The myth of a 10-year grace period, Brexit and trade talks with the EU

Conservative MP Jacob Rees-Mogg was wrong about this but he’s never corrected his mistake, and the myth persists. What is the claim and why is it wrong?

The need for an interim free trade agreement is a myth

By Peter Ungphakorn

It’s not often that hard Brexiters make WTO rules more complicated than they need to be.

Usually their error is to over-simplify.

But the mistaken identity of interim free trade agreements in the WTO is one rare instance.

The idea had apparently been knocking around for some time, at least back to March 2017 in a Politico article.

It reappeared back in May 2018, when Conservative MP Jacob Rees-Mogg claimed on television that WTO rules allow the UK a 10-year grace period to negotiate a free trade agreement with the EU.

It’s all about discrimination
The 10-year period is irrelevant
The 10-year period would be a nuisance
Contiguous territories and frontier traffic?

A real beginner’s guide to GATT Article 24, June 24, 2019
• A shorter version of this article published on UK Trade Forum, Jan 26, 2019
• Even shorter, One last go. The Article 24 red herring in less than 400 words. Think ‘highway code’, Feb 16,2019
• The final section, “The magical GATT Art24”, of “Caught up in a war — the WTO and Brexit”, April 6, 2019

The claim baffled many trade experts.

Some lawyers did recognise what he was referring to. They pointed out he was wrong.

Nevertheless, the claim has resurfaced repeatedly over the months since then, and it has been repeatedly knocked down. And yet it persists.

This is what the MP for North East Somerset said on BBC TV’s Daily Politics programme on May 10, 2018:

“If you are in negotiation for a free trade agreement, you can maintain your existing standards for 10 years under WTO rules. So we have 10 years from the point at which we leave the European Union to negotiate a free trade agreement, which would mean we could carry on with our zero tariffs.”

(That excerpt starts here, from 15:31; the whole segment is here)

A few days later on May 22, on the same programme, he went further. He said the same WTO rules allow preferential treatment between countries with contiguous borders — suggesting the provision would allow the Irish border to stay invisible — and repeated his claim about the 10-year period.

Also on the programme was Cambridge University trade lawyer Lorand Bartels who said he did not recognise the legal provision Rees-Mogg cited despite years of studying relevant WTO agreements. But he responded to one of the MP’s remarks with: “in that case we’re in agreement.”

That last comment has been interpreted by some Brexiters as proof that Bartels confirmed Rees-Mogg’s argument. In fact he did nothing of the kind.

What’s going on here? To sum up

  • The 10-year period was designed for countries setting up free trade from scratch, not weakening integration.
  • For Brexit, citing it is unnecessary because it isn’t needed
  • Citing it also adds unnecessary complications

Or to put it another way, if the UK does cite it, it will face three hurdles:

  • The UK would have to reach agreement with the EU. The UK could not do this unilaterally. So this isn’t exactly “no deal”.
  • That agreement would have to include a plan and timetable for achieving the final agreement. And it would have to have a sufficient amount of detail, including what the final agreement would look like, because …
  • … the WTO membership could demand changes, if they weren’t convinced that the plan could be achieved within about 10 years. The UK and EU would have to accept those changes or scrap the agreement. (This doesn’t apply to free trade agreements that are not interim.)

In other words, the 10-year “grace period” has been used to cite WTO rules incorrectly by people who only have a superficial understanding of the rules.

Bartels blames advisers of pro-Brexit politicians “It’s amazing how this awful misinterpretation of Art XXIV GATT won’t die, no matter how many times I point this out, directly to @ShankerASingham among others. Is there an underlying reason it is so popular with Brexiteers I wonder?”

Or as former Australian trade negotiator Dmitry Grozoubinski tweeted: “It’s utter nonsense. It relies on your being too busy to read Article XXIV of the GATT, or too confused by trade legalese to parse it.”

Most-favoured nation treatment

Myth: this is not the way to get around MFN

It’s all about discriminationBack to top

The basic legal issue here is discrimination. One of the WTO’s key rules is that countries should treat their trading partners equally. In WTO jargon this is called most-favoured nation treatment (MFN) — favour one; favour all.

As with many rules, exceptions are allowed. One of the most important is when countries unliterally give trade preferences to their poorer trading partners, such as exempting import duty on goods from developing or least developed countries.

Another — and this is the issue here — is when two or more of them have free trade agreements. The EU’s standard non-discriminatory import duty rate (or tariff) on some types of shoes is 8% of the price. That normal rate is charged on imports from the US, but the same types of shoes are traded preferentially, duty-free, between France and Germany or Denmark and the UK, since they are all in the EU’s single market, a particularly deep form of free trade agreement.

For goods such as shoes or milk, the rules governing free trade agreements come under Article 24 (XXIV) of the WTO’s General Agreement on Tariffs and Trade (GATT). This is the article Jacob Rees-Mogg was citing. (There are separate rules for services.)

