I. INTRODUCTION

The negotiation for the Trans-Pacific Partnership (TPP) Agreement, a mega regional trade agreement among 12 countries, was concluded on 5 October 2015.1 One of the new features of this regional trade agreement is that it contains a stand-alone chapter on state-owned enterprises (SOEs). Some of the rules contained therein are not completely new, as it draws upon related disciplines that already exist. However, there had never been a separate, integrated discipline specifically covering SOEs before. In particular, the restriction on what is called “non-commercial assistance” is a significant new addition to what the current trade agreements provide. The TPP SOE discipline therefore sets a new standard, a minimum floor, which will affect other negotiations such as TISA, TTIP and RCEP that are still on-going. Since many prominent SOEs are engaged in the services sector, its ramifications for services trade rules are potentially great. So far, there are no comprehensive discipline governing subsidies in services. Given such significance, it is worthwhile to analyze the substantive rules of the TPP SOE discipline and understand the background from which they were born.

Three distinct sources of influence on the TPP SOE discipline can be traced. First of all, the TPP SOE discipline integrates and further strengthens relevant rules in the existing trade regimes such as the WTO and bilateral FTAs. The TPP therefore has a consolidating role. The main multilateral instrument on SOEs in the WTO is the GATT Article XVII on state trading enterprises (STEs), which was inspired by the “fear that some government-sanctioned monopolies might play fast and loose by manipulating markets” (Hafbauer and Cimino-Isaacs, 2015: 686). However, the article applies only to STEs that are monopolies or those that have special rights and privileges, and are limited to trade in goods. The article is rarely invoked these days.2

Since NAFTA, many bilateral trade agreements the US has pursued typically contain provisions regarding SOEs under competition chapters. Here again, main concern is about those state controlled entities that have special privileges and designated monopolies rather than SOEs in general. In the NAFTA, the SOE provision has been specifically included to address state dominated energy and telecommunication sectors in Mexico. Mexico had not liberalized these sectors under the NAFTA, and the US sought to discipline possible anti-competitive actions of state enterprises through the competition chapter (Yun, 2007). In the KORUS FTA, the concern with state enterprises were minimal and only non-discrimination obligation was introduced with respect to state enterprises. The pinnacle of SOE discipline is embodied in the very one-sided US-Singapore FTA, where Singapore’s state enterprises are subject to extensive transparency rules.3 The TPP SOE discipline draws upon all these existing regimes, building on them by extending the scope of coverage and sometimes “re-organizing” existing obligations to overcome what is seen to be shortcomings of the current regime in dealing with past trade disputes.

Second, as described above, it has broken the tradition of placing SOE rules under the competition chapter, and set up a stand-alone chapter on SOEs and designated monopolies. In this, it is firmly based on the principle of what is known as “competitive neutrality.” The competitive neutrality concept provides an explanation as to why SOEs may behave differently from private firms, and offers a justification for setting up a separate regulatory mechanism apart from competition law.

No doubt, frustration with increasingly aggressive use of SOEs for industrial policy by China, allegedly propped up with subsidies, has probably played a big role in this attempt to create an international regime on SOEs. Whether or not China is a member of the TPP, the TPP SOE regime will significantly pressure China, as the new SOE regime is set to become the model for developing international trade regime regulating commercial activities of SOEs. In particular, disagreements between the US and China on trade and other legal issues seem to have influenced the shaping of some of the specific rules in the TPP SOE chapter.

In the following, this paper analyses how these factors may have influenced the shaping of specific rules in the TPP SOE regime, and discusses implications of the result for other on-going trade agreements, especially the Trade in Services Agreement (TISA). Section II goes through the main elements of the SOE regime, simultaneously interpreting or analyzing them in light of existing international rules and trade disputes. Section III discusses at length on how the competitive neutrality principle has been incorporated into the TPP SOE regime. Section IV draws implications for the TISA negotiations, with the aid of analysis undertaken in preceding sections, and Section V concludes. At the time of this writing, the agreement had not yet entered into force. The text released after the conclusion of negotiations is still under legal review, and the current analysis is based on the text made public on 5 October 2015.

1)The participating countries are Australia, Brunei Darussalam, Canada, Chile, Japan, Malaysia, Mexico, New Zealand, Peru, Singapore, US, and Vietnam.

2)For earlier literature on state trading, see Cottier, Mavroidis and Schafer (1998).

3)This one-sidedness has been rectified by the TPP Agreement. “The US-Singapore Letter Exchange on SOE Transparency” which confirms that Singapore is deemed to have complied with its competition chapter obligations under the US-Singapore FTA if it complies with the TPP SOE Chapter (which applies to all members equally), is fully incorporated into the TPP Agreement.

II. MAIN ELEMENTS OF THE TPP SOE REGIME

Chapter 17 of the TPP Agreement on “State Owned Enterprises and Designated Monopolies” (SOE Chapter, hereafter) consists of 15 articles and six Annexes, with a separate Annex IV, containing a list of non-confirming activities by country. The main purpose of the SOE chapter is to ensure that special benefits bestowed upon SOEs or designated monopolies (DM, hereafter), do not breach market access promised in the other chapters of the agreement, and to fundamentally limit such benefits to SOEs that adversely affect trade and investment interests of contracting parties. The following section discusses the main obligations of the TPP SOE regime. The focus here is on SOEs rather than designated monopolies, although designated monopolies include state monopolies, which themselves may be SOEs as defined in the TPP.

1. The Scope

TPP SOE rules are applied to the “activities of state-owned enterprises and designated monopolies of a Party” that affect “trade or investment between Parties within the free trade area.” An SOE is defined as an enterprise that is principally engaged in commercial activities and is owned or controlled by the state. The definition is important, since it determines the scope of the agreement’s effect.

Under the TPP SOE definition, the SOE is owned by the state when the government directly owns more than 50% of the share capital. Presumably, more than 50% ownership gives the state managerial control of the SOE. The state can also control the SOE when it controls more than 50% of the voting rights through ownership interests or holds the power to appoint a majority of members of the board of directors or any other equivalent management body. The emphasis is therefore on “control” by the government rather than mere “ownership.”

There are three criteria with which to determine whether activities of SOEs are commercial. First, the activities must be meant for profit. Activities based on not-for-profit basis or on cost-recovery basis are not subject to SOE discipline. Second, the good or service is supplied in the “relevant market in quantities.” Third, the SOE is able to determine the price on its own. SOEs are still construed to be able to determine pricing, production or supply decisions when there are measures applied to the “relevant market” which are “general.” The enterprise is thus an SOE when it fulfils these two criteria of commercial activity and state ownership or control.

The discipline only applies to SOEs of significant size, with annual revenue derived from commercial activities of more than 200 million Special Drawing Rights in any one of the three previous consecutive fiscal years. The threshold is to be adjusted for inflation at three year intervals according to a formula using a composite SDR inflation rate specified in Annex 17-A (Art.17.13.5). For developing members such as Brunei Darussalam, Malaysia or Vietnam, the base threshold is 500 million SDR for five years after the entry into force of the Agreement (footnote 35).

Coming up with a fairly concise definition of state owned enterprise can said to be a great feat of the TPP SOE discipline. Given the extremely variable forms and definitions of state owned enterprises among different countries, this would not have been an easy task. Although there are some distinguishing characteristics that set SOEs apart from private enterprises, there is currently no internationally agreed definition of an SOE. In some countries they are part of a government department whereas in others, they are fully incorporated, or even listed companies, where government may only have partial ownership. In many cases, they are set up to serve some public purposes and undertake various delegated mandate of the government, while undertaking commercial activities as well (OECD, 2009: 26-27). The variation is extremely wide even among the TPP member countries.

