It's a Bird, It's a Plane: Some Remarks on the Airbus Appellate Body Report (EC and Certain Member States – Large Civil Aircraft, WT/DS316/AB/R)

Abstract

The emergence of Airbus transformed the market structure of the LCA industry into a duopoly of similar-sized full-range manufacturers. The financing of Airbus's upfront investment expenditures came in a significant proportion from public funds, which violated, in the US's opinion the SCM Agreement. While the Appellate Body follows this view of things to a large extent, it does so in a measured way: the category of per se illegal export subsidies is interpreted with a view to the manipulation of normal market conditions; the distortion on competitive conditions matters, not the increase of exports as such. Other aspects of subsidies law clarified are the relationship between effect and subsidy. They are closely related but not identical; rightly, the report operates from the premise that the SCM Agreement's regime focuses on the effect, and not on the subsidy as such, which is a manifestation of a political choice by a sovereign Member state. The Appellate Body affirms that a subsidy has a ‘life’, a shorthand for a beginning and an end: it follows that the effect of a subsidy is not bound to be permanent but is bound to terminate. It is to be regretted that the Appellate Body avoided clarifying to what extent partial privatization, hence sale of assets at market prices to private investors, ‘extinguish’ subsidies.

1. Introduction

A comprehensive analysis of this seminal Report (‘the Airbus Report’) will require more than one Ph.D. thesis: After all, we are talking of a Report that was more than nine months in the making (after a Panel had toiled on the case for five years), with three distinguished Appellate Body Members (Unterhalter, Bautista, Van den Bossche) and the Secretariat lawyers¹ authoring some 650 pages, with the operative part of the Report starting at page 270. For an institution that prides itself of its Swiss-watch precision (and distinguishes it from some panels) this is, if not a record, then at least one of the longer periods spent on an appeal. We shall not undertake, in the framework of this paper, to analyze in-depth all major issues addressed in the Report. Rather, we understand this small piece as an amuse bouche that is supposed to whet the appetite for more. Hence, we will limit ourselves to highlight certain aspects of the decision we consider particularly interesting and worthy of discussion.

It is to be reminded that the Airbus Report is part of the on-going Airbus–Boeing saga², which also led, in 2012, to the Appellate Body Report United States – Measures Affecting Trade in Large Civil Aircraft – Second Complaint (WT/DS353) that will be reported next year: the United States and the European Union, respectively, claim that the other partner is unfairly subsidizing its producer of large civil aircrafts (‘LCA’).

The emergence of Airbus transformed the market structure of the LCA industry from that of a dominant incumbent with two fringe producers with a narrow product range into a duopoly of similar-sized full-range manufacturers. The financing of Airbus's important upfront investment expenditures came in a significant proportion from public funds. The rivalry between Boeing and Airbus on the global market has resulted in new product launches which, together with competition between the LCA manufacturers for orders, significantly increased global demand, but which has shown a relative decline in the preponderant position of Boeing.

The Airbus case was initiated by the United States in late 2004;³ the Panel was established on 20 July 2005 and issued its Report some five years later on 30 June 2010, exceeding considerably the nine months allocated to the Panel phase by the DSU.⁴ The European Union appealed, and the Appellate Body Report was adopted on 1 June 2011.

At issue were more than 300 instances of alleged support measures by France, Germany, Spain, the United Kingdom, and by the EC itself over almost four decades. The measures included: the provision of financing for design and development to Airbus companies (‘launch aid’); the provision of grants and government-provided goods and services to develop, expand, and upgrade Airbus manufacturing sites for the development and production of the Airbus A380; the provision of loans on preferential terms; the assumption and forgiveness of debt resulting from launch and other large-civil-aircraft production and development financing; the provision of equity infusions and grants; the provision of research and development loans and grants in support of large-civil-aircraft development, directly for the benefit of Airbus, and any other measures involving a financial contribution to the Airbus companies. The alleged subsidies relate to the entire family of Airbus planes (A300 through the A380).

Whereas all these measures were, in the US view, subsidies pursuant to Articles 1 and 2 of the WTO Agreement on Subsidies and Countervailing Measures (‘SCM Agreement’ or ‘SCM’) causing adverse effects (Articles 5, 6 SCM), the United States further alleged that certain launch aid provided for the A340 and A380 violated the absolute prohibition of Article 3 SCM Agreement to make subsidies contingent on export performance.

2. Claims before the Appellate Body⁵

The main issues that were presented to the AB related to:

1. (i) the temporal scope and life of a subsidy in application of Article 5 SCM,

2. (ii) the interpretation of extinguishment, extraction, and pass-through of subsidies under the provisions of Articles 1, 4.7, 5, 6, and 7.8 SCM,

3. (iii) the appropriate calculation of benefit of certain subsidies (‘Launch Aids’) according to the market-investor principle in line with Articles 1 and 14(b) SCM,

4. (iv) the determination of what is a de facto export subsidy pursuant to Article 3.1 (a) SCM,

5. (v) the weight to be given to a realistic product and geographic-market definition in a case involving Article 6.3(a) and (b) SCM, and

6. (vi) the assessment of product displacement and lost sales under Article 6.3(a) and (b) SCM.

