Negotiating the UK's Post-Brexit Trade Arrangements

Free trade areas and rules of origin

The critical difference between a CU and an FTA is that whereas the former allows goods, once inside the area, to circulate without facing any additional tariffs because they will have faced the same tariff wherever they entered the area, the latter cannot do this. If one member of an FTA has a zero tariff on, say, oranges, while others have positive tariffs, exporters would seek to send their oranges to the first country and serve the others from there, thus avoiding the other members' tariffs. To avoid this trade deflection, FTAs have rules of origin (RoOs) to determine whether a good has been produced within a member country, in which case it is exempt from tariffs under the FTA agreement, or whether it has been produced outside, in which case it has to pay the tariff of the country of destination.

Applying a RoO is straight-forward for simple goods like oranges, but most manufactured goods are produced by combining many inputs, some of which may come from third countries. In these cases, the RoOs can be complex, but typically a product needs to contain 60 per cent local value added to be eligible for duty free import into the EU under the European Economic Area agreement and we might anticipate the same for the UK. Enforcing such RoOs means customs checks of some kind between the EU and the UK even in an FTA. These could be minimal if electronic documentation can be used but they would give French customs at Calais the right to inspect UK trucks to ensure that the RoOs are satisfied. All this adds cost, delay and uncertainty to the trading relationship. A major concern, especially if one is hoping that the UK economy will respond to Brexit in a flexible and innovative way, is that, while for large companies the cost of establishing origin is mainly a fixed cost which can be spread over a large volume of sales, for small, and even more so, one-off, exporters the bureaucratic costs can be proportionately very high – even prohibitive.

‘Regulatory Unions’

For the sectors that are not granted conditions equivalent to the Single Market similar problems arise with standards and the certification process by which conformity to standards is checked. The EU is a not only a Customs Union but also a ‘regulatory union’, which means that all members incorporate EU mandatory standards into their legislation, along with the rules for establishing conformity with the rules. Hence products produced within the EU can be assumed to conform to mandatory standards and can circulate without further inspection to any other part of the EU. At present, for the French authorities to stop a UK lorry at Calais to check that its contents met EU standards would be legally equivalent to stopping a lorry passing from Kent to Sussex for that purpose – that is, a gross violation of the right of citizens to trade with each other. If the UK were outside the Single Market, however, such inspections could not be ruled out, absent an agreement specifically to that effect. And in order to sell in the EU the UK trader would have to have her product checked by an officially accredited agency, which status is, in the EU, probably restricted to EU agencies. ³ This brings further cost, delay and uncertainty.

In services, the absence of a regulatory agreement could prevent cross-border trade altogether. The most discussed case of this is the so-called passport for financial services firms. This allows a firm established in one EU country to sell services in any other EU country without establishing there. In this sector establishment entails being regulated by the local authorities which means not only preparing all sorts of documentation and accounts, but meeting local prudential capital requirements. These typically require a minimal level of assets to be held and the fragmentation of doing this in several places instead of just one is a major cost; moreover aggregating activities over several locations allows a much freer movement of funds across borders and hence more efficient financial operations.

It will be possible to negotiate approximations to singlemarket arrangements sector by sector, as indeed the Swiss have done, but it will be a major operation. Moreover, our guess is that if the UK curbs inflows of labour, it will be permitted to do so in only a subset of service sectors at best. An interesting and important issue will be which sectors the UK government chooses to favour in this way, especially given the intensive lobbying that we see (and presumably much that we don't see) from various sectors – e.g. motor cars, financial services and pharmaceuticals.

Relative UK and EU negotiating strength

It is often argued – sometimes from very senior positions – that because the EU exports more to the UK than the UK exports to the EU, the EU will be keener to strike a free trade agreement and therefore more accommodating in negotiations. This is seriously misguided. While 45 per cent of UK exports go to the EU, an average of 6.7 per cent of EU countries' exports go to the UK. ⁴ In other words, while the average UK exporter faces disruption to 45 per cent of its exports the average EU exporter faces problems with 6.7 per cent of its exports. Who is going to suffer the bigger disruption and need a deal more?

The European Economic Area (EEA)

One proposal has been that the UK could circumvent these difficulties by joining the European Economic Area (EEA), thus obtaining Single Market conditions across the board. We do not see this as a likely solution. The EEA formally consists of the EU plus three members of EFTA (Norway, Liechtenstein and Iceland). The latter become part of the EU ‘regulatory union’: they have to adopt all EU legislation in the covered sectors (all goods and services except agriculture and fisheries), including free movement of labour and with no say on them, but in return get access to the whole Single Market free of all technical barriers. The ‘regulatory union’ also implies, however, that individual members cannot sign agreements with third countries that affect technical regulations.