Defining the boundary between the public and the private is a tricky matter, especially for non-market economies, as demonstrated by recent trade disputes between China and the US (see Section II.2.2) below). This difficulty can be seen in the case of defining state trading enterprises for GATT Article XVII on state trading enterprises. The negotiators failed to come up with a clear definition and to date only a “working definition” exists.4 How well the TPP SOE definition serves different types and forms of SOEs of TPP members needs to be tested over time, especially because there is a built-in uncertainty in the definition: as to which entities will constitute an SOE will shift continuously with changes in the share of commercial activity and ownership (eg, through privatization).

Nevertheless, despite the variety in the legal status or corporate form of SOEs, the TPP definition seems to capture the broad conceptual denominator of what constitutes an SOE in its two criteria; commerciality and government control. This is close to the World Bank definition, which refers to SOEs as “government owned or government controlled economic entities that generate the bulk of their revenues from selling goods and services.”5

2. Major obligations

There are three major obligations arising from the SOE Chapter: “Non-discriminatory treatment and commercial considerations (Art.17.4),” “Non-commercial Assistance (Art.17.6~8)” and “Transparency (Art.17.10).” These obligations are examined in turn.

1) Non-discriminatory Treatment and Commercial Considerations

SOEs are required to act “in accordance with commercial considerations in its purchase or sale of a good or service.” Commercial considerations mean “price, quality, availability, marketability, transportation and other terms and conditions of purchase or sale; or other factors that would normally be taken into account in the commercial decisions of a privately owned enterprise in the relevant business or industry.” Exception to this rule is provided for any public service mandate that the SOE has to fulfil. Public service mandate means government mandate pursuant to which an SOE “makes available a service, directly or indirectly, to the public…” Services here include distribution of goods and supply of general infrastructure services. The SOE however have to fulfil its public service mandate in ways that are “not inconsistent” with its non-discrimination obligation with respect to enterprises investing in the “relevant market” in its territory.

The non-discriminatory treatment obligation prevents discrimination on the basis of nationality when the SOE sells and buys goods and services. The SOE must provide non-discrimination to three categories: (1) goods and services of another Party, (2) goods and services supplied by an enterprise that is a covered investment in the relevant market, and to (3) enterprises that are covered investments in the “relevant market.” The language used is treatment “no less favorable than” those accorded to like goods, like services, and enterprises in the relevant market “of the Party, of any other Party, or of any non-Party.” The obligation therefore includes both national treatment and most favored nation treatment.

The “non-discriminatory treatment and commercial considerations” article is the most familiar, as it is substantively not so different from what existing trade agreements provide. However, the way the relationship between non-discrimination and commercial consideration is constructed is different from, for example, the GATT Article XVII on state trading enterprise (STE), and has an effect of over-turning a ruling of the WTO Appellate Body. GATT Article XVII.1(b) reads “The provisions of subparagraph (a) of this paragraph shall be understood to require that such enterprises shall, …, make any such purchases or sales solely in accordance with commercial considerations,” where “subparagraph (a)” refers to the non-discrimination obligation. In Canada-Wheat Exports and Grain Imports (2004), the Appellate Body ruled that the two paragraphs must not be read separately. That is, unless the STE engages in discriminatory conduct, it is not relevant to inquire if the STE is acting commercially, subordinating the commercial consideration obligation to non-discrimination obligation.6 No such subordination exists in the TPP SOE discipline. The non-discrimination clause and commercial consideration clause are spelled out separately and independently, with commercial consideration being set out before non-discrimination.

For designated monopolies (which can simultaneously be an SOE), there is an added obligation preventing DMs from acting anti-competitively. More specifically, designated monopolies should not abuse its monopoly position in the monopolized market to engage in anticompetitive practices in the non-monopolized markets through dealings with their “parent, subsidiaries, or other entities the Party or the designated monopoly” owns.

This indicates that although the SOE chapter has been separated from the Chapter on “Competition,” the rules of non-discrimination and commercial considerations are to be clearly understood in terms of competition principles, rather than merely in terms of “market access.” That is, the purpose of the SOE chapter is partly to capture any anti-competitive behavior that cannot be dealt with through the competition chapter rather than creating any new market liberalizing concessions or setting up a comprehensive code of conduct for SOEs. This may have been further necessitated by the fact that, perhaps in consideration for developing country members, the TPP competition chapter is quite loosely framed, exempt from dispute settlement, and tolerates domestic competition laws which may provide exemptions for SOEs.

2) Non-commercial Assistance

The TPP SOE discipline restricts provision of subsidies that are specific to SOEs. Such SOE specific subsidies are referred to as non-commercial assistance (NCA, hereafter). It means assistance provided to SOEs “by virtue of” that SOE’s government ownership or control. Four criteria are used to determine whether the assistance is “by virtue of” government ownership or control:

i) the assistance is limited only to SOEs

ii) the assistance is predominantly used by SOEs

iii) a disproportionately large amount of assistance is provided to SOEs

iv) SOEs are favored in the provision of assistance.

Assistance means “direct transfer of funds or potential direct transfers of funds or liabilities,” such as:

i) grants or debt forgiveness, loans or loan guarantees or other types of financing on terms more favorable than those commercially available to that enterprise;

ii) equity capital inconsistent with the usual investment practice of private investors;

iii) goods or services other than general infrastructure on terms more favorable than those commercially available to that enterprise.

The NCA obligation prevents TPP member governments, state enterprises or SOEs from providing NCA to SOEs that cause “adverse effect” or “injury” to trade and investment interests to other TPP member countries. Causality between NCA and these two negative effects need to be “demonstrated.”

Adverse effect means causing displacement or impediment from the market due to NCA. More specifically, such displacement or impediment are said to occur if the market share of the NCA receiving SOE significantly increases, stays constant while it would have declined without the NCA, or falls at a significantly slower rate than it would have without the NCA. With respect to price, adverse effect arises if there is “significant price undercutting” or “significant price suppression, price depression or lost sales.” Price comparisons should be made at the “same level of trade and at comparable times.” Factors affecting prices should be accounted for, and if a direct comparison is difficult, “some other reasonable basis” such as unit values can be used for comparison. Three market locations in which adverse effect can arise are identified:

i) the NCA giving country’s own domestic market, in the case of goods;

ii) the markets of another TPP member for both goods and services;

iii) the markets of a non-TPP member countries, in the case of goods.7

There are two important exceptions to adverse effect obligation. First, it does not apply to domestic services. Second, it does not apply to NCA provided “pursuant to a law enacted or contractual obligations undertaken” before the signing of, or within three years after the signing of the TPP Agreement.

With respect to injury, the obligation applies only to the Party (ie, only to governments, and not to state enterprises or SOEs) as providers of NCA, and to NCA given specifically to SOEs that are covered investments in another TPP member country. Injury is said to occur when the NCA is provided to the production and sales of an SOE that is covered investments in another Party, where a like good is produced and sold (Art.17.6.4). Injury means “material injury to a domestic industry, threat of material injury to a domestic industry or material retardation of the establishment of such an industry” (Art.17.8.1). Determination of injury is to be made comprehensively, based on “positive evidence,” with respect to for example, volume of production, the effect on prices, and the effect on the production of domestic industry.

