Artikel als PDF herunterladen

Would the United Kingdom really apply high EU external tariffs in the case of a hard Brexit and carry out extensive physical checks at their border? Recent developments in London call this into question. A hard-but-smart strategy would fundamentally change Britain’s position vis-à-vis the EU and potentially result in a constructive alternative.

Most of the studies on the economic effects of Brexit assume that the UK will impose tariffs and controls. This leads to the conclusion that a hard Brexit is much more expensive for the UK than for the EU. But such a strategy would hardly be rational for London and perhaps technically unfeasible. With a ‘hard-but-smart’ strategy, on the other hand, the UK would unilaterally renounce the introduction of any new barriers, and, as required by WTO law, even reduce tariffs with third countries. This fundamentally changes the negotiating situation. With hard-but-smart, the percentage welfare loss of the UK and that of the continent is no longer substantially different. Such a Brexit would not put the UK in a worse position than the EU itself. The analysis suggests that the EU Commission should not stubbornly insist on the withdrawal agreement, which is unacceptable for London, but should work out constructive alternatives.

Previous mainstream consensus: With a hard Brexit, the UK loses more than the EU

A central premise of the EU’s Brexit negotiation strategy is that the failure of a comprehensive contractual reorganisation would damage economic relations between the UK and the EU, especially on the British side. The reason for this is obvious: if trade between the island and the continent is burdened with new tariffs and administrative barriers, the UK suffers much more because its economy is more dependant on trade with the rest of Europe than vice versa.

The EU27 countries account for almost 45% of British goods exports and 53% of goods imports; conversely the UK accounts for only seven percent and four percent of EU27 countries’ goods exports and imports, respectively. In addition, the UK receives 15% of its imports, from countries with which the EU has free trade agreements; 19% of British exports go to these countries. In other words, if the UK were to impose duties vis-à-vis all of the countries with which it has no free trade agreement in the case of Brexit, 68% of imports would be subject to duties from one day to the next and new bureaucratic burdens would be imposed on these trade volumes. If, on the other hand, the EU imposes tariffs on its imports from the UK, this will only affect four percent of its imports.1 So there would hardly be any effect on the shopping basket of Europeans.2

Almost automatically, all standard trade models predict that the macroeconomic losses due to the reintroduction of customs duties and administrative barriers are much greater for the island than for the continent.3 Although the existing analyses differ with regard to the precise quantification of non-tariff barriers (NTBs), as well as with respect to the models used and their parameterisation, the lesson is crystal clear: the UK is losing a much higher proportion of its economic output. In the various simulation studies carried out by the ifo Institute, the loss for the UK ranges from 3.5% to 1.5% and the EU27 average loss ranges from 0.25% to 0.6%.4

The bargaining power in the Brexit negotiations appears very unevenly distributed: The UK has a lot to lose, while the EU, on average, has little. So far, the EU Commission has cleverly used this perception to its advantage. But what if the long lever of the Europeans could be shrunk to a short branch by a cheap trick of the British?

What actually is a ‘Hard Brexit’?

A common trait in all known studies on Brexit is that they assume the reintroduction of customs duties and non-tariff barriers on both sides in a symmetric form in what is known as the ‘crashing-out’ scenario. Thus, it is always assumed that both the UK and the EU27 will apply the tariffs previously notified to the World Trade Organization (WTO) as the common external tariff. That is, for example, 10% on cars, 15% on small trucks, more than 20% on many foods, 68% on beef and so on. In addition there will be new regulatory and bureaucratic barriers that are often assumed to make trade in goods and services more expensive by a flat rate of 10%. The latest study by Felbermayr, Gröschl and Steininger, on the other hand, assumes that the reductions in trade costs empirically attributable to EU integration will be reversed in the case of a Brexit.5 Even if trade cost increases may be symmetric, they nevertheless have an asymmetric effect because trade structures naturally differ between countries and thus the tariff burden, as well as the so-called tariff incidence, is different.6 The UK imports a great deal of food from the EU, and food is subject to much higher customs duties than, for example, mining products or machinery. This alone results in a higher burden on the UK from new tariffs. In addition, the UK has a large trade deficit with the EU27, so that the tariff and NTB burden on the demand side is much higher than on the supply side.

