10 - The Concept Design of International Rules on State Intervention in the Economy

Summary

Regulating subsidies and state enterprises is a challenging exercise since, arguably, no other area touches more directly on the state intervention in the economy and on the various ideas and interests surrounding it. Taking stock of the acquis in regulating subsidies and state enterprises in the World Trade Organization (WTO) and in preferential trade agreements (PTAs), this chapter aims to develop a conceptual framework to help think about new rules, using the notions of principles, standards, and rules, as well as those of institutions, procedures and remedies. The chapter also delves into the process necessary to produce such rules. Using this conceptual framework, the gist of the analysis is that, to a large extent, there is no need to reinvent the wheel but to build upon the status quo under the WTO and many PTAs – which, importantly, already reflects the common agreement of many governments. In terms of process to follow to generate new rules, the importance of transparency, knowledge-generating activities and epistemic communities is duly underlined.

10.1 Motivation

Taking stock of the acquis in regulating subsidies and state enterprises in the World Trade Organization (WTO) and in preferential trade agreements (PTAs), this chapter aims to develop a conceptual framework to help thinking about new rules.

Regulating subsidies and state enterprises is a challenging exercise since, arguably, no other area touches more directly on state intervention in the economy and on the various ideas and interests surrounding it.¹ Talking about subsidies and state enterprises means referring to the various public policy objectives governments pursue (in some cases in the attempt to protect their own domestic industries) and to the role of markets. The topic of subsidies evokes the use of taxation and expenditure prerogatives to create incentives (and disincentives) for certain activities. State enterprises (broadly intended as any entity controlled by governments and operating in the market) constitute the economic arm of governments in the economy. Quite often, the two (subsidies and state enterprises) go hand in hand. Governments rely on direct participation in the economy through enterprises, and the same inevitably receive financial support from the public coffer in various forms. To a large extent, then, the issue of state enterprises is a subsidy issue. This factor alone would justify their joint analysis.

If the main goal of the chapter is to help readers in thinking afresh about the issue of how to regulate subsidies and state enterprises at the international level, this exercise does necessarily need to build up on an assessment – albeit cursory – of the status quo. Do PTAs regulate subsidies and state enterprises? If so, how? Are there common denominators among these PTAs? Are there any differences? Why? What works and what does not work in the current regulation? And, what can the ‘zeroth level’ of regulation, that is, the WTO rule book, say about subsidies and state enterprises? Does it constitute an ‘efficient’ standard? If not, why?

This exercise provides a useful diagnosis of the problem and enables us to offer a better prognosis for the ‘out-of-the-box’ thinking which aims to constitute the original contribution of the chapter. The chapter is thus organised as follows. After assessing the multilateral regulatory standard, represented by WTO rules, Section 10.3 will take stock of the chapters on subsidies and state enterprises in the PTAs currently in force. Section 10.4 invites the readers to think about the various issues in conceptual terms, using the notions of principles, standards, and rules. Sections 10.5 and 10.6 will then apply these conceptual ‘bricks’ to subsidies and state enterprises, respectively. Section 10.7 will conclude by moving from the substance of new rules to the process necessary to produce such rules.

10.2 WTO Rules on State Intervention: The (now) Inefficient Multilateral Standard

Not many learned observers will be surprised by the statement that WTO subsidy disciplines do not really work. Nor that state enterprises, especially in their more troublesome activities, are virtually unregulated at the multilateral level.

As regards subsidies, the structure of the WTO Agreement on Subsidies and Countervailing Measures (ASCM) is fairly straightforward. The ASCM defines subsidies (Article 1) and provides that, in order to be regulated, they need to be specific to certain enterprises and industries (Article 2). The regulation was then used to divide these subsidies into three categories (under Articles 8, 5, and 3, respectively) – expressly permitted, potentially actionable, and expressly prohibited (the first category is now expired).² The original version of the ASCM (under Article 6.1) also provided that, for certain more troublesome subsidies, the serious prejudice triggering actionability should have been presumed (the rule that has too expired).³ Building on the Tokyo Round Subsidy Code, the ASCM also includes important rules regulating the investigations and imposition of countervailing duties. Importantly, Members should have duly notified their subsidies to the relevant WTO Committee to enable peer control and discussion (see Articles 25 and 26 of the ASCM). Interestingly, this Committee could have set up auxiliary bodies, as appropriate (Article 24.2) – which has never happened – and could have been aided by an Expert Group (Article 24.3).⁴ Equally subject to transparency were countervailing duty measures which the ASCM strictly regulated to ensure that they could not be captured by protectionist tendencies.

