes in the cost of shipping
goods via air relative to sea as an instrument for trade in a regression of income on trade. Since
the instrument is time varying, Feyrer (2009) is able to improve upon the cross-section estimates of
Frankel and Romer (1999) by using country fixed effects to control for time invariant unobservable
that affect income levels. Feyrer (2009) concludes that the elasticity of income to trade is probably
between one-half and three-quarters. In other words, a 10% increase in trade increases income by
5% to 7.5%. The estimation strategy of Feyrer (2009) implies that his estimates capture both the
direct effect of higher trade on income and also other indirect effects of increased proximity between
countries such as variation in FDI and knowledge diffusion. Thus, the estimates we obtain in this
section should be interpreted as including some of the non-trade channels through which leaving
the EU will affect UK income in addition to the direct effect of changes in the UK’s trade.
Combining these numbers we can obtain a reduced form estimate of the effect of leaving the EU
and joining EFTA on UK income. Since 50:4% of the UK’s trade is with the EU, a 25% fall in trade
with EU members will reduce the UK’s overall trade by 12:6%. Combining this with the estimate in
Feyrer (2009) that the elasticity of income to trade is between one-half and three-quarters implies
Therefore, if a country leaves the EU and joins EFTA trade with EU members declines by (e0:19 –e0:48)=e0:48 = 25%.
To avoid confusion when interpreting the coefficient estimates in Baier, Bergstrand, Egger, and McLaughlin (2008)
note that their EEA dummy variable is defined equal to one for a country pair when one country is in EFTA and
the other country is in the EU. Baier, Bergstrand, Egger, and McLaughlin (2008) do not estimate the effects of EEA
membership on trade, probably because the EEA was only established in 1994 and they use data from 1960-2000.
117
that leaving the EU and joining EFTA will reduce the UK’s income by between 6:3% and 9:5%.12
Interestingly, these calculations are similar to estimates of the historical benefits of EU membership
for the EU. For example, (Crafts, 2016) considers a range of papers that have sought to estimate
historically what the net benefit has been of EU membership. He concludes that the there was an
increase in UK GDP of around 8% to 10%.
The reduced form approach used in this section has two principal advantages over the structural
approach used earlier in the paper. First, it requires less detailed assumptions about what the
relationship between the UK and the EU would be following a UK exit. The structural estimates
required assumptions about both the future level of tariffs between the EU and the UK and the
extent to which the UK would share in future reductions in non-tariff barriers within the EU. By
contrast, the reduced form estimates are based on the simple and plausible assumption that if the
UK leaves the EU it will join EFTA. Second, while the quantitative trade model used above is
designed to capture only the static gains from trade, reduced form estimates of the effect of trade
on income should capture both static and dynamic effects.
The disadvantage of the reduced form approach is that it relies on the existence of unbiased
empirical estimates. While we have based our calculations on estimates obtained using best practice
empirical methodologies, sampling error and identification challenges inevitably mean that some
degree of uncertainty must be attached to the estimates. Overall, the calculations in this section
should be viewed as a robustness check on the plausibility of the predictions obtained from the
quantitative trade model. The reduced form estimates of the income effect of leaving the EU are
higher than those obtained from the quantitative trade model, but they reinforce the conclusion
that leaving the EU is likely to have a sizeable negative effect on UK welfare.
6 Conclusion
We have looked at different ways of estimating the change in UK living standards following a
decision to leave the EU. Using the Costinot and Rodriguez-Clare (2013) methodology, we generate
counterfactual scenarios and show that future losses in the UK due to this move can sum up to
1.28% of the GDP in real terms in our optimistic scenario, and to 2.61% in our pessimistic one.
12These estimates will understate the cost of leaving the EU if Baier, Bergstrand, Egger, and McLaughlin (2008)
underestimate the decline in trade from leaving the EU and joining EFTA. Using the estimates in Table 5, column
1 of Baier, Bergstrand, Egger, and McLaughlin (2008) implies the UK’s trade with EU members would decline by
(e0:19 – e0:65)=e0:65 = 37% which implies a decline in UK income of between 9:3% and 13:9%. We chose to use the
estimates in Table 6, column 1 to obtain a more conservative estimate of the costs of Brexit.
118
There are good reasons for thinking these under-estimate the real costs of Brexit as the evidence
looking at the historical impact of countries joining the EU has generated more trade and more
income than the static trade exercises we perform here would suggest. Using the reduced form
approach finds welfare losses of between 6:3% and 9:5%.
In any case, we should have in mind that these numbers are likely to be larger in reality, since
many other welfare improving channels associated with EU trade such as immigration, increases
in productivity, increases in R&D intensity, vertical production chains, to cite just a few, are not
considered in our analysis.
References
HM Treasury (2013): European Union Finances 2013: statement on the 2013 EU Budget and
measures to counter fraud and financial mismanagement,” Discussion paper.
Albornoz, F., H. F. Calvo Pardo, G. Corcos, and E. Ornelas (2012): Sequential exporting,” Journal of International Economics, 88(1), 17{31.