3. Issues of particular interest

I

1

The WTO's subsidy's regime is not about prohibiting subsidies per se (the lone exception, export subsidies, will be discussed immediately, Subsection II). Rather, it is merely imposing the obligation to do no harm to fellow Members through the granting of state aid.⁶ Maybe clearer than ever before, the Appellate Body explains that subsidy and caused effect are two different issues, and that the former is – from a substantive perspective⁷ – of interest to Article 5 SCM only as a cause to the latter (unless it is not a priori verboten, pursuant to Article 3 SCM).⁸

2

Several consequences arise from this. Firstly, as the SCM focuses on effect and not on cause, the EU argument that subsidies granted – and, possibly, subsidy programs started – before the entry into force of the SCM were immune from its disciplines was a non-starter.⁹

Secondly, this constellation gives occasion to rethink the issue of the ‘life of a subsidy’ and the Appellate Body takes this invitation up with almost palpable gusto. While it flatly rejects the European Union's proposition ‘that there must be “present benefit” during the reference period’,¹⁰ the proposition that a subsidy has a ‘life’ is explicitly recognized:

[It] may come to an end, either through the removal of the financial contribution and/or the expiration of the benefit … where it is so argued, a panel must assess whether there are ‘intervening events’ that occurred after the grant of the subsidy that may affect the projected value of the subsidy as determined under the ex ante analysis. Such events may be relevant to an adverse effects analysis because they may affect the link that a complaining party is seeking to establish between the subsidy and its alleged effects.¹¹

Focusing on the nexus between subsidy and its alleged effect, the Appellate Body explained:

At the time of the grant of a subsidy, the subsidy will necessarily be projected to have a finite life and to be utilized over that finite period. In order properly to assess a complaint under Article 5 that a subsidy causes adverse effects, a panel must take into account that a subsidy provided accrues and diminishes over time, and will have a finite life. The adverse effects analysis under Article 5 is distinct from the ‘benefit’ analysis under Article 1.1(b) of the SCM Agreement and there is consequently no need to re-evaluate under Article 5 the amount of the benefit conferred pursuant to Article 1.1(b). Rather, an adverse effects analysis under Article 5 must consider the trajectory of the subsidy as it was projected to materialize over a certain period at the time of the grant.¹²

In sum, a panel's analysis of the adverse effects must take into account how the subsidy has materialized over time. As part of this analysis, a panel must assess how the subsidy is affected, both by the depreciation of the subsidy that was projected ex ante and the ‘intervening events’ referred to by a party that may have occurred following its grant.¹³

Thus, one may visualize the life-span of the subsidy as such and its effects as parallel curve graphs, which are not identical but rather in a sequencing order and may be partially overlapping.

3

Of course, it is one thing to embrace the concept that a subsidy must have an end, and another one to officially determine the ‘time of death’ – and the effect on the adverse-effect analysis under Article 5 SCM. The European Union had argued that ‘sales of shares between private entities, and sales conducted in the context of partial privatizations’ had eliminated all or part of past subsidies.¹⁴ In support of this proposition, it had invoked the Appellate Body Reports in US–Lead and Bismuth II ¹⁵ and US–Countervailing Measures on Certain EC Products.¹⁶

Rightly, the Appellate Body highlights that both Reports only ‘stand for the proposition that a presumption of extinction ¹⁷ arises where there is a full privatization’ which involves sales at fair market value, at arm's length, and a subsequent full transfer of ownership and control. In Airbus, the Appellate Body had to examine to what extent a partial privatization and private-to-private sales had an equivalent effect on the lifespan of the subsidies in question. However, the three Appellate Body members tasked with DS316 could not agree on one consolidated position and chose to offer insights into their personal thinking:

1. (a) Noting that the Appellate Body has previously ruled in privatization cases that a full privatization, conducted at arm's length and for fair market value involving a complete or substantial transfer of ownership and control, ‘extinguishes’ prior subsidies, one Member is of the view that this rule does not apply to partial privatizations or to private-to-private sales.

2. (b) One Member noted that, as discussed above, the Appellate Body ruled in US–Countervailing Measures on Certain EC Products that … full privatization at arm's length and for fair market value may result in extinguishing the benefit received from the nonrecurring financial contribution bestowed upon a state-owned firm¹⁸ … This Member considers the rationale underlying the Appellate Body's case law on full privatization in the context of Part V of the SCM Agreement equally to apply in situations of partial privatization and private-to-private transactions and in the context of Part III of the SCM Agreement. However, this Member also notes that, as the Appellate Body emphasized in US–Countervailing Measures on Certain EC Products, there is ‘no inflexible rule’ that a ‘benefit’ derived from pre-privatization financial contributions expires following privatization at arm's length and for fair market value.¹⁹ Rather … ‘[i]t depends on the facts of each case’.²⁰ An important question in this context is to what extent the partial privatization or private-to-private transactions resulted in a transfer of control to new owners who paid fair market value for shares in the company.