After Brexit, the UK has options that the EU doesn’t

The existing studies make the assumption described above without asking whether this actually makes economic sense. Both the UK and the EU could, in the case of a hard Brexit, completely dispense with customs duties and with costly border controls or certification procedures. As regards customs duties in the case of a hard Brexit, the WTO rules naturally apply, in particular the most-favoured-nation (MFN) principle. This principle means that - in the absence of a free trade agreement (FTA) – each WTO member must treat the other members equally: in other words, after a hard Brexit, both the EU and the UK would have to charge each other the same tariff rate they would charge China, India and Brazil. But, at any time, countries are allowed to apply tariff rates below the maximum ones agreed in the Uruguay Round as long as they abide by the MFN principle, i.e. by applying the same tariff to all trading partners. In fact, it is optimal for a rational UK government to do just that and simply forego (fully) applying the bound tariffs. The country would avoid burdening 68% of its imports with high new tariffs.7 The political price for this would be to also make the rest of the imports, which are not from the EU or from countries with free trade agreements, exempt from customs duty. This applies in particular to China and the US and may cause problems in some sectors. But the sudden imposition of tariffs on more than two-thirds of imports certainly generates higher costs for the British economy, particularly for consumers, than the renunciation of tariffs on less than one-third of imported goods. For the EU, the situation is exactly the opposite, with UK imports accounting for just four percent of total imports. If customs duties were waived, trade with third countries without trade agreements would also have to be exempted from customs duties. But this is estimated to affect 80% of EU imports from countries outside of the Union.8 After Brexit, an asymmetric situation would arise: the EU demands customs duties, the UK renounces.9

In the case of NTBs, there is only a partial MFN obligation under the WTO rules. This means that the UK and the EU could in most cases simply wave through goods and services from the EU or UK or their existing free trade partners without having to forego controls over countries such as China or India. Such a laissez-faire strategy has risks; but its value is much higher for the UK than for the EU, similar to the case of tariffs.

In addition, the UK has a large trade deficit with the EU (in 2015 it amounted to more than 100 billion euro). This means, that consumers in the UK are more intensely burdened by tariffs and NTBs on imported goods than in a counterfactual situation where imports are lower, domestic production higher and trade roughly balanced. In the EU27, the opposite is true. In the case of food in particular, there would be extreme price increases in the UK, presumably in the range of 12-15%,10 similar to cars, textiles and shoes. These are all products that have a major impact on the daily lives of consumers. It is hard to imagine that a British government dependent on re-election would implement such a policy.11

For the same reasons, expensive border controls, including hygiene control, are very unlikely. There would be no real reason for this, at least initially, as the production conditions in the EU at the time of Brexit were fully in line with UK law.

Finally, it is also questionable whether the UK is at all technically capable of applying new import duties and carrying out product controls on two-thirds of its imports. Although this would also be a challenge for the EU, it would not be as great because the EU imports significantly less from the UK and has more resources available overall.

This is how London could shift the balance: The hard-but-smart approach

How will the economic costs of a hard Brexit be affected if the UK renounces the levying of customs duties on all trading partners and extensive controls on the EU and the existing FTA partners, while the EU collects customs duties and carries out controls?

Figure 1 shows simulation results of a general equilibrium model. Scenarios S1 up to S3 come from Felbermayr, Gröschl and Steininger;12 S4 is new. First, the analysis confirms that the hard Brexit – as it is usually defined in the studies – causes a damage equivalent to 2.8% of real consumption in the UK and that the real consumption per capita of an average Briton decreases almost four times as much as that of an average EU27 citizen. So far, so familiar.13

Figure 1
Effects on real consumption in the UK and EU27 in different Brexit scenarios
Percentage of baseline (year 2015)
Effects on real consumption in the UK and EU27 in different Brexit scenarios