The Agreement on Agriculture offered a partly different approach based on baskets and scheduling. While the idea that subsidies within a certain given cap should be permitted (together with certain non-distorting laudable subsidies) remains (see Annex II, the so-called Green Box), the idea that certain other subsidies should be prohibited (notably export subsidies) has now taken grip of the agriculture regulation too (see Article 9 and ‘Nairobi’ Ministerial Decision on ‘Export Competition’).⁵

This brief overview gives a rough idea of the regulatory scheme for subsidies introduced by the WTO. That being said, the main indicator of the inefficient level of regulation is the limited number of measures of support that have been challenged before the WTO dispute settlement system since 1995. To be sure, this system has very often been confronted with subsidy issues (at the time of writing, for example, complaints on the ASCM account for 22 per cent of all the disputes filed with the WTO dispute settlement)⁶ but, as far as domestic subsidies are concerned, complaints have been extremely rare (representing only 5 per cent of all consultations requests).⁷ More crucially, however, ‘actionable subsidies’ claims have been pushed through by complainants and accepted by the adjudicating bodies in less than five cases! The measures that have been consistently condemned, in the form of export or local content requirement subsidies, are commonly held to be protectionist and are hence strictly prohibited.

The argument that the multilateral dispute settlement mechanism has, in fact, been used in a fairly limited way can be better appreciated if one considers the massive use of public budgets to face the numerous financial and economic crises that have confronted the world since the WTO was created. It is quite simply not possible that the several instances of this ubiquitous and massive governmental support cannot be configurated as domestic subsidies distorting international trade flows. Against this scenario, the increasing use of trade remedies (and in particular countervailing duties) has been the only significant constraint on foreign subsidisation. This is understandable since trade remedies have always been resorted to and truly represent the final protection of the inner citadel of the domestic industry.

Leaving litigation aside, it is very difficult to determine whether multilateral subsidy disciplines have really constrained the granting of distorting subsidies.⁸

One further legal observation can be made in support of the claim that the current disciplines are probably under-inclusive overall and thus do not constitute a big disincentive against distorting financial support. This refers to the increasingly high legal standards that have to be met in order to win a subsidy claim. One of the reasons why there have not been many successful actionable subsidy claims in Geneva is that the ‘serious prejudice’ standard under Article 6.3 of the ASCM is very high. Similarly, with time, the Appellate Body has made the legal standard for de facto contingency in prohibited subsidies in Article 3 higher and higher.⁹ The gate is so narrow that very few will get through. Crucially, it may well be that such a narrow scope was not what the negotiating parties had originally in mind. However, this is what has come out from the prevailing judicial interpretation (Rubini 2019; Crivelli and Rubini 2020).

Equally, current WTO subsidy disciplines do not include a single iota espousing the fact that subsidies may be legitimate if they pursue, certainly under given conditions, public policy goals. Accordingly, it would be difficult to argue that WTO subsidy disciplines provide an express and clear affirmation of countries’ policy space. If this is ensured, it is done indirectly and implicitly (Rubini 2017). But the boundaries between what is legal and what is not are far from clear – and this happens at all levels, from the definition to the specificity test¹⁰ up to the determination of adverse effects. Legal certainty is elusive (Rubini 2012).

Importantly, one of the sore points of the WTO subsidy regime is the poor transparency on subsidies. This can be explained by various factors, very often linked to the nature of subsidies themselves (Collins-Williams and Wolfe 2010). The fact remains that this is an area of gross inefficiency of the system. Without information on the measures supposed to be regulated, no real governance and regulation are possible.

A final, important gloss is that this assessment of WTO subsidy rules does not intend to constitute a critique of the work of the Uruguay Round negotiators and drafters. Being largely constituted of flexible norms, within a fairly flexible institutional setting, different interpretations and different directions could have been taken. Members’ practice and WTO jurisprudence have given WTO subsidy rules their own life. However, the path that has been followed has essentially deactivated the potential of these important rules.