Non-commercial assistance obligation is a new addition to SOE rules that cannot be found in earlier trade agreements containing SOE related provisions. Although this obligation is similar to actionable subsidies in the WTO Agreement on Subsidies and Countervailing Measures (SCM Agreement, hereafter), there are significant departures on two accounts. First, unlike the SCM Agreement which regulates subsidies provided by the government or public bodies to mostly private enterprises, NCA is specifically about subsidies provided both by the government and state enterprises or SOEs, to only SOEs.8 Second, unlike the SCM Agreement which only applies to goods, the NCA obligation is extended to services and investment.

Further, in contrast to the SCM Agreement which specifies three criteria, existence of financial contribution, economic benefit, and specificity, the TPP does not require the examination of whether any economic benefit has been actually conferred. The presumption is inherent in the definition of “non-commercial assistance,” which assumes that the assistance given to SOEs means advantages that are not commercially available in the market.

Otherwise, the NCA obligation of the TPP SOE regime adopts many concepts, terminology and criteria used in the WTO SCM Agreement. For example, the four criteria to measure specificity of assistance to SOEs are similar to the criteria to determine specificity of subsidy in the WTO SCM Agreement.9 The concepts of adverse effect and injury are also borrowed from the WTO SCM Agreement, although they are constructed differently in the TPP. In the TPP SOE Chapter, adverse effect and injury are provided in separate paragraphs, with different scope of application. In the case of adverse effect, two different modes of supplying service by the NCA receiving SOEs to a foreign market are identified:

i) the supply of a service by the SOE from the NCA giving country to another TPP Party (ie, cross border supply);

ii) the supply of a service in the territory of another TPP country through an enterprise that is a covered investment in that TPP country or a third TPP country.

What the second mode of supply exactly means is not obvious, and would need some clarification. According to the current wording, the service is supplied not directly by the NCA receiving SOE, but through an “enterprise” that is a covered investment. According to the TPP Agreement definitions, an enterprise can either be privately or governmentally owned, and in the case of cross-border supply of services, includes branches. It is difficult to envision what exact form of supply the second mode would entail. Does it refer to arm’s length transaction between the NCA receiving SOE and the “enterprise?” Does it involve equity participation of the SOE in the enterprise, or merely setting up a branch? What kind of relationship between the two entities would allow benefits the SOE gets from the NCA to pass-through to the “enterprise?”

Whatever the answer to these questions may be, the presumption here is that the adverse effect can be “indirectly” attributed to NCA provided to an SOE even if the service is actually delivered by another entity (which can be a private enterprise) that did not directly receive the subsidy, without a pass-through analysis.10 The NCA obligation also covers “indirect” NCA on the giving side. Indirect provision occurs when the government “entrusts or directs an enterprise that is not a state owned enterprise” to provide subsidies to SOEs (footnote 18). Therefore, the NCA obligation is extended to private entities involved with NCA receiving SOE.

Matters are simpler for injury, which only applies to commercial presence, that is, when SOEs that are themselves covered investments in another TPP country and to the production and sale of a good by that investing SOE. In the WTO SCM Agreement, “injury to the domestic industry” constitutes one of the categories of adverse effect, along with “nullification or impairment of benefits accruing directly or indirectly to other members under GATT 1994…” and “serious prejudice to the interests of another member.” Of the serious prejudice, the TPP SOE only incorporates displacement and impediment from the market, and not the four others which are currently ineffective.11

Specification of SOEs as providers of assistance to other SOEs has been clearly motivated by recent trade disputes between the US and China in which the US was unable to convince the Appellate Body that a Chinese SOE was a “public body” in the sense of the SCM Agreement Article 1.1a(1). In US-Anti-Dumping and Countervailing Duties (China), the Appellate Body reversed the Panel’s finding that the term public body means “any entity controlled by a government” and instead found that it covers only those entities that possess, exercise or are vested with governmental authority: “…the mere fact that a government is the majority shareholder of an entity does not demonstrate that a government exercises meaningful control over the conduct of the entity, much less that the government has bestowed it with governmental authority.”12 By specifying that the government as well as SOEs and state enterprises are all subject to the obligation as providers of NCA, the TPP simply does away with the controversial issue of having to determine whether SOEs and state owned commercial banks are public bodies, as was required in the SCM Agreement. Although, of course, one can still quarrel about whether an entity under issue should be considered an SOE or not, the TPP Agreement seems to have by far the clearest definition of SOE than any other trade agreements: and as discussed above, that definition is based on ownership and control which US had argued for in the WTO disputes.

Since the TPP Agreement does not contain a separate discipline on subsidies to private enterprises, the NCA obligation creates a decidedly uneven playground against SOEs. Presumably, adverse or injurious effects of subsidies to private enterprises can be countervailable under the WTO SCM Agreement as TPP members retain their rights and obligations under the SCM Agreement. However, since the SCM Agreement does not exclude SOEs from its discipline, different conclusions are likely to be obtained in the TPP compared to that of the WTO as far as treatment of SOEs are concerned. This could result in a serious challenge to the multilateral trade regime. Further, as in the countervailing duty cases of the SCM Agreement, issues as to what constitutes non-commercial assistance and determining adverse effect or injury are expected to be contentious. The success of this discipline will therefore depend on how judiciously the panel under the dispute settlement handles these issues.

3) Transparency

The TPP SOE discipline contains extensive transparency rules. Each Party is required to provide to the other Parties or make publicly available a list of its SOEs and DMs (including an expansion of the scope of an existing monopoly and its designation) (Art.17.10.1~2). Upon request, a Party should provide following information regarding SOEs and government monopoly (ie, not private designated monopoly), provided the activities of such entities affect trade or investment between the Parties. Noticeably, in addition to general information, the information that should be provided concern governance mechanisms that gives special rights to the state or other share-holding state-related enterprises:

i) the percentage of shares that the Party, its SOEs or designated monopolies cumulatively own, and percentages of votes that they cumulatively hold;

ii) a description of any special shares or special voting rights or other rights that the Party, its SOEs or designated monopolies hold, to the extent the rights are different than the rights attached to common shares;

iii) the titles of any government officials serving as an officer or member of the entity’s board of directors;

iv) annual revenue and total assets over the most recent 3-year period for which information is available;

v) any exemptions and immunities from which the entity benefits under the law;

vi) any other information that are publicly available, such as annual financial reports and third-party audits.

A 5-year transition period is allowed for developing members of the TPP such as Brunei Darussalam, Vietnam and Malaysia with respect to these transparency rules. Upon request, detailed explanation of any policy or programs for the provision of NCA should be provided, as long as the policy or programs affect trade and investment between the Parties. The response should contain the form of the NCA, the names of government agencies, SOEs or SEs providing the NCA, names of SOEs that receive or are eligible, the legal basis and policy objectives, the amount or amount budgeted for the NCA, the amount of loans (loan guarantees, interest rates, and fees charged), price charged if any, as well as amount and characteristics of investment.

These transparency rules are quite extensive and detailed, going further than the US-Singapore FTA, which contains the most onerous transparency rule for (non-US) SOEs so far. The US has a history of emphasizing transparency in order to discipline subsidies. In the WTO DDA Rules negotiation, the US has repeatedly proposed strengthening transparency rules through for example, setting deadlines to respond to questions posed by other members. In particular, the US has strongly complained against lack of or incomplete notification by such Members as China and India, under the transparency obligation. US has even made over 300 counter notifications on Chinese subsidies by 2014 (USTR and US Department of Commerce, 2015: 9~22).