3. (c) One Member of the Division, though affirming the general test that an extinction of benefit is to be determined upon a consideration of all relevant facts, entertains no small measure of doubt that an acquisition of shares, concluded at arm's length and for fair market value, constitutes relevant circumstances warranting the conclusion that an extinction of benefit has taken place. A subsidy granted to a recipient company contributes to the net asset value of that company. The value of that asset permits the recipient to enjoy an enhanced stream of future earnings over the life of the asset. The asset is the property of the recipient. The recipient's shareholders enjoy the right to the dividends that may be declared by the recipient and to any capital gains that arise from the enhanced earnings attributable to the recipient. When shares change hands on an arm's-length basis and for fair market value, the buyer pays a price that, in the estimation of the buyer, places a proper value on the future earnings of the recipient. Those earnings derive from all the assets of the recipient, including the benefit of any subsidy paid to the recipient. One shareholder may not accurately value or properly manage the assets of the recipient. Precisely for this reason, sales of shares take place: the buyer believes that the assets, properly managed, will be worth more over time than the price paid, and the seller believes the opposite. Time will tell who is correct. The central point is that a sale of shares, whether or not it conveys control, transfers rights in the shares to a new owner. The assets of the company, to which the shares attach, do not change at all. Nor could it be otherwise, because the buyer would then not acquire the full benefit of the bargain: the buyer would pay for an asset (the subsidy) that had in the very sales transaction been ‘extinguished’. Shares in listed companies are traded on stock exchanges with great frequency and without any fear that sales on the market diminish the underlying value of the assets owned by these companies. The changing price of listed securities reflects the different valuations that buyers and sellers place upon companies and their underlying assets. However, nothing about these trades extracts the value of any asset, including the benefit of any subsidy granted. That subsidy continues to benefit the recipient, even if the ownership of the recipient's shares changes from one day to another. Given that the Appellate Body in this case does not need to come to any final view on the issue of extinction in the context of a partial privatization or private-to-private sales, these matters do not require more definitive determination.

The Appellate Body had the luxury to not decide the issue as it was of the opinion that the Panel's analysis had drastically fallen short of what was required in the matter: arguments of the European Union with regard to the quality of the transactions in question had, in the Appellate Body's view, not received due attention, both with regard to factual findings and the care with which the arguments of the European Union had been dealt with. This reluctance of the Appellate Body is to be regretted; more clarity on this point would have been most welcome.

4

In addition, the Appellate Body rejected the European Union's proposition that the removal of cash from two Airbus partners constituted an ‘extraction’ of subsidies. Recognizing this possibility in principle, it decided that the transactions brought forward by the European Union were not ‘of such a nature, kind, and amount as to be relevant to its adverse effects analysis’. In particular, it doubted the existence of a specific relationship between subsidy and cash removal:

we do consider that, at a minimum, the European Communities was required to explain how the specific subsidies received by Dasa and CASA were reflected in the balance sheets of those companies, and how the cash removed or ‘extracted’ represented the remaining or unused value of these subsidies. The mere assertion by the European Communities, without more, that subsidies to Dasa and CASA increased the value of those companies and that therefore any cash taken out represents the subsidy or its ‘incremental value’, does not in our view satisfy the requirement of establishing a ‘causal relationship’ between the ‘cash extraction’ and the subsidy.²¹

A second criterion, namely ‘whether it had been demonstrated that the extracted cash permanently left or moved beyond the reach of the “company-shareholder unit”’ could be left to be analyzed on another day.

5

How to calculate the benefit provided by Launch finance from the four EU Member states concerned? When answering this question, the Appellate Body had the benefit of having had been exposed to numerous arguments from the parties to the dispute on how to define the relevant benchmark for the ex-ante-rate-of-return that would have been the threshold for market investors to undertake investment. For all practical purposes, the question focuses on the risk premium a rational private investor would demand in order to engage financially with regard to the different LCA product launches. Emphasizing that point, the Appellate Body's Report²² stresses that the assessment has to take place on an ex ante perspective, i.e. as if it was taken when the profit-maximizing lender and cost-averse borrower commit to the transaction. How the loan actually performed over time is irrelevant: the benefit of hindsight is denied even to the most astute private investors.

6

The central issue analyzed by the Appellate Body in regard to assessing whether the challenged financing measures confer benefit concerns can be separated into two questions: firstly, whether a single fixed project specific-risk premium for all LCA product launches (derived from a 2004 study of venture capital investments) represents an appropriate quantification or whether a more accurate estimate is that obtained from anticipated returns of a sample of risk-sharing suppliers to A 380. In its assessment, the Appellate Body identifies²³ technical problems and inconsistencies in both alternatives. Nevertheless, as both attempts to quantify the appropriate risk premium come to the conclusion that the challenged Launch Aid/Member States’ Financing measures do not fully reflect market conditions at the time, the Appellate Body accepts that the LA/MSF measures confer a benefit within the meaning of Article 1.1(b) SCM Agreement. A significant missing element is, however, the lack of a quantitative estimate of the subsidy element represented by LA/MSF measures, thus impeding a realistic and robust analysis of the actual displacement orders data when considering that Boeing was receiving subsidies over much of the same period.