But membership of the EEA does not mean that there are no barriers. Being a Free Trade Agreement rather than a customs union, the EEA needs Rules of Origin. Duty-free circulation applies only to goods ‘originating’ in member states. For example a Toyota car that had more than 40 per cent of its components imported from outside the EU would not be considered a UK car and would have to pay the EU's MFN tariff to enter. This would be very unattractive to foreign investors, as the comments from Carlos Ghosn, Chief Executive of Nissan, on 29 September demonstrated. ⁵

Article 50 and the timing of negotiations

The key objective with the EU should be to start discussions on the future trade relations as soon as possible. The content of Article 50 negotiations and the role of trade negotiations in the exit process are actually ambiguous. ⁶ One interpretation suggests that negotiations on trade cannot begin until after Brexit, ⁷ but a more forgiving interpretation can be drawn from Article 50(2), which states that an exit agreement should take “account of the framework for its future relationship with the Union”. This seems to include the possibility of trade negotiations taking place in parallel to, or as part of Article 50. This is, however, a slender hook on which to hang our hopes and the EU 27 may refuse to play ball. Moreover, we have to recognise that while the Article 50 negotiation is effectively time-limited and approved in the EU27 by qualified majority vote, a trade agreement is not so constrained and requires unanimity among EU Member States. On the precedent recently established for the FTA with Canada, the latter seems likely to entail approval by up to 38 different European parliaments.

However, such a clause could be challenged in the WTO because it would give the UK and the EU preferential access to each other's markets while not constituting a customs union or free trade area as defined by GATT's Article 24 or GATS Article 5. And even more challenging would be the UK politics: the peace clause might be taken as avoiding Brexit by the back door and might well involve the UK having to continue to allow the free movement of labour for its duration. Clearly these objections may be sufficient to prevent such a scheme and so we have to accept that there could well be a period during which the UK had exited the customs union and the Single Market and yet had not secured a free trade agreement. This has come to be termed a ‘hard’ Brexit.

‘Hard’ Brexit

A ‘hard Brexit’ entails trading with the EU27 purely on WTO terms and not bothering even to try to negotiate special arrangements. This does not seem attractive to us. As table 1 shows, The EU takes around 45 per cent of UK exports of goods and provides around 54 per cent of UK imports of goods. All of this trade is carried out with zero tariffs and very low non-tariff barriers courtesy of the customs union and the Single Market. If that trade were carried out on an MFN basis the EU tariff would be 5.3 per cent. ⁸ At a more detailed level around 16 per cent of UK exports to the EU-27 would face tariffs of more than 7 per cent, with half of that trade comprising motor cars which would face a tariff of 10 per cent.

It is very difficult to measure the effect of non-tariff barriers on trade volumes, but they are generally held to be high. The costs of administering the RoOs will reduce trade – in some estimates by up to 9 per cent (Cadestin et al., 2016) – and differences in standards and the additional burden of certification is bound to increase this. It is not that trade will cease, but that the extra cost will discourage some trade and cut the returns to that which remains. The costs of the bureaucracy have to be found somewhere – will it come out of profits or wages? Turning to services, a hard Brexit is even harder. The data behind table 1 suggest that around a third of UK exports to the EU are of services. As noted previously, the loss of the financial passport is likely to cut exports and cause firms to relocate abroad. An overall assessment, however, is not possible without a massive research effort, because there is no uniform EU external trade policy for services. The EU's GATS schedule sets out a framework for market access which is punctuated by individual countries' derogations in particular subsectors and modes of supply. For example, while the conditions for establishing commercial presence (mode 3) in life insurance are similar across most EU members, those pertaining to the legal sector (advice on foreign law) differ widely; e.g. in France non-EU firms are not allowed to establish branches under their own name or to form partnerships with locally licensed lawyers.

It is universally accepted that dropping out of the, albeit imperfect, Single Market for services and trading under the EU's provisions for third countries will cut access considerably. Moreover, these provisions are typically more liberal than those incorporated in the EU's GATS schedule, and thus while they may be quite favourable, they could be removed at any time and thus are afflicted by considerable uncertainty that does not apply while the UK is within the EU Single Market.