The TPP SOE transparency rules would be particularly burdensome for developing members of the TPP, although transition period has been granted for them. The rules would also be difficult to comply by those SOEs which undertake a variety of commercial and non-commercial activities where budget and accounting are not completely clear cut.

While transparency is a key element in enforcing the SOE discipline by enabling monitoring, given the disastrous experience with respect to the STE reporting mechanism, and less than satisfactory compliance under the notification obligation of the SCM, it is not clear how well this would work outside of the dispute settlement mechanism which contains stringent obligation to cooperate with information gathering (see Section II.3, below).

3. Other Rules

The SOE chapter is subject to dispute settlement, and has an interesting feature regarding the information gathering process. In the TPP SOE discipline, a greater transparency mechanism is set to operate once the dispute settlement process begins with respect to alleged claims of violation of “non-discriminatory-treatment and commercial considerations” and “non-commercial assistance” obligations. A particular information developing process is to be used as set out in Annex 17-B which works to enforce information disclosure through time limits and obligations to respond. In particular, the tribunal “should draw adverse inference from instances of non-cooperation by a disputing Party in information-gathering process” (paragraph 9) and “the tribunal shall not request additional information to complete the record where the information would support a Party’s position and the absence of that information in the record is the result of that Party’s non-cooperation in the information-gathering process.” Such detailed and forceful mechanisms for information gathering and exchange cannot be found in the earlier generation of competition chapters containing SOE provisions. Clearly, the concern is that it is difficult to elicit information from state-controlled entities, which in some countries may be exempt from disclosure rules normally applicable to commercial entities, and that in some cases, involved firms may refrain from divulging information about SOEs, if repercussions from host states are expected.

A case in point is China-Electric Payment Service (DS413, 2012), a case where China Union Pay (CUP) allegedly held exclusive market rights for electronic payment services for transactions in Renminbi. US argued that this violated China’s national treatment and market access obligations under the GATS, but the WTO panel rejected the claim that CUP represents an across-the board monopoly supplier, due to a lack of evidence. Some have complained that because of the close relationship between the CUP (which can be considered to be an SOE) and the Chinese government, multinational corporations were reluctant to actively cooperate with the USTR, fearing retaliation from the Chinese market. It is alleged that VISA, one of the main informant to the USTR, saw some of its business activities blocked during the dispute process (Kowalski et al., 2013: 36).

The USTR categorically rules out the possibility of an Investor-State Dispute being initiated on the grounds of violating the SOE Chapter (USTR, 2015). It is however possible that state enterprises, when undertaking government mandate, can be subject to Investor-State Dispute, on the grounds of violating obligations under the Investment Chapter.

Another rule concerning claims that can be made against a foreign SOE is Article 17.5 on “Courts and Administrative Bodies” which specifies that each Party “shall provide its courts with jurisdiction over civil claims against an enterprise owned or controlled through ownership interests by a foreign country based on a commercial activity carried on in its territory,” as long as such jurisdiction cover claims against domestically owned enterprises. It is made clear through a footnote that the jurisdiction need not be limited to civil claims only, indicating that criminal claims is possible. In the words of the USTR (2015), this would ensure that “SOEs cannot evade legal action regarding its commercial activity merely by claiming sovereign immunity.” Sovereign immunity, especially regarding SOEs, is far from a settled issue in the international arena, although currently dominant doctrine seems to be “restricted immunity” for commercial, private or non-governmental acts. This would again, clash with China’s view, which holds on to the “absolute immunity doctrine.”13

Challenging situations can arise when domestic courts reach decisions that conflict with rulings by international bodies in trade disputes regarding the same act of state. For example in a private anti-trust case against the vitamin C export cartel, a US district court did not accept foreign sovereign compulsion defense by Chinese enterprises, despite the amicus curie submitted by the Chinese government acknowledging that minimum export price requirements were compelled through trade associations controlled by the government. At the same time, in a WTO dispute, US had charged China with imposing export restraints of various raw materials, by coordinating minimum export price requirements through government controlled trade associations. The US presented Chinese government’s amicus curie in the vitamin C case as one of its evidence, and the Panel ruled in favor of the US.14 Instead of providing a mechanism to facilitate coordination between domestic courts and international trade bodies, Article 17.5 of the TPP SOE regime leaves the possibility wide open to similarly conflicting situations where domestic court decision on commercial activities of SOEs as private enterprises could conflict with a WTO dispute proceeding, where the status of SOEs could be ruled either public or private, as we saw in the preceding discussion. Again, there is a potential for this TPP SOE provision to undermine multilateral dispute settlement mechanism. Some scholars argue that one possible solution to this problem would be to adopt a principle of domestic courts deferring to international proceedings in a situation similar to the vitamin C case.15

4. Exceptions

The SOE regime does not apply to regulatory or supervisory activities, conduct of monetary and related credit policy and exchange rate policy by a central bank or monetary authority, as well as exercise of regulatory or supervisory authority over financial services suppliers (Art.17.2.2~3). Activities of SOEs or state enterprises to resolve a failing or failed financial institution or an “enterprise principally engaged in the supply of financial services” are also beyond the scope of the TPP SOE regime (Art.17.2.4). In addition, the rules do not apply to sovereign wealth fund and independent pensions, except for the non-commercial assistance obligation (Art.17.2.5~6).

Certain allowance is made for financial services supplied by SOEs to support export or import. For example, the “non-discriminatory treatment and commercial considerations” obligation does not apply to financial services by an SOE pursuant to a government mandate if this supports exports, imports, or foreign investment, as long as it is not intended to displace commercial financing nor offered on terms more favorable than market conditions, or if the offer is consistent with the terms and scope of the OECD’s Arrangement on Officially Supported Export Credits (Art.17.13.2).

Further, under similar conditions, if the SOE supplies such financial services through local presence in another Party, non-commercial assistance shall not be deemed to adversely affect hosting Party’s domestic financial industry (Art.17.13.3). Similarly, adverse effects are assumed not to arise when SOEs assume temporary ownership over enterprises located outside of the Party, as a result of default associated with such financial services. The condition here is that the non-commercial assistance to the temporarily owned enterprise is to recoup the SOE’s investment in accordance with restructuring or liquidation plan (Art.17.13.3).

Non-discriminatory treatment and commercial considerations or non-commercial assistance obligations do not prevent Parties from implementing temporary measures to deal with national or global economic emergency, including temporary measures with respect to SOEs for the duration of the emergency (Art.17.13.1).

The TPP rules are targeted at commercial activities of SOEs and contain explicit exceptions for SOE’s public activities. For example, SOEs are not prevented from providing goods or services exclusively to the government for the purposes of carrying out governmental functions. Further, the three major obligations (non-discriminatory treatment and commercial assistance, non-commercial assistance, and transparency) do not apply to services supplied in the exercise of governmental authority in the sense of GATS (Art.17.2.10).

More specifically, the non-discriminatory treatment and commercial considerations obligation do not apply to purchases and sales of goods or services of SOEs or designated monopolies in case of any existing non-conforming measures with regards to Investment, Cross-border Services and Financial Services chapters (Art.17.2.11). That is, these SOE obligations apply to the already liberalized services sectors only.