II

1

To the extent they entail the damaging effects on fellow WTO members, enumerated in Article 5 SCM Agreement, subsidies stop being a purely internal method of states influencing the behavior of their economic operators and turn into a matter of interest for WTO law. If a state can show that another state's subsidy has the negative impact described in Article 5 SCM Agreement, it can react, either by attacking the subsidy in the multilateral DSU procedure, or unilaterally by offsetting the harming consequences through countervailing duties.

However, there is a mortal sin that is so wrong (according to the contracting parties to the SCM Agreement) that no concrete adverse-effect damage needs to be shown: a subsidy is, pursuant to Articles 3 and 4 SCM, per se prohibited (and thus: attackable and countervailable in a particularly swift and sharp manner) if it is (de iure or de facto) conditional upon either export performance or upon the use of domestic over-imported goods.²⁴ Article 3 SCM reads in its relevant parts:

[T]he following subsidies … shall be prohibited:

(a) subsidies contingent, in law or in fact⁴, whether solely or as one of several other conditions, upon export performance, including those illustrated in Annex I; []

Original Footnote 4 reads:

This standard is met when the facts demonstrate that the granting of a subsidy, without having been made legally contingent upon export performance, is in fact tied to actual or anticipated exportation or export earnings. The mere fact that a subsidy is granted to enterprises which export shall not for that reason alone be considered to be an export subsidy within the meaning of this provision.

2

The measures attacked by the US complaint were not contingent in law: none of the legal documents forming the legal basis for the support measures established that the benefit granted was conditional upon Airbus exports.

3

Of course, Airbus operates in a global market (see infra, Subsection III.2). If it was to be a commercial success, it had to acquire market share in the global and (provided they really exist) in regional markets. And while many details about the future champions of the global commercial airline industry were not foreseeable at the time when decisions were made, it was crystal clear that without export markets the project would, quite literally, not fly: how could, for example, the A 380 not be an economic debacle, unless it acquired significant market share in the Gulf and in the Asia-Pacific region? Despite the US suspicion that EU states subsidize no matter what, European taxpayers expected, given the amounts at stake, at least a significant part of their money back. And that was only to happen if non-EU orders would fill Airbus books.

Relying on detail-rich and not necessarily always completely clear prior Appellate Body case law, the United States could hence make a plausible claim that what the Europeans had done amounted to an a priori illegal export subsidy, and should be subjected to the most far-reaching sanctions²⁵ in WTO law: wasn't the grant of subsidies based on the expectation that Airbus would be an export success? The link between subsidy and export expectation fulfilled, in the complainant's view, the essence of what an export subsidy is: pay-outs tied to export performance. The Panel bought into that narrative: without reasonable expectation of export performance, no business case for Airbus; without a business case, no subsidies; de facto thus, the Panel opined, the subsidies were tied to export performance, contingent in fact upon exports. In doing so, it relied heavily on the fact that sometimes the documentations of the financing included specific references to exports while this was not the case in others. In the latter case, the Panel accepted that these were actionable subsidies whereas in the former the rough justice ²⁶ of the SCM's Part II applied.

4

Prior case law with regard to de facto contingency may be summarized as follows: the Appellate Body Report on Brazil–Aircraft (Article 21.5–Canada) ²⁷ clarified that a subsidy coming under the purview of the Illustrative List ²⁸ was ipso facto prohibited; a complainant need not show in detail that the general requirements of Article 3.1 SCM are met. To the contrary, if a support measure does not appear on the Illustrative List, the complainant will have to demonstrate that the subsidizing state ‘either in law or in fact’ made payment conditional on either exports or the use of domestic goods.

In Canada – Measures Affecting the Export of Civilian Aircraft (‘Canada–Aircraft’),²⁹ the Appellate Body, paying particular attention to the SCM's footnote 4, decided that mere knowledge of the beneficiary's exporting activities would not suffice for the de facto threshold to be met. Something more is required, the Appellate Body explained:

The second substantive element in footnote 4 is ‘tied to’. The ordinary meaning of ‘tied to’ confirms the linkage of ‘contingency’ with ‘conditionality’ in Article 3.1(a). Among the many meanings of the verb ‘tie’, we believe that, in this instance, because the word ‘tie’ is immediately followed by the word ‘to’ in footnote 4, the relevant ordinary meaning of ‘tie’ must be to ‘limit or restrict as to … conditions’. This element of the standard set forth in footnote 4, therefore, emphasizes that a relationship of conditionality or dependence must be demonstrated … In any given case, the facts must ‘demonstrate’ that the granting of a subsidy is tied to or contingent upon actual or anticipated exports. It does not suffice to demonstrate solely that a government granting a subsidy anticipated that exports would result.