Moving now to state enterprises, it is pretty accurate to conclude that the current regulation is fairly limited (Mavroidis and Janow 2017). The WTO has not really managed to go beyond the non-discrimination standard, which has emerged, through dispute settlement clarifications, as the main (arguably the exclusive) applicable rule to these enterprises.¹¹ Once again, a different direction could have partly been taken if, for example, the provisions of the General Agreement on Tariffs and Trade (GATT) Article XVII¹² had each been given distinct dignity by recognising both an obligation of equal treatment and an obligation of commercial practice (Mavroidis and Janow 2017; Mavroidis and Sapir 2021).

The General Agreement on Trade in Services (GATS) simply added a non-circumvention provision for monopolies and exclusive service suppliers (Article VIII),¹³ not really changing the general picture. Individual scheduling of market access and national treatment commitments can help address some of the concerns state enterprises raise,¹⁴ but, by definition, scheduling implies an approach which is individualised and hence liable to create fragmentation and less legal certainty than a generalised approach would entail.

Moreover, since nowadays the issue with state enterprises is largely a subsidy issue, the above assessment of subsidy disciplines can be made for state enterprises too. The current rules do not manage to clearly capture the concerns raised by state enterprises operating in international trade nor to recognise the specific missions some of them might have (Li and Rubini 2021).

In conclusion, there are several indicators that the current WTO rule book on subsidies is – at the time of writing – ‘inefficient’ insofar as it does not achieve its main purpose of regulating troublesome subsidies while explicitly recognising which subsidies are legitimate.¹⁵ Equally unsatisfactory is the discipline of state enterprises which has recently emerged as one of the most troublesome areas in international trade relations.

10.3 Taking Stock of the Existing Preferential Standard: Common Ground and Unresolved Issues

The review of the PTAs currently in force bears significant results. Recent research carried out by the World Bank in the context of their ‘deep integration’ research agenda has reviewed 283 PTAs, signed between 1956 and early 2016 and all currently in force (see Rubini 2020; Rubini and Wang 2020). What has emerged is that provisions on subsidies and state enterprises are particularly common in PTAs. In particular, while subsidy chapters were present in 94 per cent of the reviewed PTAs, rules on state enterprises were present in 77 per cent. While subsidy provisions have consistently featured in PTAs across time, state enterprise disciplines were rare before 1991 and have really become frequent from 2012 onwards.

With respect to subsidies, the most common rules are those on transparency (78%), followed by those on countervailing duties (74%) and those that regulate export subsidies (55%) and subsidies distorting trade and competition (48%). Interestingly, 75% of the PTAs that include subsidy disciplines are enforceable inasmuch as they are subject to binding dispute settlement.

As far as state enterprises are concerned, the most common rules include those that prohibit anti-competitive behaviour (87%), that regulate subsidies (71.5%), and that prohibit discrimination (64%). Importantly, though both competition and subsidy disciplines are, in principle, applicable to any enterprise, state enterprises are not specifically mentioned. Rules confirming the neutrality of the parties vis-à-vis property regimes and liberalisation decisions were found in half of the PTAs with state enterprises provisions. Relatively less common were provisions committing parties to ensure that state enterprises or monopolies do not act in a manner contrary to the commitments accepted by the parties (38%) or that ask state enterprises to act in accordance with commercial considerations (29%). Importantly, only thirty-six PTAs included transparency rules and only twenty-four included corporate governance requirements.

How does all this compare with the WTO standard? If subsidy chapters are considered, the large majority is essentially in line with the WTO benchmark. WTO-plus provisions, in particular the prohibition of export subsidies to agricultural products between the parties (which often anticipated the ‘Nairobi’ Ministerial Decision), were rare. Occasionally, additional transparency obligations between the parties were provided for. Furthermore, while PTAs concluded by the United States are usually centred around CVDs only, European Union (EU) PTAs, the major player in preferential deals, are more complex. In particular, the closer the trade partner to the EU is, the closer subsidy disciplines are to internal EU state aid laws (see Rubini 2021). This is obviously the case for those countries that are in line to enter the EU. But even outside this relatively knit, and deep, European network, there are many other more recent PTAs that do arguably include the ‘key principles’ of EU state aid law and policy.¹⁶