Note: Simulation of scenarios as in G. Felbermayr, J. Gröschl, M. Steininger: Quantifying Brexit: From Ex-post to Ex-ante using Structural Gravity, CESifo Working Paper No. 7357, 2018, CESifo; additionally: hard-but-smart scenario as described above. All effects are statistically significant from zero at the 5 percent level. In all scenarios it is assumed that the UK will no longer pay net contributions to the EU (2015: 0.25 percent of GDP) and that the EU will spread this loss evenly across all remaining members. The disequilibrium in the trade balance between the EU and the UK is taken into account, but is kept constant. Perfectly flexible exchange rates. The approach takes into account, that NTB reductions within the EU to date have been heterogeneous according to econometric data analysis. The model covers 22 goods and 28 services industries and 44 countries (more than 90% of world GDP). S1 (Hard Brexit): MFN duties and NTB on both sides and UK vis-à-vis existing free trade partners; S2: as S1, but the UK concludes free trade agreements with the Commonwealth countries and the USA; S3: as S1, but the UK concludes free trade agreements with the EU (modelled on EU-Canada); S4: like S1, but the UK renounces all tariffs erga omnes and does not increase NTB tariffs vis-à-vis the EU and existing free trade partners. Details for 44 countries are listed in Table 1 in the Annex.

Source: G. Felbermayr, J. Gröschl, M. Steininger: Quantifying Brexit: From Ex-post to Ex-ante using Structural Gravity, CESifo Working Paper No. 7357, 2018, CESifo; calculations by the author.

Scenarios S2 and S3 also qualitatively reproduce what the literature reports. If the UK concludes free trade agreements with the US and its former colonies, the damage will be halved; this ‘global Britain’ strategy cannot compensate for the loss of full EU membership. However, the loss of the EU will increase slightly because, in addition to the new trade barriers, European buyers and sellers now face tougher competition in the British market from other suppliers such as those from the US. And even if, as in S3, the UK were to conclude a far-reaching free trade agreement along the lines of the EU’s modern treaties, the economic damage – although it would be very small for the EU countries on average – would remain. In addition, such a strategy requires a lot of time and concessions from the trading partners – not a matter of self-evidence in times of increasing protectionism.

Scenario S4, in which the hard-but-smart Brexit strategy is used, reduces the economic damage in the UK to half a percent. The reason for this is that now British consumption would not be burdened by Brexit because no new barriers would arise; on the contrary, by lowering tariffs to other WTO members to zero, consumer prices would fall even further. However, the EU is presumably introducing barriers, which is hampering British exports and lowering nominal incomes. In sum, some damage remains, probably in the form of additional depreciation of the British pound; no other scenario is more bearable for the UK, even though the EU is building new barriers.14

Table 1
Detailed effects for 44 countries
S1: Hard Brexit S2: Global Britain S3: Canada Dry S4: Hard-but-smart