Anderson, J. E., and E. van Wincoop (2004): Trade Costs,” Journal of Economic Literature,
42(3), 691{751.
Arkolakis, C., A. Costinot, and A. Rodriguez-Clare (2012): New Trade Models, Same
Old Gains?,” American Economic Review, 102(1), 94{130.
Armington, P. S. (1969): A Theory of Demand for Products Distinguished by Place of Production (Une th´eorie de la demande de produits diff´erenci´es d’apr`es leur origine)(Una teor´ıa de la demanda de productos distingui´endolos seg´un el lugar de producci´on),” Staff Papers-International
Monetary Fund, pp. 159{178.
Baier, S. L., J. H. Bergstrand, P. Egger, and P. A. McLaughlin (2008): Do Economic
Integration Agreements Actually Work? Issues in Understanding the Causes and Consequences
of the Growth of Regionalism,” The World Economy, 31(4), 461{497.
Baldwin, R. E., and A. J. Venables (1995): Regional economic integration,” Handbook of
international economics, 3(3), 1597{644.
Berden, K., J. Francois, S. Tamminen, M. Thelle, and P. Wymenga (2009): Non-Tariff
Measures in EU-US Trade and Investment: An Economic Analysis,” Final report, Ecorys.
119
(2013): Non-Tariff Measures in EU-US Trade and Investment: An Economic Analysis,”
IIDE Discussion Papers 20090806, Institue for International and Development Economics.
Bloom, N., M. Draca, and J. Van Reenen (2015): Trade induced technical change? The
impact of Chinese imports on Innovation, IT and Productivity,” Review of Economic Studies,
83(4), 87{117.
Caliendo, L., M. Dvorkin, and F. Parro (2015): Trade and labor market dynamics,” NBER
Working Paper, 21149.
Caliendo, L., and F. Parro (2014): Estimates of the Trade and Welfare Effects of NAFTA,”
The Review of Economic Studies, p. rdu035.
Costinot, A., and A. Rodriguez-Clare (2013): Trade Theory with Numbers: Quantifying
the Consequences of Globalization,” CEPR Discussion Papers 9398, C.E.P.R. Discussion Papers.
Crafts, N. (2016): The Growth effects of EU Membership for the UK: A Review of the Evidence,” mimeo, University of Warwick.
di Giovanni, J., A. Levchenko, and F. Ortega (2015): A Global View of Cross-Border
Migration,” CReAM Discussion Paper Series 1218, Centre for Research and Analysis of Migration
(CReAM), Department of Economics, University College London.
Dietzenbacher, E., B. Los, R. Stehrer, M. Timmer, and G. de Vries (2013): The
Construction Of World Input{Output Tables In The Wiod Project,” Economic Systems Research,
25(1), 71{98.
Eaton, J., and S. Kortum (2002): Technology, Geography, and Trade,” Econometrica, 70(5),
1741{1779.
Egger, P., M. Larch, K. E. Staub, and R. Winkelmann (2011): The Trade Effects of
Endogenous Preferential Trade Agreements,” American Economic Journal: Economic Policy,
3(3), 113{43.
Feyrer, J. (2009): Trade and Income { Exploiting Time Series in Geography,” NBER Working
Papers 14910, National Bureau of Economic Research, Inc.
Frankel, J. A., and D. H. Romer (1999): Does Trade Cause Growth?,” American Economic
Review, 89(3), 379{399.
120
Harari, D., and G. Thompson (2013): The economic impact of EU membership on the UK,”
Commons library standard note, Commons Library.
Head, K., and T. Mayer (2013): Gravity Equations: Workhorse, Toolkit, and Cookbook,”
CEPR Discussion Papers 9322, C.E.P.R. Discussion Papers.
(2014): Gravity Equations: Workhorse, Toolkit, and Cookbook,” Handbook of International Economics, 4.
House of Commons (2013): Leaving the EU, no. 13/42.
Ilzkovitz, F., A. Dierx, V. Kovacs, and N. Sousa (2007): Steps towards a deeper economic
integration: the internal market in the 21st century,” European Economy – Economic Papers
271, Directorate General Economic and Monetary Affairs (DG ECFIN), European Commission.
Keller, W. (1999): How Trade Patterns and Technology Flows Affect Productivity Growth,”
Working Paper 6990, National Bureau of Economic Research.
(2002): Trade and the Transmission of Technology,” Journal of Economic Growth, 7(1),
5{24.
Krugman, P. (1980): Scale Economies, Product Differentiation, and the Pattern of Trade,”
American Economic Review, 70(5), 950{59.
LooiKee, H., A. Nicita, and M. Olarreaga (2009): Estimating Trade Restrictiveness Indices,” Economic Journal, 119(534), 172{199.
Magee, C. S. (2008): New measures of trade creation and trade diversion,” Journal of International Economics, 75(2), 349{362.
Mejean, I., ´ and C. Schwellnus (2009): Price Convergence in the European Union: Within
Firms or Composition of Firms?,” Working Papers hal-00354190, HAL.