What are these ‘common’ elements? They certainly include the definition of subsidy/state aid; what that covers (which encompasses both goods and services); the stricter regime for certain more troublesome subsidies; the highlighting of the importance of public interest goals pursued through subsidies; the introduction of exceptions for certain legitimate subsidies (which, in some cases, reproduce EU law verbatim); and the strengthening of transparency mechanisms. This comes out especially in some (though not all) so-called second-generation trade agreements concluded by the EU. See, for example, EU–Korea (2010), EU–Moldova (2014), EU–Ukraine (2014), EU–Mexico (2019), EU–Japan (2018), EU–Singapore (2019), EU–Viet Nam (2020).¹⁷ are much closer to the WTO baseline.

Against the presence of common elements in many PTAs, and with so many interesting precedents, which extend to negotiations with countries as economically, politically, and geographically diverse as the United Kingdom (UK), Viet Nam, Singapore, and Japan, one wonders whether the normative process collectively emerging reflects the development of a ‘common law’ of subsidies. Apart from the inevitable variations that can be explained given the specific circumstances of the respective trade relations, if the analysis is maintained at the level of general principles regulating subsidies, a ‘European common law’ is indeed emerging (Rubini 2023).

The provisions on state enterprises, by contrast, are more composite, at times reflecting key EU law principles (such as the protection of public services), and at times interestingly following a mixed model, based on both EU provisions and US-led provisions.¹⁸ While transparency and corporate governance provisions are limited, and binding dispute settlement is provided for in only two-thirds of the PTAs, EU PTAs often include an express recognition that state enterprises pursue public services or other legitimate public policy objectives. With time, we also notice longer and increasingly detailed provisions.

For example, the provisions of the EU–Canada PTA (2016) show various origins. While some provisions elaborate the GATT (and in particular Article XVII), clearly separating and distinguishing non-discrimination from commercial considerations, and others echo the NAFTA, some show an interesting common ground between European and Canadian sensitivities with respect to public policy goals. Thus, the PTA provides for limitations and exceptions to the application of commercial considerations that would hinder the fulfilment of public service or regional development obligations or the specific public mission assigned to state enterprises. Finally, and importantly, the chapter on state enterprises is subject to binding dispute settlement.

But for language, the EU–UK PTA (2020) clearly bears numerous resemblances to the EU state aid rules, which, for level-playing-field considerations, is fully understandable. While many provisions are extremely interesting and may provide useful guidance for law reform (the extensive chapter on tax incentives probably provides the best example), this PTA is sui generis (being the expression of a disintegration – and not integration – process).

It is time for a few conclusive remarks. What does this brief overview of PTA practice suggest? First, the content and ambition of PTAs clearly depend on the objectives pursued by the parties and on the relative negotiating powers of the same. Second, while many PTAs simply replicate the WTO standard, there are increasingly numerous examples of innovative provisions, quite often finding origin in ‘domestic’ legal systems or in other PTAs. But these are not simply ‘cut and paste’ exercises. By contrast, what can be seen are interesting changes, specifications, omissions, etc. – in a word: variations.¹⁹

The time is probably ripe for digging deep into all these PTAs and provisions and learning what they have to say about countries’ preferences on trade and other issues.

10.4 Thinking Afresh: Legal-conceptual Bricks

When thinking afresh about a new regulatory framework, it is useful to consider the very basic legal-conceptual bricks that may be used in treaties.

10.4.1 Principles, Standards, and Rules

First, one needs to think about ‘principles’, ‘standards’, and ‘rules’.

‘Principles’ express political, economic, and/or legal ideas, for example, that certain subsidies are overall so negative for the system that they should be automatically prohibited or that state and private enterprises should be treated equally or that there should be transparency on state action in the market. A good example of the use of principles can be traced back to the Tokyo Round Subsidies Code, where Article 11, paragraph 1 highlighted that:

Signatories recognize that subsidies other than export subsidies are widely used as important instruments for the promotion of social and economic policy objectives and do not intend to restrict the right of signatories to use such subsidies to achieve these and other important policy objectives which they consider desirable.