-0,00 0,12 -0,00 0,01


-0,35 -0,38 -0,09 -0,28


-1,40 -1,46 -0,29 -0,96


-0,51 -0,50 -0,24 -0,46


-0,01 -0,01 0,00 0,00


0,00 0,26 -0,01 0,01


-0,01 -0,04 0,04 0,12


0,05 0,13 0,02 0,06


-1,37 -1,36 -0,35 -1,08

The Czech Republic

-0,75 -0,84 -0,35 -0,51


-0,72 -0,80 -0,20 -0,48


-0,89 -0,91 -0,12 -0,71


-0,39 -0,42 -0,13 -0,29


-0,70 -0,71 -0,27 -0,62


-0,50 -0,52 -0,08 -0,45


-0,52 -0,54 -0,10 -0,40


-2,76 -1,43 -0,93 -0,50


-0,39 -0,37 -0,12 -0,37


-0,34 -0,34 -0,04 -0,29


-0,87 -0,94 -0,34 -0,60


0,01 0,00 -0,00 0,02


0,02 0,20 0,00 0,07


-8,16 -8,22 -3,08 -5,39


-0,40 -0,43 -0,09 -0,31


-0,00 0,06 0,00 -0,00


-0,03 0,15 -0,09 0,06


-0,51 -0,55 -0,07 -0,42


-5,23 -5,46 2,15 -3,15


-0,58 -0,58 -0,16 -0,51


-0,01 0,04 -0,01 0,01


-5,19 -5,16 -0,76 -3,36

The Netherlands

-1,64 -1,71 -0,37 -1,06


0,52 0,61 0,23 -0,15


-0,69 -0,73 -0,25 -0,47


-0,45 -0,46 -0,12 -0,40


-0,37 -0,39 -0,16 -0,32

Rest of the world

-0,02 -0,02 0,02 0,02


0,01 -0,02 -0,08 0,05


-0,73 -0,77 -0,33 -0,38


-0,42 -0,46 -0,17 -0,33


-0,75 -0,79 -0,11 -0,64


-0,04 -0,08 -0,07 0,05


0,13 0,09 0,06 0,10


-0,01 0,11 -0,00 -0,01

Source: See note to Figure 1.

More important than this finding is that there is no longer any statistically significant difference between the effects in the UK and the EU. What applies to the average of the EU Member States also applies to e.g. Germany, which at 0.48% loses about as much as the UK, or France at 0.40% (see Table 1 for details). When the UK plays hard-but-smart, it suddenly no longer gets the short end of the stick.

Hard-but-smart: Not unrealistic

Now, one might argue that the British government did not want to implement such a scenario at all because it was interested in protectionism. However, there is little evidence of this; many Brexiteers do indeed demand more control at the borders, but by this they mean protection against uncontrolled immigration and not against imports. On the contrary: Theresa May has repeatedly spoken of a “truly global Britain” without referring to the hard-but-smart approach.15 The same goes for Trade Minister Liam Fox, who emphasises unilateral approaches in his speeches that already come very close to our Scenario S4. Similarly, there are (a few) economists who think Brexit is a good thing, most prominent among them is Patrick Minford, professor at the University of Cardiff and chairman of the association ‘Economists for Free Trade’. Critics rightly point out that the ‘Global Brexit Plan’, as modeled in Scenario S2 in Figure 1, can only mitigate the damage from a hard Brexit, but cannot reverse it. Even the hard-but-smart approach cannot convert economic costs into profit. But this approach definitely changes the negotiating position of the EU – clearly to its disadvantage.

It is often assumed that the UK would give up all its leverage by unilaterally lowering tariffs to zero in FTA negotiations. This is not true because the UK could return to WTO tariffs at any time.

Compared to all other scenarios, the hard-but-smart strategy leads to stronger sectoral distortions. The opening of agricultural and food markets would hit UK farmers hard. However, the effects in industry would presumably be very heterogeneous (see Table 2). The simulations suggest that where the UK has high imports of intermediate products and low exports of final goods to the EU, local value added could increase (pharmaceuticals, mechanical engineering and electronics). Imported inputs would become cheaper. The automobile industry, with its high exports to the EU, would have to accept significant losses. But, of course, the UK government need not apply zero tariffs in this sector but could rather maintain the EU’s external tariff of 10%. The service sector, on the other hand, would tend to lose out compared to other scenarios.

Table 2
Price-adjusted value-added effects for selected sectors
in percent
S1: Hard Brexit S4: Hard- but-smart S1: Hard Brexit S4: Hard-but-smart



-1,4 -0,5 7,9 -2,2


Food and beverage

-1,5 -0,6 1,9 -3,1



-1,1 -0,8 -5,7 -3,7



-0,7 -1,9 -3,1 8,7


Metal industry

-0,8 -0,6 -0,5 1,2



-0,6 -1,1 -8,5 3,7



-0,1 -1,0 -6,9 8,4



-0,7 -0,5 -0,5 -0,7


Postal delivery services

-0,9 -0,4 0,0 -0,3



-0,6 -0,5 -0,8 0,2



-0,8 -0,8 -1,6 -0,2


Computer + IT services

-0,4 -0,4 -0,6 -0,2


Financial services

-0,8 -0,5 0,4 0,2


Legal services

-0,5 -0,3 -1,5 -0,9


Business services

-0,4 -0,1 -2,6 -2,1

Source: See note to Figure 1.