Transhipment and global value chains

Why might anyone object to a mere relabelling exercise? Possibly out of spite, but let us discount that. The real problem is that an x per cent tariff levied by the UK and the EU as separate markets is not perfectly equivalent to an x per cent tariff by a Single Market comprising the EU27 and the UK. Because of the customs union and the Single Market, a third country facing an x per cent tariff on imports into the EU28 (i.e. including the UK) pays the tariff and then can circulate its goods freely around the whole union. If the UK is not part of the customs union and Single Market, on the other hand, goods cannot flow freely between the UK and the EU27; there may be tariffs to pay or inspections of standards at the border between them, which add to costs and uncertainty. For example, whereas Kenya currently exports its cut flowers to Rotterdam, from where they are distributed rapidly to the UK among other places, after Brexit they may have to pay UK import duties as they enter the UK or be inspected for pests or disease. The tariff may be the same in the EU and the UK (and so it does not matter to the exporter where she pays it), but after Brexit the market separation imposes more costs on Kenyan exporters, so Kenya may challenge the rectification, arguing that its rights under the Uruguay Round are impaired and that it needs to be compensated by a tariff reduction.

The position is more complicated for intermediate goods and value chains. Imagine a case where an intermediate good is imported into Britain, incorporated into a manufactured good and then sold in Germany. It pays its x per cent on the way into the UK; prior to Brexit the final good can be sold in Germany with no further questions asked, but once the markets are split the final good may no longer meet EU RoOs and so face an additional tariff on entering Germany. No tariff rate has changed but the exporter of the intermediate good is now disadvantaged.

Similarly, RoOs create concerns that value chains for EU final products could be disrupted. Final goods produced in the EU using parts and components from the UK risk losing their preferential status in EU FTA partners because the UK content could no longer be counted towards EU ‘local content’ and thus the FTA partner's RoO is not met. ¹⁰

Separating UK and EU agricultural commitments

There are also potential problems in agriculture. Under the WTO Agreement on Agriculture, some agricultural imports from the rest of the world face tariff rate quotas (TRQs), which allow the entry of a specified quota of imports at a below-normal tariff rate. These quotas are defined at the EU level, and any division agreed between the EU27 and the UK could be challenged in the WTO as impairing the rights that others had acquired under the Uruguay Round Agreement on Agriculture. For example, suppose that a product is imported by France but not by the UK and that it pays the below-normal tariff as it enters the former. If the whole of the TRQ were assigned to the UK, France could then raise its tariff on the products to the normal level. No tariff rate would have been changed, but tariff payments would have increased. Similarly, the cap on expenditure on tradedistorting agricultural subsidies that the EU negotiated in the Uruguay Round (the so-called Amber box) needs to be divided. According to whether it goes to the UK or to Italy, it may be spent on sheep-rearing or on olives, with obviously different implications for different third parties.

Hence although the division of the TRQs and the Amber box would properly form part of the Article 50 negotiations on UK withdrawal, they would still then, in principle, have to be agreed with the WTO membership. ¹¹ If the UK does not manage to do this before it leaves the EU, it could emerge with an ambiguous position despite applying exactly the same tariffs and customs procedures, possibly opening itself up to legal and diplomatic complexities and possible trade conflicts.

Pragmatism vs. legalism within the WTO

One bright side of this situation is that the WTO is a very pragmatic organisation, which does not generally trip over legal niceties. (For example, the EU commitments to the WTO have not yet been formally updated to take account of the enlargements of 2004 and 2007, and yet trade continues without challenge.) Thus the key to preventing these sorts of problems from arising, as well as those surrounding a possible EU ‘peace clause’ is rapid and active diplomacy with partners in the WTO and in the EU. In the WTO it might be sensible to find a sympathetic WTO Ambassador to set up a discreet ‘friends of the UK’ group to help the UK navigate the shoals.

This is because the problems noted above affect the value of EU27 concessions to trading partners as much as they affect the value of the UK's. Thus the UK's departure potentially opens the EU to demands from WTO Members to re-negotiate its schedules, so there is a shared interest in regularising relations with WTO partners as quickly as possible.

None of this should be viewed as necessarily ruling out changes to the UK's relationships with non-EU countries, merely a way of ordering such changes in a way that allows them to be properly considered. For example, it is possible that the UK may want to reduce its level of agricultural protection and it may wish to seek deeper market access agreements for services. From the other side, developing country partners may wish to extend services talks to include Mode 4 (movement of natural persons), and developed country partners to break new ground by providing rights of establishment for foreign services providers. While all of these issues could be broached multilaterally in the WTO, with the exception of agriculture, they are more likely to be tackled in the context of bilateral or regional FTAs. It is to these that we now turn.