The commercial considerations obligation does not apply to an SOE’s purchase or sale of a good or service to fulfil any terms of its public service mandate. Similarly, the commercial considerations obligation does not apply to a designated monopoly if its purchase or sales of the monopoly good or service is to fulfil the terms of its designation. However, SOEs and designated monopolies must carry out these actions in a non-discriminatory manner (Art.17.4.1~2). Further, SOEs are exempt from TPP SOE obligations when exercising delegated governmental authorities. But again, SOEs are required to carry out these activities “in a manner that is not inconsistent” with obligations of the Agreement (Art.17.3). Examples of delegated authority include “the power to expropriate, grant licenses, approve commercial transactions, or impose quotas, fees or other charges.” Services supplied by an SOE within its own domestic market “shall be deemed to not cause adverse effects,” except in the case where that service itself is a form of non-commercial assistance (Art.17.6.4 and footnote 21).

These requirements put SOEs in a spotlight: it cannot get away pretending to be a “public body” in the commercial world, nor can it get away pretending to be purely acting on commercial considerations when carrying out public actions on behalf of the government, and must take up full contractual responsibility on either front as a commercial entity or as the government. In many ways, the TPP SOE regime is an effort to discipline the slippery, amphibian character of SOEs, which carry out both commercial and publicly mandated activities.

4)The “Understanding on the Interpretation of Article XVII of GATT 1994” defines state trading enterprises as follows: “governmental and non-governmental enterprises, including marketing boards, which have been granted exclusive or special rights or privileges, including statutory or constitutional powers, in the exercise of which they influence through their purchase or sales the level or direction of imports or exports.” This definition is closer to the definition of designated monopolies in the TPP SOE Chapter, and only applies to the purchase or sales of goods, and not to services, production or investment.

5)World Bank (1995), “Bureaucrats in Business: The Economics and Politics of Government Ownership,” requoted from OECD (2009: 26).

6)The Appellate Body opined that “…a panel inquiring whether an STE has acted solely in accordance with commercial considerations must undertake this inquiry with respect to the market(s) in which the STE is alleged to be engaging in discriminatory conduct…The disciplines of Article XVII.1 are aimed at preventing certain types of discriminatory behavior. We see no basis for interpreting that provision as imposing comprehensive competition-law type obligations on STEs, as the United States would have us do.” This ruling overturns a former Panel decision which found violation of commercial consideration obligation to be sufficient to show a violation of the whole Article XVII.1, in Korea-Various Measures on Beef. (WTO, 2012: 274-276)

7)Appendix 17-C(b) specifies that negotiation to extend this obligation to services should start within 5 years after the entry into force of the TPP Agreement.

8)It should be noted that due to Article 1.1(a)(1)(iv), the SCM Agreement includes under its discipline indirect subsidy through private entities which government and public bodies have “entrusted” or “directed” to undertake government mandates.

9)Regulation of SOE specific subsidies is not entirely new. For example, the Section 10.2 of China’s Accession Protocol to the WTO Agreement directly targets subsidies to SOEs by viewing “…as specific, if inter alia, state-owned enterprises are the predominant recipients of such subsidies or state-owned enterprises receive disproportionately large amounts of such subsidies.” This obligation is certainly disadvantageous for China where specific markets may consist predominantly of SOEs. For a critique, see Qin (2004).

10)Determining whether such “indirect subsidy” exists is not a straightforward matter. Two situations of indirect subsidization where benefit of a subsidy is transferred from the entity that is the original recipient of the subsidy to an entity that is supplying the product (or in this case services) without explicit government delegation or command can be identified: when the upstream supplier uses subsidized downstream inputs, and when the original subsidy receiving SOE has been privatized. Qin (2004: 880) notes that according to WTO jurisprudence, if the two entities are unrelated, a pass-through analysis is required, and that when an SOE is completely privatized at arm’s length and for fair market value, there is a rebuttable presumption that the benefit from the original subsidy ceases to exist. On this issue, also see Shadikodjaev (2012).

11)The four other situations where serious prejudice are “deemed to exist” are ad valorem subsidization of a product exceeding 5%, subsidies covering operating losses sustained by an industry or an enterprise, and direct forgiveness of debt.

12)WTO Analytical Index (2012: 1025-1026). For a critique of Appellate Body’s judgement, see Cartland et al. (2012). For a comprehensive discussion on this problem of “public body” with respect to China, see Ding (2014). Lee (2015) also provides an insightful commentary on how the WTO decisions may be “misinterpreting” the definition of state organs in the International Law Commission Draft Articles on Responsibility of State for Internationally Wrongful Acts. For a general analysis of China’s experience in WTO dispute settlement see Chi (2012), and more specifically regarding countervailing duty cases against China, see Ahn and Lee (2011).

13)For greater detail on this issue, see Aaken (2013).

14)There were two other similar private anti-trust cases against Chinese export cartels in the US, both reaching different conclusions regarding foreign state compulsion defense in relation to the WTO dispute than in the vitamin C case. See Martyniszyn (2012). The related WTO dispute is China-Measures Related to Exportation of Various Raw Materials (DS349, 2009-2012). Although, the Appellate Body ruled the Panel’s decision regarding minimum export price requirement to be legally ineffective, this was not based on substantive facts of the case but on legal technicalities. Panel and Appellate Body reports are available at https://www.wto.org/english/tratop_e/dispu_ e/cases_e/ds394 _e.htm. The author would like to thank an anonymous reviewer who drew the author’s attention to the private antitrust cases in the US. These cases embody complex issues touching upon sovereign immunity, extra-territorial enforcement of export cartels, and the interface between competition policy and trade policy. These are important issues of relevance to our discussion of the TPP SOE regime that need to be addressed, but dealing fully with all of these complex issues goes beyond the scope of the present paper.

15)For detailed discussions on this, see Wang (2012) and Lee (2010).

III. SOES AND COMPETITIVE NEUTRALITY

State presence in the market place has receded during the 1980s throughout the industrialized countries, with a shift to liberal market policies emphasizing privatization and deregulation. The policy shift was joined by the former Eastern Bloc and developing countries in the 1990s. Nevertheless, presence of the public sector is still significant in many countries, particularly in the less industrialized countries. In many cases, privatization of state owned enterprises (SOEs) remain partial or incomplete. This has created mixed markets in which SOEs and private enterprises are competing in the same market.

If SOEs receive preferential treatments from the government which owns them, this may give SOEs competitive advantages against private enterprises. This potentially leads to market distortions, in terms of both efficiency and equity. Such distortions which arise when SOEs and private enterprises compete are now known as “competitive neutrality (CN)” problems. At the level of international trade, such CN problems can cause trade friction. A number of SOE-related trade disputes between China and the US attest to such concerns of the international trade community.

As UNCTAD (2014) observes, many developing countries, including countries such as China and Vietnam, active use of SOEs for specific government policies, especially industrial policy in strategic sectors, afford preferential treatment to SOEs. Despite the recognition that abiding by the CN principles will bring benefits in the long run, it is not a policy priority in most developing countries wishing to rapidly catch up. This stance is increasingly being challenged by developed countries which fear that government backed SOEs will create unfair and uneven playground against their private enterprises in international transactions. Such fear is well demonstrated by the US, which has already filed 28 countervailing cases against China by the end of 2014. These cases involve industries ranging from steel, aluminum products, textiles, paper, various chemicals, wood, and non-ferrous metals to new energy technology industries among others. (USTR and US Department of Commerce, 2015: 12). Thus, incorporating the CN principle into trade negotiations has become a key issue.