Businesses and the EU Commission have repeatedly pointed out that London is not prepared to impose new controls on more than two-thirds of its imports. If this is the case, the smooth waving through of imports would not only be the rational economic response to Brexit, but also the only one that is technically possible. Moreover, the UK does not give away any options because, when it is technically ready, it can raise the applied tariffs to the higher most-favoured-nation tariffs in full compliance with WTO rules. This is to be expected when the first free trade agreements with major trading partners are concluded.

Finally, political pressure in the UK may well ensure that customs duties and the like are discarded. Indeed, it would not be easy for the government to explain why beef should become 70% more expensive and dairy products 20% more expensive as a result of ‘taking back control’ when the UK is not at all obliged to apply the maximum tariffs agreed upon in the Uruguay Round by the EU. On the contrary, the abolition of all customs duties would be a first and very visible act of a government that will be sovereign in terms of trade policy from the day the country leaves the EU onwards and a tangible sign in citizens’ wallets of the newly achieved independence from Brussels.

The EU suddenly under strong pressure

If the UK were to play the hard-but-smart strategy on a disorderly Brexit, the EU27 would have made the wrong bet. For British consumers, many goods prices would not only not rise, they would fall because tariffs vis-à-vis all trading partners would be omitted.16 The same applies to industrial inputs. There would be no supply bottlenecks because all goods and services would simply be waved through as before. There would be no congestion on the French side of the Channel or in Oostende; the queues would form on the British side, caused solely by the EU, which is hardly better prepared for customs clearance than the British.

Continental lovers of scotch and cheddar cheese as well as European manufacturing industries buying raw materials and intermediate products from the UK (with respect to their weight in trade statistics) would be confronted with higher prices. They would express their displeasure quickly and clearly. The EU would be under great pressure to conclude a free trade agreement with the UK in order to quickly negotiate away the tariffs that had just been introduced, in a way that is compatible with WTO rules, without having to renounce barriers vis-à-vis China and other third countries.

What to do now: Postpone article 50, search for a solution

It would be much better to put off Article 50 and immediately negotiate a far-reaching new comprehensive free trade regime, for example within the framework of a new European Customs Association.17 The EU must offer much more than an agreement in the style of the EU-Canada-FTA. As Figure 1 shows, the hard-but-smart strategy for the UK is better than such a free trade agreement.

Scenario S4 makes one point quite clear: the strategy of ‘punishing’ the UK with a hard Brexit in order to deter followers can dramatically backfire as the British lose no more than continental Europeans. The assertion of the dogma “extra ecclesiam nulla salus” has already failed to save the old Catholic Church from division.18 Applying it to the EU might not be the most promising strategy.

In view of these findings, the EU should urgently consider whether the danger of a hard Brexit may actually be greater for the EU itself than previously thought. It’s time to take a more constructive approach to either avoid Brexit altogether or to adapt the divorce agreement in such a way that the hard-but-smart scenario could be avoided as well. This would allow both parties to better position themselves and it would be a fantastic side effect if the important peace issue in Ireland were also to be resolved in the framework of a more constructive approach.

* This is the translated and updated version of G. Felbermayr: Brexit: Eine “Hard-but-Smart”-Strategie und ihre Folgen, in: ifo Schnelldienst, Vol. 72, No. 4, 2019, pp. 27-33.