Principles elucidate points of general importance for the design of treaties. They are thus often included in preambles. While principles guide negotiations and the drafting of new disciplines (and, if becoming embedded into a system, can take the nature of ‘paradigms’), the specific legal bricks that are used to build the substantive norms of treaties are ‘standards’ and ‘rules’.

Differently from overarching principles, ‘standards’ and ‘rules’ are norms that can already guide the behaviour of operators.²⁰ This is particularly true for ‘rules’ that aim to define a specific situation with precision, with the result that operators who have to apply them are normally left without any significant doubt of what has to be done.²¹ By contrast, to be operative, ‘standards’ need to be specified into ‘rules’. In other words, while the prescriptive content of rules is determined beforehand, when the same rules are created, the content of standards is intentionally left vague so that it can be shaped when its application becomes necessary. Using economic language, while ‘rules’ are usually linked to nearly complete contracts,²² ‘standards’ are the best example of incomplete contracts. Negotiators and drafters may decide it is not desirable, or even possible, to agree on the specific content of a norm and may thus simply adopt a standard that will need to be given more specific content when it needs to be applied in practice.²³

Continuing with the use of economic concepts, and using a principal-agent framework, while with rules principals determine (at least) the key elements of the norm and leave to agents their simple implementation, with standards principals essentially delegate the normative act to agents.²⁴ One interesting form of delegation, which may involve more technical issues, would involve continuous or regular expert discussions and deliberations to develop or codify best practices.

It should be noted that, in practical terms, the agent may always be required to determine the content of the norm, the difference between rules and standards, thus becoming a matter of degree of definition and delegation. If the essential content of the norm/behaviour is already defined by the principal, we have a rule; if not, we have a standard. The use of a few examples taken from the WTO ASCM can illuminate this distinction.

Article 1.1(a)(2) of the ASCM reads that, for a subsidy to be deemed to exist, there must be a ‘benefit’. This is a ‘standard’ which needs to be specified each and every time it has to be applied. Hence, the case law made it clear that the term ‘benefit’ implies a ‘comparison’ and that the best, the only logical comparator is the ‘market’. This does not solve any issue since it is then necessary to identify the precise ‘market benchmarks’ applicable to the financial contribution and the circumstances at hand.

By contrast, Article 14 of the ASCM, on the ‘calculation of the benefit’, already outlines various ‘rules’ to the interpreter and, in particular, various ‘market benchmarks’ to use. Thus, for example, a provision of equity capital shall not be considered as conferring a benefit if the investment is in line with the usual investment practice of private investors (Article 14 (a) of the ASCM). This reference is certainly more precise than a general indication that there must be a benefit. That said, it is clear that those delegated to apply the rule will have to carry out an assessment. A more precise rule could be found in the now-expired Article 8.2 (a). Assistance for research activities was permitted if it covered no more than 75 per cent of the costs of industrial research or 50 per cent of the costs of pre-competitive development actions. Footnotes specified the meaning of industrial research and pre-competitive development actions.

10.4.2 Institutions, Procedures, and Remedies

Importantly, norms, whether they are principles, standards, or rules, do not live in a vacuum. They are made operational within ‘institutional frameworks’, following certain ‘procedures’. And, crucially, their breach ensues certain ‘remedies’. Thinking about new rules on subsidies and state enterprises at the multilateral level will thus require considering this important level of regulation too. One cannot think about reforming the rules governing subsidisation or state enterprises’ action without also considering where, how, and by whom these rules should be applied, enforced, monitored, and reviewed.

For example, subsidies are ambivalent measures where design is key to achieving the stated objectives while keeping international spillovers at bay. Their success often depends on tackling new challenges, which in turn is based on the proper assessment of policies and their effects. The same holds true for ensuring that they are designed in such a way that across-the-border distortions are kept to the minimum. The ‘institutional’ structure of subsidy governance should thus duly reflect this fundamental need for information and assessment, by encompassing bodies of new knowledge, on how best practices and suggestions for law reform are generated. Crucially, serious thinking should be directed at how information should be channelled into this process, how assessment and deliberation should take place, and how any output should be made available. These are essential ‘procedural’ issues.