Developing countries

The largest group within the set of third countries is the developing countries that receive unilateral preferences in the EU through the Generalised System of Preferences (GSP), including the provision of duty-free quota-free imports for nearly all goods under the Everything But Arms provisions. These are relatively simple to deal with. The GSP is a unilateral concession by the importing country and so it is entirely within the power of the UK government just to announce that it will offer exactly the same terms as the EU has on Brexit-day. This would not offer precisely the same value to developing countries as did the GSP for the EU28, but the losses would generally not be huge and might be handled ad hoc if they were pressing. The loss of value stems first from the separation of the EU and UK markets, as explained above, and second from the operation of the competitiveness clauses. These clauses require that when a developing country supplies more than 17.5 per cent (14.5 per cent for textiles) of the donor's imports of a product it loses its preference. Given that the percentages would now be calculated in the UK and the EU27 separately they may disrupt certain existing trades. A third possibility is that if the UK reduces its tariffs on other, developed, countries the preferences under the GSP are eroded. However, as we argue below, we do not see this as a very strong possibility.

This will involve a modicum of diplomacy but would create certainty for developing countries and postpone a large number of potential interactions. After five years, when the main elements of UK trade policy had been clarified, the UK should undertake a full review of its GSP provision.

Free trade agreements

As noted above, a strict interpretation of EU treaty obligations might suggest that UK negotiations with third countries cannot begin until after Brexit. The EU might put diplomatic pressure on trading partners not to jump the gun with the UK even informally, although the signs are that British commercial diplomacy is already underway.

Just as with the WTO and developing countries, this may still entail inconveniences for our partners, the extent of which will depend critically on what access the UK retains to the EU market. ¹² There were 51 countries with FTAs with the EU in 2015, accounting for 14 per cent of UK exports and 11 per cent of imports. If instead of negotiating an FTA of some sort, the UK traded with these countries under WTO rules, UK exports would face average MFN tariffs that varied from under 5 per cent (Israel) to almost 30 per cent (Egypt) and perhaps, most notably, an average 17 per cent tariff into Korea. ¹³

Once the arrangements with the EU and the WTO are reasonably clear (‘fully determined’ may be too high a threshold), there are still around 20 potential new UK FTAs. These include FTAs with China, India, Brazil and various ASEAN members among rapidly growing emerging markets, as well as the USA, Australia and New Zealand among developed countries. These will need to be prioritised according to the size and diversity of the production in and markets of potential partners, the size of barriers to trade and investment, the growth of domestic demand in areas of British competitiveness, and complementarities with UK economic structures (i.e. products that could fit with UK supply chains).

The government also needs to consider how arduous the negotiations would be and how much effect marginal effort would have on the outcome. For example, Allee and Lugg (2016) show relatively formally that a negotiation with the USA largely results in the partner accepting US positions on issues such as patent protection, ISDS and agriculture. This is not just a matter of relative size because we see that Japan has not managed to win many concessions in the TPP, and so statements about how well the UK will do in a negotiation, let alone how quickly agreement will be reached, are not very convincing.

Negotiating agreements with countries with the same language, where the legal system and trade objectives are broadly similar – so there are likely to be fewer opportunities for misunderstandings and mistakes – is attractive. The Commonwealth is one potential source, as recent preliminary discussions between the UK and Australia underline. However, as Gasiorek (2016) points out, the Australian government has said that it intends to agree an FTA with the EU before entering one with the UK and trade in goods with Australia is quite small, so it hardly represents an alternative to the EU. The prize with Australia is in services, which account for over half of total UK goods and services exports to Australia. But services agreements are arduous.

Another option for ease of negotiation is applying to join an existing mega-regional agreement, most notably, in violation of geography, the Trans-Pacific Partnership (TPP), subject to it being ratified/implemented by the USA. The upside of TPP is having an FTA with twelve countries including the US, Japan, Singapore, Australia, New Zealand, Malaysia and Vietnam: all likely targets for the UK's independent trade policy. Essentially, however, negotiation would mean turning up for the signing ceremony. It is unlikely that any member would be interested in recasting an agreement that took more than five years to craft for the sake of UK membership.

Joining the Trans-Atlantic Trade and Investment Partnership (TTIP) – should it be signed or implemented – is another possible approach to cutting the negotiation burden. Like TPP however, TTIP would likely be a take it or leave it deal. Nonetheless, signing TTIP could, at a stroke, provide a better deal than MFN with the USA. However, the TTIP would provide only the palest imitation of the market access that the UK would receive in the EU if it remained in the Single Market. Also, since TPP and TTIP have significant regulatory provisions they would also result in an apparent loss of sovereignty and thus prompt public opposition.