  • 1 These figures come from the IMF’s Direction of Trade Statistics and refer to the year 2015. These data have not yet been distorted by the Brexit referendum of 23 June 2016 and are therefore more suitable for analysis than those from 2018. The free trade agreements, on the other hand, are those that are currently in force according to the WTO. For these agreements, a flat-rate zero tariff regime has been assumed. The unilateral preferential agreements EBA, GSP and GSP+ have not been taken into account.
  • 2 Of course, there is a high degree of heterogeneity within the EU: Ireland, for instance, receives 32% of its imports from the UK and delivers four percent of its exports there.
  • 3 There are already some surveys that summarise existing economic estimates, such as G. Tetlow, T. Stojanovic: Understanding the Economic Impact of Brexit, London 2018, Institute for Government; T. Sampson: Brexit: The Economics of International Disintegration, in: Journal of Economic Perspectives, Vol. 31, No. 4, 2017, pp. 163-184; B. Busch, J. Matthes: Brexit: The Economic Impact – A Survey, in: CESifo Forum, Vol. 17, No. 2, 2016, pp. 37-44.
  • 4 See R. Aichele, G. Felbermayr: Kosten und Nutzen eines Austritts des Vereinigten Königreichs aus der Europäischen Union, Study for the Bertelsmann Stiftung, 2015, available at; G. Felbermayr, J. Gröschl, I. Heiland, M. Braml, M. Steininger: Ökonomische Effekte eines Brexit auf die deutsche und europäische Wirtschaft, Study commissioned by the BMWi, ifo Forschungsbericht No. 85, Munich 2017, ifo Institute; G. Felbermayr, J. Gröschl, M. Steininger: Quantifying Brexit: From Ex-post to Ex-ante using Structural Gravity, CESifo Working Paper No. 7357, 2018, CESifo.
  • 5 G. Felbermayr, J. Gröschl, M. Steininger, op. cit.
  • 6 Customs duties are paid by importers, but importers can often pass them on at least in part to exporters abroad. This depends on the elasticities of supply and demand, which naturally differ from sector to sector. NTBs are also borne both by importers and exporters.
  • 7 For the sake of completeness, it should be noted that the EU does not levy customs duties on about a quarter of all products in the customs classification. The 68% mentioned refer to the so-called “dutiable trade”.
  • 8 China, the US, Russia, India and Brazil alone, with which the EU has no preferential agreements, account for almost 50% of extra-EU imports.
  • 9 WTO law even allows the UK to raise tariffs back to the maximum level at a later date. If the maximum duty laid down in the WTO agreements is undercut by the duty actually applied, this is called “water in the tariff”. Many developing countries use this strategy. They can raise their tariffs back to the ‘bound level’ at any time. In this way they retain their negotiating power and can turn the applied customs duties into contractually binding concessions in bilateral agreements.
  • 10 According to the WTO, the EU’s external tariff in the agro-food sector averages 14%. However, as the UK imports mainly processed foods from the EU for which the tariff rate is significantly higher (20%), it can be assumed that part of the tariff is absorbed by EU suppliers; hence the estimate of 12-15%.
  • 11 WTO law allows maximum tariffs to be differentiated for only a few product lines. This would allow the UK to continue to protect particularly sensitive industries.
  • 12 G. Felbermayr, J. Gröschl, M. Steininger, op. cit.
  • 13 In other studies the difference is even more pronounced; this is because the current ifo figures allow a high degree of sectoral heterogeneity and asymmetry between the UK and the EU27 and do not artificially zero trade balances as other studies do.
  • 14 These figures refer to the long term. If UK exports were burdened by barriers but imports were not, the trade deficit would increase. With flexible exchange rates and a given propensity to save, however, this leads to a depreciation of the British pound, so that the deficit remains constant with the new trade policy constellation.
  • 15 See Theresa May calls for ‘truly global Britain‘, BBC, 2 October 2016, available at
  • 16 One may argue that in a hard-but-smart strategy, the British pound would depreciate to restore some of the lost competitiveness of the UK’s export businesses. However, the exchange rate being a forward-looking price, this reaction is probably already accounted for in the current rate.
  • 17 This proposal is detailed in G. Felbermayr, C. Fuest, H. Gersbach, A.O. Ritschl, M. Thum, M. Braml: Hard Brexit ahead: breaking the deadlock, EconPol Policy Brief No. 12, 2019, EconPol Europe.
  • 18 The dictum “Outside the (Roman Catholic) Church there is no salvation” was raised to the status of dogma at the Council of Ferrara/Florence (1438-1445) in times of looming church division. This approach neither prevented the separation into an Eastern and a Western church nor the successful emergence of Protestantism.


Search results on EconBiz

Under reconstruction

Due to technical conversion work in our archive and therefore in Intereconomics articles, formatting errors may occur. We apologize for this temporary inconvenience as well as the disruption of some website functions (e.g. search, comments, keywords ...).