10.5 New Rules for Subsidies

Subsidies may distort international trade and competition – hence we regulate them internationally – but they may also be the first best to tackle market failures and pursue other important societal objectives.²⁵ Unfortunately, one does not exclude the other, and, at least theoretically, the final worth of a subsidy should be the result of a balancing of the negatives with the positives. This balancing involves difficult trade-offs between different interests, a process which may be certainly more difficult if it has to be performed across different jurisdictions (Rubini 2017). That being said, while more complex balancing scenarios necessarily require the existence of a very deep level of economic and political integration (see the EU), what can certainly be done in any international discipline regulating subsidies is to identify more clear-cut welfare situations and to regulate them.

It is also known that the design of subsidies is key, not only to achieve public policy goals, but also to avoid negative spillovers. It is, therefore, clear that any new rule should take into due consideration significant design elements.

Prohibitions should focus only on those subsidies that are more likely to produce a more negative impact on trade or that include discriminatory or protectionist elements. Other subsidies, which do not necessarily involve a negative impact on trade, should be objected to only if a requisite legal standard is satisfied. Crucially, however, if disciplines aim to be meaningful, this standard cannot be exceedingly difficult to satisfy. Governments can also decide that certain subsidies, pursuing laudable objectives and satisfying certain conditions of neutrality and non-discrimination, should be accepted. Variations on these themes are certainly possible with further sub-categories between the various types of subsidies depending on their expected distortedness (or otherwise) and their design. For example, governments can also decide to follow another approach, perhaps with those subsidies which, though troublesome, are difficult to remove from the statute book immediately or should deserve protection. They can negotiate caps on them and envisage only progressive reductions/phase-outs.

The reader will have surely realised by now that these suggestions do not depart dramatically from some of the main features of the trade law acquis resulting from the multilateral rule book (especially as it was originally devised) and many PTAs. To a large extent, there is no need to reinvent the wheel but to build up on the status quo – which, importantly, already reflects the common agreement of many governments.

One area where rules at the multilateral level have shown a high degree of deficiency is remedies. Following the GATT/WTO law mantra that remedies can only be prospective, and in derogation from the prevailing rule in general international law,²⁶ remedial action for subsidies granted in contravention of trade laws has often lacked teeth.²⁷ Clearly, the strength – or lack of strength – of remedies is a key consideration that may be connected to the degree of ambition of a legal system. Accepting that trade law remedies are retrospective would be considered a Copernican revolution in trade circles. This does not exclude, however, that, against the prevailing standard of public international law, toothless remedies are a legal and political oddity.

Arguably, the area where more progress should be made – and could be more difficult to achieve – is transparency. Subsidies are complex, and there inherently is a great degree of uncertainty in their definition, classification, and evaluation (see Collins-Williams and Wolfe 2010; Hoekman and Nelson 2000a, 2000b). Furthermore, and leaving aside all these inherent difficulties, the issue is that the incentives are simply not there for being transparent. The only real solution to the question of transparency would be to go beyond the simple governmental channel. Data and information on public expenditure at the disposal of all international organisations (such as the International Monetary Fund (IMF), World Bank, International Energy Agency, and Organisation for Economic Co-operation and Development (OECD)) should be linked and harmonised if necessary. Moreover, any interested party – including individuals, companies, and trade associations – should be put in a position to inform the WTO of relevant subsidy measures. Any information should, of course, be subject to verification and comments by the relevant government; see Casier et al. (2014).

This liberation of information should be coupled with a suitable institutional framework where important knowledge-generating activities should take place in addition to the traditional peer-review function. In other words, the first best institutional governance of subsidies cannot accept ignorance of measures and their effects, due to the inherently difficult nature of subsidies and the opaque stance of governments. Knowledge on subsidies should be considered a ‘public good’, for the benefit of all. Not only may this help in reducing distorting practices, to the benefit of all, but it may also generate positive spillovers for the domestic governance of subsidies, which should be particularly welcome in an era of increasing budgetary constraints and vital common challenges like the fight against climate change.

All these substantive and governance principles can then be translated into ‘rules’ or ‘standards’. The choice of one or the other should mainly be dependent on design considerations linked to the objective and nature of the activity that has to be carried out.