Until now it has been customary to frame the rules on SOEs in trade agreements within the competition chapter. In doing so, no general principle regarding why we need special rules regarding commercial activities of SOEs has been provided. This has changed in the TPP. The TPP SOE rules adopt some of the same disciplines available in existing free trade agreements but in establishing a separate chapter on SOEs apart from the competition chapter, it seems to have adopted the competitive neutrality principle as its underlying, general “theoretical” framework, although it does not explicitly mention the term competitive neutrality in the text.

Overall, the TPP SOE obligations closely reflect what may be recommended under competitive neutrality principles. The concept of competitive neutrality originates from an Australian experiment to discipline the SOEs and has been developed most extensively in the OECD with respect to public sector reform discussions. The term “competitive neutrality” was first coined by the Hilmer Review (Hilmer et al., 1993), the document issued by an independent committee which set out principles for comprehensive competition reforms in Australia during the early 1990s. In the Australian context, competitive neutrality refers to a very specific situation where government owned public enterprises are in competition with private business. Therefore, competitive neutrality concerns the commercial activities of government enterprises rather than the government enterprise as a whole. Consequently, the Hilmer Review specifically rules out “non-business, non-profit” activities of SOEs from being subject to competitive neutrality regulations.

The Hilmer Review clearly distinguishes nature of competitive neutrality between private entities and that between public and private entities. The Review argues that in the former case, discrimination generally arises from deliberate and transparent policy actions through legislations (eg, entry restrictions, regulations regarding permissible market conduct) and therefore can be resolved through regulatory review processes, which in turn can be dealt with within the traditional realm of competition advocacy.16 The competitive neutrality between public and private enterprises however, was seen to involve distortions fundamentally arising from incomplete reforms of the government business, which may still retain bureaucratic or monopolistic characteristics in parts of its commercial activities, requiring extra regulation beyond ordinary competition law and advocacy. Thus, the writers of the Hilmer Review clearly viewed ordinary competition law, which mainly regulates the competition processes between private enterprises, to be inadequate in dealing with competitive neutrality problems arising from competition between government owned business and private business.17

The Hilmer Review specifies that CN applies to the case where competitive advantage arises purely from the status of being state owned, and not from any other firm characteristics such as size, expertise, efficiency or managerial competence. The intention of CN regulation is not to put every firm on a completely equal footing, but simply to eliminate the “undue,” additional advantages that accrue merely from the status of being government owned (eg, outright subsidies or exemptions from certain requirements, ability to borrow from the government budget, public perception that its debt would be guaranteed by the government).

It is important to note that the Hilmer Review presents neutrality in “net” terms. While government business was seen to enjoy many competitive advantages over private business, it was recognized that government business may suffer from certain disadvantages, such as the obligation to provide public service obligations, sometimes at below market price, or restrictions against diversifying into other business areas. It is argued that non-commercial activities of the public entity should be fully taken into account and compensated for, and that such disadvantages of the government business ought to be deducted from the competitive advantages they enjoy.

Competitive neutrality reforms do not require reducing the size of the public sector per se, nor require privatization or commercialization of in-house provision of goods and services, nor removing community service obligations of government business. Neither is competitive neutrality built upon the assumption that government business could not inherently be as competitive as private business. The single most important emphasis of competitive neutrality is that when publicly owned entities do choose to adopt commercial principles, then they ought to compete on the same basis as private firms, under “neutral” competitive environment where neither party has competitive advantages or disadvantages due to their ownership status.

Nevertheless, to the extent that CN problems are mainly seen to arise from incomplete reforms of SOE governance, CN regulations should ultimately be seen as moves to complement and to further facilitate the continuation of the pro-market public sector reforms of the 1980s and 1990s. The OECD support for competitive neutrality policies is also strongly based on the perceived positive benefits of pro-market public sector reforms.18 The OECD (2012c) recommends that structural separation of commercial and non-commercial activities would be the key in running the commercial activities according to market principles. It also recommends that corporatizing commercial activities would help to ensure exposure to competition, transparency and accountability. Accurate cost identification is essential so as not to under or over compensate costs of public service obligations, as this would remove any cost advantage or disadvantage that could arise due to public ownership.

Further, forcing SOEs to achieve commercial rate of return would prevent cross-subsidizing between commercial and non-commercial activities, and enable comparison of their performance to similar business activities in the same industry. Tax, finance, and regulation are particular sectors where SOEs might enjoy advantages from public ownership, and therefore special attention is needed to ensure neutrality in these fields. Public procurement is another area where competitive, non-discriminatory and high transparency should be upheld to be consistent with competitive neutrality. OECD recommends that general procurement rules should apply to SOEs, with clear selection criteria set forth in advance. Where discriminatory preferences exist, they should be made transparent in the selection criteria and be shared with potential bidders in advance.

The above OECD recommendations show that often competitive neutrality distortions arise from weak corporate governance of SOEs such as soft budget constraint, lack of accountability, and protection from bankruptcy. OECD therefore emphasizes that competitive neutrality is part and parcel of sound corporate governance mechanism for SOEs, much of which depends on its relationship with governments as their owners.

The Chapter 1 of OECD Guidelines on Corporate Governance of State Owned Enterprises is effectively a competitive neutrality requirement. It states that “The legal and regulatory framework for state-owned enterprises should ensure a level-playing field in markets where state-owned enterprises and private sector companies compete in order to avoid market distortions. The framework should build on, and be fully compatible with, the OECD Principles of Corporate Governance.” The six sub-clauses of this chapter is compatible with areas of neutrality calling for particular attention to such aspects as transparency, legal and regulatory neutrality, financial access neutrality, debt neutrality, along with separation of government function as a regulator versus player in the market (as in the newly deregulated or partially privatized network industries, or in carrying out other state functions such as industrial policy). Conflation of “ownership function” and “state function” may hinder optimal management, resulting in too passive management of the SOEs or excessive intervention, without due consideration for objective, commercial interest of the SOE. It is argued that a separation of state function and ownership function will enhance transparency in defining objectives and monitoring performance.

Even though the TPP SOE regime is not a full code of conduct (or a manual of corporate governance) for SOEs, it closely reflects the competitive neutrality principle. As set out in its definition of SOEs and scope of application, the TPP Agreement does not prevent setting up of SOEs or DMs, nor deny the value of public services these organizations provide, but aims to create a “neutral” competitive environment between commercial activities of SOEs and private enterprises in situations where it could be marred, through for example discriminatory treatment and non-commercial assistance. Further, the TPP Agreement does not condemn subsidies to SOEs per se, but seeks to redress, if at all, the harmful effects on foreign competitors, of such subsidies that are only available due to their special position as SOEs and not generally available in the market. Regulatory neutrality is also ensured by the article on administrative bodies. However, the non-discrimination obligation of the TPP should be viewed as the traditional national treatment and MFN clause to prevent discrimination based on nationality rather than discrimination on the grounds of competitive neutrality.