There are a few important points to remember about GATT Article 24:

  • It covers: ordinary free trade agreements between two or more countries — where trade among them is duty-free (and may also be covered by other rules), but they are free to set their own tariffs on imports from elsewhere — and customs unions — where the participating countries all also set the same tariffs on imports from outside the union. The EU single market includes a customs union.
  • The rules are not as clear as they should be. In particular, whether a free trade agreement or customs union, the deal is supposed to cover “substantially all the trade”. This means if only one sector is covered, such as a free trade deal in cars, that would violate GATT Article 24. But it also allows some products to be excluded, and this creates a grey area. Some agreements exclude all or most agricultural products. There has never been a legal ruling on whether this violates Article 24. And in their committees WTO members have never been able to agree on what they mean by “substantially all the trade”.
  • The original GATT dates back to 1948. Since then it has been amended, updated and clarified, including a new “Understanding” on Article 24 agreed in 1994. (All versions along with a WTO General Council decision and legal interpretations can be found here or on the WTO website here including as a pdf file here.)
10-year interim agreement doesn't make sense
Unhelpful myth: let’s stop wasting time on it

The 10-year period is irrelevantBack to top

Where did Jacob Rees Mogg get the idea of a 10-year grace period? You have to dig down to the 1994 “Understanding” (a new agreement providing interpretation) on GATT Article 24 to find it.

It’s all about an “interim agreement” leading eventually to a fully-fledged deal. Article 24 allows countries to announce an interim agreement so long as they supply a plan and timetable for setting up the complete deal “within a reasonable length of time”. The 1994 “Understanding” says that should normally be no more than 10 years.

But, as Bartels said on May 22, these provisions were designed for a situation where countries start with no agreement and move to free trade. They are allowed to set up an interim arrangement while they negotiate the final deal.

Brexit is different. The UK is closely integrated into the EU single market and its customs union. It is leaving that, and intends to negotiate a looser free trade agreement of some kind with the EU.

At a key moment in the TV programme, Rees-Mogg and Bartels interrupted each other, which means we didn’t get to hear what they meant in full.

Rees-Mogg said: “A preliminary agreement can be a continuation of what you’ve already got …”. Bartels cut in: “Oh, in that case we’re in agreement”. Rees-Mogg welcomed the reply.

But did they really agree? Crucially, not on the 10-year period.

The point is that the UK and EU can simply agree between themselves to continue with what they’ve “already got”. In that case they don’t need to notify anything to the WTO until the present arrangements are terminated. The 10-year period is irrelevant because there is no “interim agreement”. But that’s not what Rees-Mogg was saying at all.

Alternatively, if UK and EU go for the transition-period customs union written into the Withdrawal Agreement, that could also be notified to the WTO simply as a new customs union understood to be in place until it’s terminated or replaced. Again, as far as the WTO is concerned, no need for an “interim” agreement or a 10-year period.

That seems to be the consensus among trade lawyers who have studied Article 24 and Brexit.

There is a further technicality to consider. The current arrangement of what the UK has “already got” is notified to the WTO in April 2013 under the WTO’s goods and services agreements, with a link to a single EU text (the notification is here). That suggests that for Brexit, the UK and EU will have to notify any post-Brexit arrangement since the WTO notification is based on the UK being an EU member.

GATT extract on how contracting parties (WTO members) can scrutinise an interim agreement
Conflict: the myth could lead to a fight with WTO members

The 10-year period would be a nuisanceBack to top


Out of 12 agreements notified to the WTO as “interim” under GATT Article 24, only one could be considered significant with provisions for further negotiations: the “Interim economic partnership agreement between the European Union and the ESA states (Madagascar, Mauritius, Seychelles, Zimbabwe)” — Signed on August 29, 2009; entered into force on May 14, 2012; notified to the WTO on July 27, 2012; full implementation by January 1, 2022. (See this WTO Secretariat note; or search for all documents)

The only other major agreement notified as interim is the EU-Chile Association Agreement, but replying to questions from WTO members, the EU and Chile both said they notified it as “interim” because it contained a 10-year transition period, but that the agreement itself was final and would remain in force after the transition.

Of the rest, three are between the EU, EFTA and Turkey, and the Palestinian Authority. One is EU-Lebanon (12 years from March 1, 2003). Another is EU-North Macedonia (as it’s now called). The rest are between the EU and countries that have now joined it.

This search link will produce all WTO documents on free trade agreements that include “interim” in the title (but note that the one between Albania and Kosovo uses “interim” in a different sense).

Before the WTO replaced GATT in 1995, interim agreements under GATT Article 24 were more common — see pages 33–36 of this GATT document on agreements notified 1986–94.)

It gets worse. Countries tend to avoid going through the rigmarole of notifying and defending interim agreements in the WTO. Here’s why:

First, even though the agreement is interim, the notification cannot be a simple piece of paper saying “Hey, WTO! We’re negotiating a free trade agreement. It may take 10 years. While we’re doing that, we might violate some of your non-discrimination rules.”