Canada–Aircraft then cautioned:

The second sentence of footnote 4 precludes a panel from making a finding of de facto export contingency for the sole reason that the subsidy is ‘granted to enterprises which export’ … The second sentence of footnote 4 is, therefore, a specific expression of the requirement in the first sentence to demonstrate the ‘tied to’ requirement.³⁰

… De jure export contingency is demonstrated on the basis of the words of the relevant legislation, regulation or legal instrument. Proving de facto export contingency is a much more difficult task. There is no single legal document which will demonstrate, on its face, that a subsidy is ‘contingent … in fact … upon export performance’. Instead, the existence of this relationship of contingency, between the subsidy and the export performance, must be inferred from the total configuration of the facts constituting and surrounding the granting of the subsidy, none of which on its own is likely to be decisive in any given case … We note that satisfaction of the standard for determining de facto export contingency set out in footnote 4 requires proof of three different substantive elements: first, ‘the granting of a subsidy’; second, ‘is … tied to … ’; and third, ‘actual or anticipated exportation or export earnings.’³¹

5

The Airbus Report develops that body of law:³² far from breaking with prior precedents, it clearly adds new elements to the definition of an export subsidy under Article 3 SCM. Whereas the Panel had ‘equated the standard of export contingency with the reasons(s) for granting a subsidy’³³ – and probably could do so reasonably expecting to be in line with prior Appellate Body jurisprudence – the Appellate Body, this time in no uncertain terms, rejects a ‘standard that requires anticipated exportation to be the reason for the granting of the subsidy’.³⁴ It explicitly states that ‘the standard for de facto export contingency is [neither] met … by showing that anticipated exportation is the reason for granting the subsidy … [nor by showing] the subjective motivation of the granting government to promote the future export performance of the recipient’.³⁵

Rather, building on its prior jurisprudence, the Appellate Body developes the notion that a subsidy must be ‘geared to the promotion of exports’:

Because anticipated exportation ‘alone is not proof that the granting of the subsidy is tied to the anticipation of exportation’,³⁶ the legal standard for de facto export contingency under Article 3.1(a) and footnote 4 of the SCM Agreement further requires that there exists a relationship of conditionality between the granting of the subsidy and anticipated exportation. Where a subsidy is alleged to be ‘in fact tied to … anticipated exportation’, the relationship of conditionality is, unlike in the case of de jure export contingency, not expressly or by necessary implication … can be established by recourse to the following test: is the granting of the subsidy geared to induce the promotion of future export performance by the recipient? ³⁷

The new operative word for describing the special nexus between export performance and the granting of a subsidy is the term geared. The Report uses it 20 times, indicating that its drafters wanted to let the world know that a new concept has been added to pre-existing case law. What does it mean, then? The Appellate Body answers this question in para. 1045 of its Report:

Rather … the standard for de facto export contingency under Article 3.1(a) and footnote 4 of the SCM Agreement would be met when the subsidy is granted so as to provide an incentive to the recipient to export in a way that is not simply reflective of the conditions of supply and demand in the domestic and export markets undistorted by the granting of the subsidy.

In line with other tests established by the Appellate Body that are supposed to avoid the appearance to examine, horibile dictu, intent,³⁸ the Appellate Body establishes pro forma objective criteria that are, nevertheless, not free from highly subjective connotations:

The standard for determining whether the granting of a subsidy is ‘in fact tied to … the existence of de facto export contingency … ‘must be inferred from the total configuration of the facts constituting and surrounding the granting of the subsidy’, which may include the following factors: (i) the design and structure of the measure granting the subsidy; (ii) the modalities of operation set out in such a measure; and (iii) the relevant factual circumstances surrounding the granting of the subsidy that provide the context for understanding the measure's design, structure, and modalities of operation.³⁹ … [This] is an objective standard … Indeed, the conditional relationship between the granting of the subsidy and export performance must be objectively observable on the basis of such evidence in order for the subsidy to be geared to induce the promotion of future export performance by the recipient.⁴⁰

To what extent does that test establish a standard that deviates from de jure export subsidies? After all, it may be imaginable that an explicit linkage between export performance and the granting of a subsidy would not qualify as an ‘incentive to the recipient to export in a way that is not simply reflective of the conditions of supply and demand in the domestic and export markets undistorted by the granting of the subsidy’. An explicit reference in (domestic) law or in an executive act would then suffice to render a previously as such perfectly legal subsidy – subject only to the limitation to not create the consequences described in Article 5 SCM – into one of the rare instances where the WTO Agreement explicitly states that a state measure is prohibited, and thus utterly illegal, subject to the harshest procedural treatment dispended by the WTO Agreement.⁴¹ So could there be a difference between the standard applicable to de jure export subsidies and those applicable to de facto export subsidies?

Upon closer inspection – and absent laboratory conditions that the Appellate Body does not view as the appropriate test environment⁴² – it seems difficult to imagine an explicit conditionality between the granting of public funds and export performance that would not constitute an incentive to export more, both in absolute and relative terms, than would have been the case under exposure to normal market influences. That would meet the standard established in Airbus – it should be recalled: an evidentiary standard! – for de facto export subsidies. Because the difference between the WTO regimes for a priori illegal export subsidies and for normal state aid are so stark (in both theory and procedural practice), some nexus between subsidy and export performance does not suffice to declare measures of a sovereign Member a priori illegal. In the case of the de iure export subsidy, the support measure is explicitly and manifestly contingent on export performance.⁴³ Of course, the interpretation of the wording of a text allegedly establishing de iure contingency may prove difficult, and in that case the interpreter will take guidance from the Airbus Report.