The TPP SOE regime explicitly outlaws cross-subsidization between designated monopoly market and non-monopoly market for designated monopolies. At the same time, the commercial consideration clause basically forces SOEs to behave just like private enterprises in their buying and selling activities. The TPP does not require separation of commercial and non-commercial activities nor force SOEs to adopt specific accounting procedures to enforce transparent cost identification. However, to meet detailed transparency obligations, SOEs basically need to have sound corporate governance enabling separate accounting of commercial and non-commercial activities, cost identification, flows of direct and indirect subsidies and so on.

It cannot be said the “net” concept has been translated very well into the TPP SOE rules. While the TPP SOE discipline applies only to commercial activities of the SOE and public functions such as “public service mandate” and regulatory or monitoring activities are exempt from SOE obligations, there are no specific compensating mechanism for disadvantages SOEs may suffer in commercial activities due to its onerous public obligations and limitations on commercial strategies it can pursue, especially if it is fundamentally difficult to completely separate the budget or accounting of the two kinds of activities. This difficulty would naturally force TPP countries to reform their SOEs to structurally separate commercial activities from public functions, so as not to become disadvantaged in competing with private enterprises in commercial markets. In this way, the new SOE discipline implicitly disadvantages SOEs and encourages pro-market public sector reforms, although it does not explicitly prevent setting up or operating SOEs per se.

Nevertheless, the TPP Agreement does make exceptions for SOE’s public good services such as public mandates, domestic and global economic crisis, financial prudence and monitoring activities. The rules also apply only to those SOEs above a certain size, measured by revenue from commercial activities. Due to a separate chapter on government procurement, government procurement, including SOE’s supply of goods and services for governmental purposes are not covered by TPP SOE obligations.

As for other neutralities relating to financial aspects such as tax neutrality, debt neutrality and direct subsidies, they are embodied in the non-commercial assistance obligation. This NCA provision is what is really new in the way of disciplining the SOEs and therefore the most significant. Of course, the subject of regulation in the TPP agreement is their “adverse effect (injury)” on private competitors (industry) of other TPP member countries, rather than elimination of such subsidies per se.

16)Competition advocacy refers to any competition promoting activities besides the enforcement of competition law that is undertaken by the competition agency. It is often difficult to directly enforce competition law against other Ministries or regulatory bodies pursuing legitimate social goals, and anti-competitive behavior created by these bodies needs to be restrained through such advocacy tools as persuasion, consultation and recommendation (Yun, 2013).

17)Australia has instituted an explicit, separate competitive neutrality regulatory system. Most other countries enforce competitive neutrality related issues through their competition laws. For more information on national practices, see OECD (2012a: 13-22).

18)The OECD defines competitive neutrality as a situation where “no entity operating in an economic market is subject to undue competitive advantage or disadvantage.” (OECD, 2012b: 15). This is a much neutralized definition, void of any reference to SOEs and failing to distinguish the subtle difference between competition between private parties versus competition between a private party and an SOE, which has been the root cause for articulating a separate competitive neutrality regulatory mechanism. Viraten and Valkam (2009), which is a rare attempt to further conceptualize competitive neutrality, likewise widens the definition to include many different market actors, or involving discrimination between actual and potential competition, or between different regions. However, OECD’s policy recommendation arising from CN analysis are basically those regarding public sector reform, so that it is safe to assume that OECD’s take on CN is fundamentally not different from what has been set out by the Hilmer Review.

IV. RELEVANCE TO SERVICES SUBSIDY NEGOTIATIONS IN OTHER FORA

Since many prominent SOEs operate in various services sectors, it is significant that the NCA obligation brings subsidies regime to services.19 No comprehensive international trade rules governing services subsidies exist to date. However, current GATS rules are not completely devoid of subsidy measures affecting trade in services. For example, GATS Article II (MFN treatment) applies to services subsidies, if measures by Members affect trade in services. Further, subsidies must be granted on a national treatment basis unless limitations have been specified in a Member’s schedule of commitments (GATS Article XVII: NT). GATS Article VIII (Monopolies and Exclusive Suppliers) and Article IX (Business Practices) also bear some relevance to subsidy insofar as they “draw attention to the role of government ownership and regulation in generating effects similar to trade-distorting subsidies” (Sauve and Soprana, 2015: 6). Not only can subsidies be subject to non-violation complaints, GATS Article XV:2 allows Members to request consultations when it considers that it is “adversely affected by a subsidy of another Member” and such requests “shall be accorded sympathetic consideration.”

In addition, services subsidy is one of the built-in agendas of the GATS. Although no time-limit has been specified, GATS Article XV calls for WTO members to enter into negotiation on extending the rules to subsidies, and for that purpose, Members “shall exchange information concerning all subsidies related to trade in services that they provide to their domestic service suppliers.” GATS Negotiating Guidelines even states that WTO members “shall aim to complete” negotiation on services subsidies prior to the conclusion of negotiations on specific commitments (WTO, 2001). The slow pace of DDA negotiations, however, has led to pessimistic views as to the possibility of concluding negotiations for new rules on services subsidies.

The greatest task in the GATS negotiation seems to be striking a proper balance between disciplining the trade distortive subsidies versus retaining policy spaces to pursue public policy objectives. As Sauve and Soprana (2015) note, there is a particular concern regarding how to treat certain kinds of subsidies, the primary purpose of which is not to create trade advantages but to pursue legitimate social goals (eg education), or to correct market failures (eg. lowering transaction or information costs when entering new markets), which nevertheless may result in adverse effects to foreign suppliers.

Indeed, some of the FTAs contain a roll back from GATS obligations, and to a considerable extent these relate to subsidies. Adlung (2015) shows that in a number of regional trade agreements more generous national-treatment exemptions for subsidies prevail, regarding limitations scheduled on a horizontal basis, and in the case of sectors, for those sectors such as education and health. Such GATS-minus provisions may have been motivated by the desire to retain “policy space” even by developed countries. Since many of the parties to these trade agreements also belong to the multilateral TISA negotiation initiated in 2013, one cannot rule out the possibility that such GATS-minus provisions would also find their way into TISA.

However, to the extent the issue of SOEs in services significantly caused frustration in the GATS DDA negotiations, it is highly likely that TISA would opt for a strengthened discipline on subsidies to services supplied by SOEs, now that TPP SOE discipline has been negotiated and can serve as a model. Except for Brunei Darussalam, Malaysia, Singapore and Viet Nam, all the members of the TPP are also negotiating in the TISA, and will have little resistance to accepting the level of TPP SOE obligations. On top of this, China, which may have the greatest reservation against any SOE discipline, has been so far deterred from entering TISA negotiations. The details of TISA are not yet officially released. The core text of April 2015 leaked by Wikileaks in 2015 does not yet contain new and enhanced disciplines, but does hold a space for subsidies. Given these indications, it is very likely that TISA will incorporate subsidy discipline specific to SOEs, and if that is indeed the case, one can expect that TPP SOE discipline would serve as a role model.

Of course, given the fundamental difference between the nature of goods trade and services trade, there would still be more general issues to be sorted out, such as definition of subsidy and its coverage. Although the TPP SOE regime modifies and reconstructs existing rules for goods to fit services subsidy, much of its non-commercial subsidy discipline borrows from the SCM agreement designed for trade in goods. Further, exempting domestic services from major obligations in respect of concerns by some TPP members may not be guaranteed in the TISA.