10.6 New Rules for State Enterprises

There is obviously no need to reiterate that the most prevailing principle currently applying to state enterprises – that of non-discrimination – should feature prominently in any new disciplines. But prospective disciplines on state enterprises cannot certainly be centred around non-discrimination norms (be they standards or rules) only.

On the one hand, the issue is not just to preclude possible unjustified differential treatment between domestic and imported products, services, and players. What any discipline should entrench is a more general principle of ‘competitive neutrality’, whereby, in principle, governments commit not to favour the public sector or state enterprises to the detriment of the private sector and private enterprises (OECD 2021; Piernaz-Lopez 2023).

One key expression of this ‘competitive neutrality’ would be the principle that state enterprises should conduct themselves ‘according to prevailing commercial practices’ and/or to subject them expressly to competition laws.²⁸ What is, in particular, necessary is a clear separation of this principle (and emancipation from) the principle of non-discrimination. This would be overturning the WTO Appellate Body’s jurisprudence, and aligning future disciplines to what most modern PTAs have done already (see Chapter 17 of the CPTTP). Benchmarking state enterprises’ actions to commercial practice would, by necessity, avoid distortions to trade and competition. More specific rules and standards can then be introduced that are informed by this principle.

Compliance with ‘competitive neutrality’ and ‘commercial considerations’ can also be ensured by inserting a reference to prevailing ‘corporate governance’ principles which may, for example, introduce separation of functions and organs, accounting requirements, and transparency principles (OECD 2015).

By the same token, should the use of state enterprises be necessary to pursue legitimate public policy goals, such as the performance of public service obligations, this should be properly and expressly recognised in treaty language. Many EU PTAs represent a good model in this respect.

At the same time, while the market needs should be properly balanced with public needs, it would go too far to specifically require governments to liberalise sectors or privatise their enterprises. Future disciplines should thus include a principle of neutrality with respect to the organisation of the economy and the regimes of property. These are truly sovereign decisions that should be left to the preferences of each society and, crucially, are not needed in order to ensure that fundamental tenets of ‘competitive neutrality’ and ‘commercial action’ are safeguarded. Of course, this does not mean that we cannot envisage state enterprises disciplines which provide for liberalisation or privatisation processes. This is certainly one option but has to be placed in the context of a deeper integration contract that also pursues other distinct economic policy objectives (this is, for example, clear in the many association agreements that the EU concluded with would-be Members).

As hinted at, transparency is of the essence and largely state enterprises’ concerns are raised by the lack of transparency, in particular with respect to their decision-making and state financing. Requirements enabling more transparency with respect to these aspects would go very far in making the previous principle credible and also in facilitating any scrutiny of subsidy disciplines.

10.7 How to Create New Rules: Fostering an ‘epistemic community’ and Creating Knowledge

At the time of writing, it is extremely difficult – even audacious! – to speculate about what political factors could trigger countries to convene, spend part of their political capital, and amend the laws. Unlike in the Tokyo or Uruguay Rounds, there is no ‘K group’ anymore (Odell 2015; Koulen 2016). More fundamentally, however, the current geopolitical climate seems to be averse to any grand collective negotiating effort.²⁹

Even assuming a real appetite for negotiating and changing the law did exist, the very preliminary step of any law-reform process would be to create the conditions for the creation of an ‘epistemic community’ or ‘community of practice’ that develops common understandings and, through the creation of a process of mutual trust, creates the conditions for deliberation and negotiations (Adler 2005; Koulen 2016). Along the same vein, the gathering of expertise must be based on data, and good data. This may require linking up various sources of knowledge in competition agencies and international organisations with expertise in subsidies and state enterprises (Hoekman and Nelson 2020a, 2020b).³⁰

One interesting signal in this respect has been sent from the secretariats of the four key international organisations with an interest in subsidies (and state enterprises) with the joint publication of a momentous and most interesting report in the spring of 2022 (IMF, OECD, World Bank, WTO 2022) and, even more promising, the launch of a ‘Subsidy Platform’ making information on subsidies publicly available in one website (www.subsidydata.org) in May 2023. A ray of hope in gloomy times – where confrontation, subsidy wars, and ‘new’ massive industrial policies are on the rise.