It has to contain details, including a plan and timetable for concluding the final agreement.

Second, as the name says, it would have to be formal agreement between the UK and EU, meaning the EU would have to be on board too.

In theory, the transition customs union and the Protocol on Northern Ireland / Ireland (the “Backstop”) in the Withdrawal Agreement could qualify. The attached non-binding political declaration on the future relationship would not, since it’s not an agreement.

Worst of all, the details are likely to be scrutinised by WTO members and they can demand changes to the agreement. This is what GATT Article 24 Paragraph 7 says (note this keeps the pre-WTO language, so “CONTRACTING PARTIES” now means WTO members):

(b) If, after having studied the plan and schedule included in an interim agreement referred to in paragraph 5 in consultation with the parties to that agreement and taking due account of the information made available in accordance with the provisions of subparagraph (a), the CONTRACTING PARTIES find that such agreement is not likely to result in the formation of a customs union or of a free-trade area within the period contemplated by the parties to the agreement or that such period is not a reasonable one, the CONTRACTING PARTIES shall make recommendations to the parties to the agreement. The parties shall not maintain or put into force, as the case may be, such agreement if they are not prepared to modify it in accordance with these recommendations.

(c) Any substantial change in the plan or schedule referred to in paragraph 5 (c) shall be communicated to the CONTRACTING PARTIES, which may request the contracting parties concerned to consult with them if the change seems likely to jeopardize or delay unduly the formation of the customs union or of the free-trade area.

No country wants to go through all that unnecessarily, which is why interim agreements are now rare  And even if the UK wanted to take this complicated route, there’s little chance the EU would agree.

In fact, there is no WTO definition of an interim agreement. We can only surmise some of what it might contain from the conditions in Article 24 and other procedures.

Would an interim agreement ever be notified? Perhaps only if the interim arrangement blatantly violated something like the “substantially all the trade” rule. For example, if the countries involved could agree on free trade across the board, except in agriculture and they wanted to implement the rest while they negotiated agriculture. Perhaps. Normally they would craft some way to get round that.

Frontier zone is only 15 km
Frontier myth: border traffic between adjacent territories

Contiguous territories and frontier traffic?Back to top

Finally, the reference to “contiguous” territories in paragraph 3 of Article 24 only applies to Trieste and is now out of date.

Another provision on frontier traffic between adjacent countries would only apply up to 15km inside the border and therefore not to the whole of Ireland or Northern Ireland, Bartels says: “It applies to products produced 15km either side. You also get grazing cattle as ‘frontier traffic’ (provided they come home). In short, this is no solution to the NI border question.”

Where does this come from? According to historical GATT documents (see page 7 of this, or the version on the WTO website), this was the intention behind the original drafting proposal from the US back in the 1940s. Although the drafting committee felt this should not be interpreted too strictly, the reference to 15km can probably be taken as an indication of the scale intended.

Understanding these WTO provisions is not going to have a dramatic impact on Brexit. But understanding that they are red herrings could leave more time for issues that really matter.

Lorand Bartels has produced a short “basic” model UK-EU free trade agreement, here, or online here. It’s not an interim agreement, but one that would last until replaced. It would still require the UK and EU to agree, and it remains to be seen if the EU would agree to it without a Withdrawal Agreement.

Since the draft’s main purpose is to meet WTO notification requirements, it only covers tariffs (keeping all at zero), rules of origin and dispute settlement. Other issues that we would expect to see in a free trade agreement, such as standards, regulations, recognition, etc, would still have to be sorted out bilaterally between the UK and EU.

Full text: GATT Article 24 (XXIV), all amendments and clarifications, and jurisprudence, from the WTO Analytical Index, is here.

• July 3, 2019 — replaced broken links to Jacob Rees-Mogg on Daily Politics
• February 14, 2019 — removing incorrect information on interim agreements never being used; replacing it with the sidebar box on their rare use
• January 26, 2019 — adding the information that interim agreements were used under GATT pre-WTO but apparently not since then.
• January 23, 2019 — adding link to actual EU notification of enlargement to 28 members, 2013; Lorand Bartels’ model free trade agreement
• January 15–16, 2019 — adding clarification that the extract of Art.24 is from Paragraph 7; link to full text from the WTO Analytical Index; link to the March 2017 Politico article

Photocredits: All original images from Pixabay, CC0

Author: Peter Ungphakorn

I used to work at the WTO Secretariat (1996–2015), and am now an occasional freelance journalist, focusing mainly on international trade rules, agreements and institutions. (Previously, analysis for AgraEurope.) Trade β Blog is for trialling ideas on trade and any other subject, hence “β”. You can respond by using the contact form on the blog or tweeting @CoppetainPU

%d bloggers like this: