India-UK CETA: Interesting things in Services, Digital Trade, and the Security Exception
In an era of “napkin deals”, it’s nice to see some countries still doing the hard work of entering into comprehensive, concrete and legal binding trade agreements. The India and the United Kingdom are the latest to sign an agreement—a Comprehensive Economic and Trade Agreement (CETA, Indian site, UK site).
Detailed analysis of the CETA is available elsewhere. In this post I’ve just set out three areas where I spotted something of interest (to me): (1) Trade in Services, (2) Digital Trade, and (3) the Security Exception.
1. Trade in Services
First, the CETA contains what looks to be India’s most detailed Domestic Regulation Article to-date (Article 8.8), and the article actually binds India to much of what is contained in the WTO’s Joint Initiative on Services Domestic Regulation (Services Dom Reg JI) (which India had some concerns with).
The biggest aspect missing from CETA is the Services Dom Reg JI’s translation of GATS Article VI(4) into a binding obligation. GATS Article VI(4) sets up a process for the development of “necessary disciplines”, “with a view to ensuring that measures relating to qualification requirements and procedures, technical standards and licensing requirements do not constitute unnecessary barriers to trade”. Such disciplines “shall aim to ensure” that these measures are:
(a) based on objective and transparent criteria, such as competence and the ability to supply the service;
(b) not more burdensome than necessary to ensure the quality of the service;
(c) in the case of licensing procedures, not in themselves a restriction on the supply of the service.
The only disciplines to come out of the GATS Article VI process so far have been the Disciplines on Domestic Regulation in the Accountancy Sector (but these have not entered into force). While waiting for such disciplines, GATS Article VI(5) obliges WTO Members to not apply qualification, licensing, etc, measures if they would nullify or impair a specific commitment “in a manner which” does not comply with the criteria listed above and could not reasonably have been expected when the specific commitments were made.
Meanwhile, trade agreements started using an amended version of Article VI(4) that requires Members to follow (or endeavour to follow) the criteria in at least sub-paragraphs (a) and (c).1 In the Services Domestic Regulation JI this looks like:
22. If a Member adopts or maintains measures relating to the authorization for the supply of a service, the Member shall ensure that:
(a) such measures are based on objective and transparent criteria;
(b) the procedures are impartial, and that the procedures are adequate for applicants to demonstrate whether they meet the requirements, if such requirements exist;
(c) the procedures do not in themselves unjustifiably prevent the fulfilment of requirements; and
(d) such measures do not discriminate between men and women. [footnotes omitted]
India has traditionally opposed this move, including because it is seen as usurping the mandate under GATS Article VI(4) obliging Members to work on more specific disciplines. So it is not a surprise to see this position continued in CETA.
Second, as compared to India’s recent trade agreement with Australia (the ECTA),2 in CETA India has provided the UK with a slightly larger list of services sectors covered by Most-Favoured-Nation (MFN) treatment (by my count its roughly five extra sectors3).
India’s MFN commitments to the UK are also missing four sectors it committed to Australia:
“Internet and internet infrastructure services”: it may be for this one that India and the UK consider this covered by other commitments on data processing, data base and other computer related services—however these are also committed by India in its agreement with Australia in addition to these internet sectors;
“Audio-visual services (CPC 9611, 9612, 9613)”: this sector is entirely carved out of CETA’s Trade in Services Chapter so could never have been committed;4
“Insurance and insurance-related services;” and
“Banking and other financial services”.
The two financial services sectors are interesting given the UK’s obvious interest in the sector. But part of the explanation is similar to audio-visual services.
CETA carves financial services out of the Trade in Services Chapter and puts them into their own chapter. This is typical for many modern trade agreements, including for example the CPTPP. However, often those agreements then include versions of the services disciplines (including MFN) in their Financial Services Chapter to enable similar commitments to be made on financial services.
CETA’s Financial Services Chapter doesn’t contain MFN (although please correct me if I’ve missed it in my review!), so it was not possible for either India or the UK to make an MFN commitment on financial services.
Conversely, in Australia’s ECTA, the Trade in Services Chapter (including its MFN provision) applies to financial services, and more specific provisions for financial services only are set out in a Financial Services Annex to the chapter. This mean India was able to commit MFN on two financial services sectors for Australia.
In the past I’ve seen some confusion on the ‘value’ of having a separate Financial Services Chapter. Some mistakenly automatically associate this with higher ambition outcomes or more substance for financial services. Clearly that is not the case—form and structure is rarely determinative.
Third, and this is not a substantive point, but I did enjoy this paragraph in the Trade in Services Transparency Article
Article 25.2 of CETA’s Transparency Chapter is a publication obligation, requiring parties to publish regulations in advance, seek comments and the like. Article 25.7 says dispute settlement does not apply “for any matter arising under Article 25.2”. I just enjoy cross-referencing drafting, particularly where it does something slightly complicated like reversing a non-application of dispute settlement clause.
2. Digital Trade
CETA’s Digital Trade Chapter is the highest ambition digital trade chapter India has agreed in a trade agreement to-date. It covers much of the same ground as the WTO Joint Initiative on Electronic Commerce (E-Commerce JI), but doesn’t prohibit customs duties on electronic transmissions or have provisions on electronic payments or personal data protection. One notable addition as compared to the E-Commerce JI, though, is that CETA does include a prohibition on forced source code disclosure (which is already under criticism).
Also of interest is the chapter’s ‘forward review mechanism’ contained in Article 12.20:
In essence, this article means that if India enters into a trade agreement in the future that contains provisions on personal data protection, cross-border data flows, or data localisation requirements, the UK could then force India to enter consultations that are aimed at amending the CETA to include similar disciplines. So if India ever does accept these disciplines with another counterparty, it provides a way for the UK to try and maintain equivalency.
Earlier examples of similar language in trade agreements (although in the services context) include Article 7 of AANZFTA’s Trade in Services Chapter (prior to the recent upgrade) and Article 8.5 of ECTA.
A related kind of provision is Article 28(5) of the Singapore-Australia Digital Economy Agreement, which triggers an immediate expansion of its source code provision to also cover algorithms once both parties have undertaken such an obligation in other international agreements. This is more than just an obligation to negotiate, instead creating a new obligation as between Singapore and Australia once triggered. But the trigger itself is both parties entering into other trade agreements containing a relevant obligation.
3. The Security Exception
Finally, on LinkedIn Mona Paulsen has noted that CETA contains a new formulation of the security exception.
Article 28.2(1) of CETA essentially reflects the standard security exceptions from the GATT and the GATS. But Article 28.2(2) then also allows for any action a Party “considers necessary for the protection of its essential security interests, including in time of national emergency or relating to the protection of critical public infrastructure, whether publicly or privately owned, including communications, power and water infrastructure, subject to the requirement that such action is not taken in a manner which would constitute a disguised restriction on trade.”
This language is an interesting amalgam of:
‘self-judging’ security exception language from agreements such as the CPTPP (“considers necessary for the protection of its essential security interests”);
a non-exhaustive list of essential security interests with critical infrastructure protection language that, as Professor Paulsen notes, has some history in other bilateral and regional trade agreement security exceptions including RCEP (“relating to the protection of critical public infrastructure”); and
part of the chapeau requirement found in the GATT/GATS general exceptions (“subject to the requirement that such action is not taken in a manner which would constitute a disguised restriction on trade”).
The use of the chapeau language is particularly interesting as it adds in an explicit hook for evaluating the application of the security measure (even if the measure itself is ‘self-judged’ as necessary). This would be in addition to the broader good faith obligation that would attach to the use of the exception. We shall see whether others pick-up this new compromise approach.
More interesting, as Professor Paulsen's 2020 article points out, in the negotiations that led to the GATT the UK actually proposed chapeau test language for the security exception (including the disguised restriction on trade test):
It’s always interesting to see some rhymes in trade negotiation history.
Sub-paragraph (b) is more controversial given its imposition of a necessity test.
My usual disclosure that many years ago I worked for Australia and also was the legal advisor to the ECTA team in the last days of the negotiations. I obviously don’t reveal any confidences here. You should though take any spruiking for Australia with a grain of salt given my biases.
Namely, Accounting and Auditing Services (CPC 862); Veterinary Services (CPC 932); Services provided by midwives, nurses, physiotherapists and paramedical personnel (CPC 93191); Whole Sale Trade services (CPC 622); Retailing Services (CPC 631, 632, 6111, 6113, 6121). Note the full list in Annex 8C looks much larger than the list in the Australia-India agreement primarily because of how it splits out some sectors into separate rows.
A UK practice inherited from the European Union, linked to sensitivities around cultural content.