To sum up this point: in line with the general rules, it behooves the complainant to show (and to the adjudicatory bodies of the WTO to establish) that the state measure in question skews market conditions in order to view support measures as per se illegal export subsidies; if it cannot be shown that a subsidy is geared to induce more exports, contrary to market conditions, then it qualifies as a regular subsidy, actionable pursuant to part III of the SCM Agreement, but not one that merits summary justice and a priori condemnation.

It is in that light that the test whether a subsidy has been geared to induce the promotion of future export performance by the recipient has to be administered. Firstly, hindsight must not come into play: the test ‘must be assessed on the basis of the information available to the granting authority at the time the subsidy is granted’.⁴⁴ The Appellate Body defines high thresholds:

[W]here relevant evidence exists, the assessment could be based on a comparison between, on the one hand, the ratio of anticipated export and domestic sales of the subsidized product that would come about in consequence of the granting of the subsidy, and, on the other hand, the situation in the absence of the subsidy. The situation in the absence of the subsidy may be understood on the basis of historical sales of the same product by the recipient in the domestic and export markets before the subsidy was granted. In the event that there are no historical data untainted by the subsidy,⁴⁵ or the subsidized product is a new product for which no historical data exists, the comparison could be made with the performance that a profit-maximizing firm would hypothetically be expected to achieve in the export and domestic markets in the absence of the subsidy. Where the evidence shows, all other things being equal, that the granting of the subsidy provides an incentive to skew anticipated sales towards exports, in comparison with the historical performance of the recipient or the hypothetical performance of a profit-maximizing firm in the absence of the subsidy, this would be an indication that the granting of the subsidy is in fact tied to anticipated exportation within the meaning of Article 3.1(a) and footnote 4 of the SCM Agreement.

6

The Appellate Body's Report in EC and Certain Member States–Large Civil Aircraft is certainly a welcome concretization of previously developed concepts. Building on previous jurisprudence, it explains for the first time in such a clear fashion that the decisive criterion for de facto export subsidies is the favoring of exports contrary to market conditions ‘over products destined for domestic consumption’.⁴⁶ Only such a restrictive interpretation avoids the absurdity that, say, most Singaporean subsidies would a priori fall into the category of prohibited subsidies, as subsidies in a small country with an export-oriented economy will more often than not be granted in expectation of export performance and in order to increase such performance.

It would seem that the Appellate Body, in order to avoid pronouncing itself too firmly on questions that were either not asked or can, in the Appellate Body's opinion, await an answer on another day, refrained from establishing truly operational standards. In fact, it seems that the Appellate Body is even keen to get academic feedback, in particular from economists, to incrementally develop this jurisprudence.

Therefore, the Appellate Body limits itself to establishing some outer parameters of a new export subsidy definition, and illustrates it with an example that shows both the practicability (in the case at hand) but also the remaining, possibly intended, imprecision.

1048. The following numerical examples illustrate when the granting of a subsidy may, or may not, be geared to induce promotion of future export performance by a recipient. Assume that a subsidy is designed to allow a recipient to increase its future production by five units. Assume further that the existing ratio of the recipient's export sales to domestic sales, at the time the subsidy is granted, is 2:3. The granting of the subsidy will not be tied to anticipated exportation if, all other things being equal, the anticipated ratio of export sales to domestic sales is not greater than the existing ratio. In other words, if, under the measure granting the subsidy, the recipient would not be expected to export more than two of the additional five units to be produced, then this is indicative of the absence of a tie. By contrast, the granting of the subsidy would be tied to anticipated exportation if, all other things equal, the recipient is expected to export at least three of the five additional units to be produced. In other words, the subsidy is designed in such a way that it is expected to skew the recipient's future sales in favour of export sales, even though the recipient may also be expected to increase its domestic sales.

Export subsidies are per se illegal, because they represent, in the mercantilist worldview underlying the GATT 1947 and to a large extent the WTO Agreements, an attack (our French colleagues still write on and teach la guerre commerciale) of a sovereign state on the (commercial) homeland of another state. Such a scenario, no doubt, would be caught by the standard offered by the Appellate Body.

The new test, however, deals with more complex and realistic constellations: imagine a scenario where the home market is saturated (yet) inexistent, or a scenario where the market is really a world market. If, in that case, a producer wants to expand, the expansion will unavoidably take place not in the domestic market. While the numerical example suggested by the Appellate Body would, at first glance, be cold comfort for such enterprises and the states subsidizing them, the test still reduces drastically the coverage of prohibited subsidies pursuant to Article 3 SCM: this is so because the determinative criterion is, according to the Appellate Body, whether the subsidy would change the export–import ratio that would normally have developed as a consequence of unfettered supply and demand.