Melitz, M. J. (2003): The Impact of Trade on Intra-Industry Reallocations and Aggregate
Industry Productivity,” Econometrica, 71(6), 1695{1725.
Melitz, M. J., and S. J. Redding (2014): Missing Gains from Trade?,” CEP Discussion Papers
dp1254, Centre for Economic Performance, LSE.
121
Morris, N. (2013): Former Chancellor Nigel Lawson calls for UK to leave European Union,”
Discussion paper, The Independent.
Novy, D. (2013): Gravity Redux: Measuring International Trade Costs With Panel Data,”
Economic Inquiry, 51(1), 101{121.
Ottaviano, G. (2014): European Integration and the Gains from Trade,” Routledge. Handbook
of the Economics of European Integration.
Piermartini, R., and R. Teh (2005): Demystifying modelling methods for trade policy, no. 10.
WTO Discussion Paper.
Sampson, T. (2016): Dynamic Selection: An Idea Flows Theory of Entry, Trade and Growth,”
Quarterly Journal of Economics., 131(1), 315{80.
Simonovska, I., and M. E. Waugh (2014): Trade models, trade elasticities, and the gains from
trade,” Discussion paper, National Bureau of Economic Research.
Springford, J., and S. Tilford (2014): The Great British trade-off,” Discussion paper, Centre
for European Reform.
Wacziarg, R. (1998): Measuring the dynamic gains from trade,” Policy Research Working Paper
Series 2001, The World Bank.
122
Appendix
A Tables
Table A.1: Sector and Non-tariff Barriers (NTB) used in the Counterfactuals
NTB Cost EU+/USA Reducible share Weight
Sector (tariff equivalent) of NTB (total trade UK/EU)
Transport Equipment 22.1% 0.53 95723
Chemicals and Chemical Products 23.9% 0.63 74797
Post and Telecommunications 11.7% 0.70 8733
Electrical and Optical Equipment 6.5% 0.41 61506
Financial Intermediation 11.3% 0.49 50145
Food, Beverages and Tobacco 5.8% 0.53 56463
Construction 4.6% 0.38 3760
Renting of Machinery & Equip. and Other Business Activities 14.9% 0.51 72628
Services Nec (*) 4.4% 0.37 13561
Basic Metals and Fabricated Metal 11.9% 0.62 44769
Textiles and Textile Products; Leather, Leather and Footwear 19.2% 0.50 20178
Wood and Products of Wood and Cork 11.3% 0.60 3413
Overall Weighted Average 14.7% 0.55 {
Source: WIOD and authors’ compilation of a subset of the sectors presented in Tables 3.3 and 4.2 of Berden,
Francois, Tamminen, Thelle, and Wymenga (2009).
Notes: The Table provides non-tariff costs (in tariff equivalent terms) of trade flows from the USA to the EU+
(column 1). It also provides the share of costs that are potentially reducible (column 2). In our counterfactuals
we assume either (i) that after Brexit the UK faces 1/4 of the reducible costs of the USA (optimistic scenario) or
(ii) that after the exit the UK faces 3/4 of the reducible costs seen by the USA (pessimistic scenario). We then
use total EU trade as weights (column 3) to compute a weighted average of these costs and apply to all sectors in
all our counterfactuals. EU is defined as EU 28 minus the UK. EU+ includes the UK. Total trade in column (3)
is the sum of all imports from the rest of the EU to the UK plus all exports from the UK to the EU (in millions
of US dollars). The overall weighted averages in the final row use column (3) numbers as weights.
(*) Includes ’Repair of Household Goods’
123
Table A.2: UK MFN tariff with Non-EU Countries
Sectors Import Tariff Export Tariff
Agriculture, Hunting, Forestry and Fishing 1.07 4.02
Mining and Quarrying 0.00 0.00
Food, Beverages and Tobacco 6.19 2.08
Textiles and Textile Products; Leather, Leather and Footwear 10.70 8.73
Wood and Products of Wood and Cork 2.74 3.16
Pulp, Paper, Paper , Printing and Publishing 0.07 0.06
Coke, Refined Petroleum and Nuclear Fuel 2.51 3.36
Chemicals and Chemical Products 2.46 1.89
Rubber and Plastics 5.25 5.28
Other Non-Metallic Mineral 4.79 3.49
Basic Metals and Fabricated Metal 1.47 1.00
Machinery, Nec 2.36 2.00
Electrical and Optical Equipment 1.84 1.70
Transport Equipment 5.43 6.26
Manufacturing, Nec; Recycling 1.45 1.76
Overall Weighted Average 3.09 2.60
Source: UN Comtrade comtrade.un.org/ and WTO http://tariffdata.wto.org/.
Note: Tariff used in the case of UK unilaterally liberalization. Actual applied MFN tariff for HS6 industries
are aggregated to WIOD sectors using the trade between UK and non-EU countries as weights. In other words
we use the total imports to the UK from non-EU countries at the HS6 level to weight the import tariffs and the
total exports from the UK to non-EU countries at the HS6 level to weight the export tariffs.
124
Table A.3: UK Trade Statistics in 2011