19)Using 50.01% government ownership as the definition, Kowalski et al. (2013) identifies 204 SOEs out of world’s 2000 largest publicly listed firms. These firms are deemed to be SOEs that are most actively engaged in international trade and investment. Overall SOE shares were highest in natural resource extraction and construction. While OECD contribution to sectoral shares was generally small, they were highest in provision of energy, tobacco manufacturing, warehousing, automobile manufacturing, and financial intermediation. BRICS countries had higher SOE shares in general, and they were highly represented in natural resources and manufacturing sectors, as well as financial intermediation, telecommunication, and air transport. (Kowalski et al., 2013: 26-27).

V. CONCLUSION

Free trade agreements are premised upon market principles, and state actors involved in commerce or international trade have always been an aberration to the system. The rules of the current trade regime such as the WTO does not seem to be very well equipped to deal with state owned enterprises undertaking commercial activities in “free market” situations. Nevertheless, the WTO has accommodated state enterprises and trade with non-market economies through, for example adopting the rules on state trading enterprises and introducing adjustments in the anti-dumping and subsidies agreements. In so doing, the WTO has taken on what Qin (2004) calls “market structure-based view” rather than an “ownership-based view.” It is argued that what matters to the world trade system is not state ownership per se, but the market structure that allows state enterprises to entertain special rights and monopolies which may impair liberalization commitments agreed upon. To the extent such special rights and monopolies are bestowed upon private parties, the resulting problems would not be fundamentally different from those emanating from state owned entities.

The competitive neutrality concept on which the TPP SOE regime is primarily based, takes on a different approach, by adopting the “ownership-based view.” Here, the state ownership itself matters. Weak corporate governance in state owned enterprises are prone to harming neutral environment for competition, creating market distortions with respect to both efficiency and equity. However, little theorizing has been done with respect to competitive neutrality. The concept has primarily developed among policy circles. Academic works supporting this proposition are difficult to come by, although Sappington and Sidak (2003) and Sokol (2009) are representative in this line of thought. Sappington and Sidak (2003) show that particular corporate governance characteristics that are said to be common among SOEs (eg. soft budget constraint, protection from bankruptcy, pursuit of multiple goals other than profit maximization such as scale expansion, statutory monopoly) make them compete more aggressively to expand scale compared to private enterprises which just maximize profits. This makes them more prone to anti-competitive behavior such as predatory pricing, raising rival’s costs and cross-subsidization between monopoly and non-monopoly sectors. They therefore recommend that for SOEs, price floor should be set higher, at levels higher than cost measures that set the floor for a profit-maximizing firm. However, Sokol (2009) argues, based on a cross-country empirical analysis, that antitrust efforts to curb predatory pricing of SOEs have been ineffective, due to difficulties in quantifying advantages and disadvantages arising from SOE status. He therefore argues that strengthening SOE corporate governance can be a substitute to competition law enforcement against SOEs. However, literature on this issue is quite limited, and more vigorous empirical analysis would be necessary before one can feel comfortable about singling out SOEs as a group presumed to be especially anti-competitive that needs special regulation. It is not even certain whether SOEs as a group generally have such a great impact on international trade and investment to be singled out. Comprehensive statistics on international activities of SOEs and their economic impact is not readily available. Kowalski et al. (2013) reports that 200 or so SOEs included in world’s top 2000 corporations take up 19% of world export value, and that 90% of these SOEs operate at least one foreign subsidiary. However, these statistics by themselves hardly show clearly the extent of internationalization of SOEs or their impact on the world economy. The international trade community should therefore be cautious about TPP style, stand-alone regime on SOEs.

For sure, rising exports and foreign investment by China, which is still largely dominated by SOEs, have made themselves visible in international trade and investment. China also does not hide its intention of relying on SOEs to be the pillars of its industrial upgrading strategy.20 It is clear that some of the TPP SOE provisions were written with China in mind, and these rules will certainly pressure China as they will serve as a model in developing international trade rules governing SOEs in other fora as well. It is true that the TPP SOE regime works against economies with high proportions of SOEs and limits policy space for governments to use SOEs as spearheads of industrial policy. Some of the articles specifically seem to have been motivated by disagreements between the US and China over their trade disputes, and trade disputes between the two necessarily touch upon such fundamental issues as to the nature of state enterprises and subsidies in China, stance regarding state immunities, and even very politically sensitive issues such as censorship. Therefore, it would be difficult to expect that international trade rule-making would be able to successfully accommodate China without further internal structural changes and reform in China.

Although at first glance the TPP SOE provisions seem onerous for China, it may in fact serve to help China carry its state sector reforms to the next level. For example, it may be more reasonable for China than China’s WTO Accession Protocol, which has been criticized on many accounts.21 The Protocol already contains high levels of obligations for SOEs that are comparable to TPP SOE obligations, including non-discrimination and commercial consideration, as well as the requirement that subsidies to Chinese SOEs will be construed to be “specific.” On the other hand, the TPP SOE discipline would apply to all members and therefore not as discriminatory as the China-specific Protocol of the WTO. It also provides various exemptions and exceptions to allow for domestic subsidies, financial prudence, and resolution of failed financial institutions and corporate bankruptcies. These exceptions have been probably incorporated to meet the needs of current members in dealing with the aftermath of recent financial crisis, but they would also serve China well in enhancing its state sector reforms and continued privatization, of which rules in the Protocol may not be so amenable because it precludes exceptions in the SCM Agreement which had been guaranteed to other developing countries and transition economies to deal with privatization and reform.

This has been a preliminary study based upon TPP SOE chapter text which is yet subject to legal scrubbing. The TPP Agreement itself has not come into effect and it is not easy to predict how each obligation contained in the SOE regime will play out, and to what extent it will impact negotiations in other fora, especially given the many built-in uncertainties. The main purpose of this paper has not been to gauge the economic effect of the SOE rules or predict the impact it will have on international trade at large. Rather, this paper attempted to evaluate various factors that influenced the formation of a stand-alone regime on SOEs which is a new feature in free trade agreements and the shaping of its specific rules, and based on such an analysis, to raise questions for future exploration. This paper identified three distinct sources of influence: the consolidating effort of the SOE regime as it draws upon various existing obligations scattered around in different trade regimes and strengthening them; the influence of trade disputes between US and China; and the role of competitive neutrality concept serving as the theoretical basis for the TPP SOE regime.

What is certain is that the TPP SOE regime has created an important benchmark on how to think about various cross-cutting issues on international trade related to SOEs, ranging from enforcement of competition policy and state immunities to SOEs to treatment of non-market economies and subsidies in services trade. In particular, we are forced to think hard again about how subsidies should be treated in international trade, and how competition policy against state owned enterprises involved in international trade can be effectively enforced. Since the theoretical foundation of competitive neutrality concept on which the TPP SOE discipline is based, is still weak, improvement of the discipline could only be achieved when these difficult issues are given further theoretical consideration. It is beyond the scope of this paper to give full treatment to the many complex issues of relevance to the SOE regime. Here, it suffices to note that addressing these issues is an urgent task, as it should facilitate bringing stability to the international trade regime, given that the TPP SOE discipline has built in uncertainties and has the potential to challenge the WTO jurisprudence.

20)Szamosszegi and Kyle (2011) provides an extensive analysis of the state sector in China. According to this study, SOEs and entities directly controlled by SOEs appear to account for more than 40% of China’s non-agricultural GDP. If contributions of indirectly controlled entities, urban collectives, and public township and village enterprises (TVEs) are included, the portion may rise to approximately 50%.

21)On this, see Qin (2004), and Yang (2000).