It would seem that in many instances of state support, measures would be far too complex to be adequately addressed by the complete prohibition of Article 3 SCM. In the Airbus constellation that was arguably the case: Airbus was also an effort to be less dependent on a foreign quasi-monopoly, closely related with the military-industrial complex of that foreign state;⁴⁷ an effort to create incentives for further technological leadership, preserve and create professional jobs, and the like. In those constellations, the regular subsidies regime seems to offer a more appropriate instrument than part II of the SCM. Arguably these constellations often reflect market situations that cannot be assimilated to the Ceteris Paribus clause (referred to in paragraph 1047) ‘all other things being equal’. The performance of a profit-maximizing entrant into a global market dominated by an incumbent with significant market power is likely to be the result of strategic interactions particularly in regard to product launches, resulting in important and persistent differentials in the margins on sales to customers depending on the degree to which the product ranges of the entrant and incumbent are substitutable. Such market situations can be expected to pose problems for the test proposed since they demand rigorous definitions of the geographic and product markets at issue so as to interpret reliably the trends in domestic and export sales.

Many of the challenges humankind faces in the next century – access to freshwater, transition to a post-carbon energy environment, global warming, food security, nano-technology, fight against multi-resistant pathogenic agents, exploration of outer space and deep-water zones, development and containment of artificial intelligence, to give just a few examples – may benefit from launching aids that will inevitably benefit projects that target the world (i.e., the global market); in this context the distinction between domestic versus foreign consumption seems irrelevant. The potential competitive dangers of such state engagements are obvious, and the SCM offers in its part III et seq. appropriate and timely antidotes. The ‘nuclear option’ of part II, however, would not seem the appropriate avenue to control abuses of such subsidies.

III

The Appellate Body upheld the Panel's view that certain launch-aid and other support measures caused serious prejudice to the interests of the United States within the meaning of Article 5(c) SCM. However, it comes as a great relief that the Appellate Body rejects the position of the Panel that in a case involving Article 6.3(a) and (b) SCM it is bound by the complainant's definition of the market at stake.

1

This is what the Appellate Body had to say in this regard:

1128 … [I]n its analysis, the Panel deferred to the United States’ subsidized product allegations [that there was – contrary to what the EC had said – only one subsidized product at issue in this dispute, consisting of all models of Airbus LCA] rather than making its own independent assessment of whether all Airbus LCA should be treated as a single subsidized product. In so doing, the Panel failed to make an objective assessment of the matter, including the ‘applicability of and conformity with the relevant covered agreements’, as required under Article 1 of the DSU … [T]he Panel's failure to comply with its duties under Article 11 appears to flow directly from its erroneous interpretation of the requirements of Articles 6.3(a) and 6.3(b) of the SCM Agreement, which led it to believe that it lacked the power and was under no obligation to assess independently the ‘subsidized product’ and the relevant product market. In the absence of such a determination, the Panel did not have a proper basis for assessing whether the alleged subsidized and like products compete in the same market or multiple markets, which is a prerequisite for assessing whether displacement within the meaning of Articles 6.3(a) and 6.3(b) could be found to exist as alleged by the United States …

1131. Clearly, there is no inhibition on how a complainant may choose to formulate its claim as to the scope of the ‘subsidized product’ … This does not mean, however, that a panel has no duty to review the complainant's formulation of the scope of the ‘subsidized product’. Rather, the panel has a duty to ascertain the relevant product market or markets in which the complainant's and respondent's products compete. The notion of ‘subsidized product’ and ‘like product’ is, in each case, to be analysed as an integral part of a panel's duty objectively to assess a particular claim of serious prejudice and its obligation to assess the relevant market under Articles 6.3(a) and 6.3(b).

This clarification is to be applauded. It is, of course, the duty of the Panel to define whether in reality pain was inflicted on other Members by subsidies. That requires competitive relationships, real ones, and not just the claim of a party to the dispute. The duty of the Panel to examine what the complaining party presents as the factual basis for its claims seems a self-evident proposition. The Panel is supposed to make an ‘objective assessment’⁴⁸ of both facts and law; that seems hardly possible with a boilerplate internalization the position of one of the parties to the dispute. On the basis of the available empirical literature on the LCA industry, the benefits of a more refined definition of the product market would not have been theoretical, but, on the contrary, quite operational for the displacement analysis. Empirical studies of LCA industry⁴⁹ suggest that LCA models within the industry show quite different degrees of substitution between product ranges: LCA products in the same cluster (defined for example in terms of passenger capacity and range) compete closely with one another but the degree of substitution between the LCA models in different clusters vary widely. The importance of defining correctly the relevant product market is to enable a robust evaluation of displacement of a product by a product that exercises on it a competitive constraint. As noted by the Appellate Body⁵⁰ ‘[a]n assessment of the competitive relationship between products in the market is required in order to determine whether and to what extent one product may displace another. Thus, while a complaining Member may identify a subsidized product and like product by reference to footnote 46, the products thereby identified must be analyzed under the discipline of the product market so as to be able to determine whether displacement is occurring.’ A similar reasoning extends to the notion of the geographic scope of the relevant product market. Where the market is cross-border or even global, the distinction between sales to the domestic ‘market’ and sales to the export ‘market’ disappear, and so do the terms domestic or export market shares. It follows that any conclusions on discernible displacement of a like product by a subsidized product have to be made on the basis of an objective determination of the relevant product market.

2

Of course, what irritates at first glance is the geographic compartmentalization of markets (e.g., Taiwan, Australia, India) whereas one would have believed that LCAs are exposed to competition in a global market. The key to the Appellate Body's approach is the wording of Article 6.3 SCM.⁵¹ Its subparagraphs (a) and (b) both concern effects of state aid in a geographically well-defined market: on the one hand ‘the market of the subsidizing Member’ and, on the other hand, ‘a third country market’.

The Appellate Body, as any applier of the law, has to recognize the normative starting point, even though it may not represent a categorization current economic thinking would subscribe to.

1117. … A plain reading of Articles 6.3(a) and 6.3(b) therefore reveals that an analysis of displacement or impedance under those provisions is limited to the territory of the ‘subsidizing Member’ or the territory of any third country at issue. The manner in which the geographic dimension of a market is determined will depend on a number of factors: in some cases, the geographic market may extend to cover the entire country concerned; in others, an analysis of the conditions of competition for sales of the product in question may provide an appropriate foundation for a finding that a geographic market exists within that area, for example, a region. There may also be cases where the geographic dimension of a particular market exceeds national boundaries or could be the world market, even though Articles 6.3(a) and 6.3(b) would focus the analysis of displacement and impedance on the territory of the subsidizing Member or third countries involved.

The Appellate Body however mentions, almost in passing in footnote 2462, that ‘in terms of the geographic dimension of markets under Article 6.3(c) of the SCM Agreement, it may be appropriate to examine the “world market” and the conditions of competition as they exist in that market’. This, however, is the only occasion to do so, as the first two subparagraphs determine a territorial approach.

IV

1

An interesting economic and quantitative analysis of displacement of orders of Boeing that could be attributed to subsidies to Airbus is attempted by the Appellate Body Report in section IX.C.⁵² The analysis is predicated on defining the relevant product market as consisting of three submarkets – Single-aisle LCA, Twin-aisle LCA, and Very Large LCA. This provisional definition of LCA was selected as it was not contested by either party. The method of analysis adopted by the Appellate Body was to examine the trends in the volume of sales and relative shares of the orders of each of the manufacturers to specific countries, over a six-year reference period. The analysis is therefore very much along the lines outlined in paragraph 1048 of the Report. The analysis of the orders data over 2001–2006 leads the Appellate Body to confirm displacement of Boeing sales to half of the specific countries put forward by the complainant:⁵³ in terms of unit sales over the period to these countries this amounted to 70% of total orders. This can be regarded as significant, if attention is limited to just these countries, but much less significant in relation to global LCA orders over the reference period.

2

In regard to lost sales, the Appellate Body Report is careful to define the term and the approach needed to identify lost sales ‘caused’ by the challenged measures. As noted by the Appellate Body Report, this case was the first instance when the issue had been examined by it, under Article 6.3(c) of the SCM Agreement. The Report underlines that the analysis compare the sales actually made by the competing firm(s) of the complaining Member with the sales resulting from a counterfactual scenario in which the firm(s) of respondent Member would not have received the challenged subsidies.⁵⁴ The central element in the counterfactual underlying the Appellate Body's Report is that displaced and lost sales for Boeing to certain geographic destinations occurred because the challenged subsidies made it possible for Airbus to launch rival offerings at the time it did. But for the subsidies, there would not have been the availability of competitive alternatives and wider choice of LCA specifications, and hence no loss or displacement of Boeing sales over the reference period would have taken place.

3

In order to fine-tune displaced and lost sales of Boeing during 2001–2006, alternative counterfactual scenarios including the ‘no-subsidy’ case for Airbus are considered in the Report. For a consistent approach, it would have been necessary to develop a global demand growth evolution corresponding to delayed development of product families by Airbus. Such a counterfactual would have highlighted greatly reduced pressure on Boeing to undertake expensive development expenditure of its own product range. The Airbus Report does not consider that there would be a different evolution of global LCA demand-growth under the scenario of a reduced Airbus operation. The common assumptions under the counterfactual scenarios considered are that the development of LCAs by Boeing over the whole period and the global demand for all types of LCAs would be as it actually was during the whole period. An incumbent, dominant or not, has no profit incentives and faces no price pressures from captive customers to escalate developmental spending so as to meet or create prospective demand. On the assumption that Boeing's product development over the period did not benefit from subsidies, the basic analytical framework is certainly correct, but simulations reported equate the time pattern of global LCA growth development and product launches by the incumbent in the ‘no subsidy’ case to the realized development over the reference period of global LCA demand growth and product launches. The core dynamics of LCA industry in the form of strategic interactions between competitors regarding developmental expenditures leading to product launches that simultaneously drive demand growth, as well as react to that demand growth, are not adequately incorporated in the counterfactual simulations.

4

The conclusions⁵⁵ drawn in the Appellate Body Report on displacement of sales and lost sales favor, based on the evidential record before the Panel, the counterfactual scenarios that assume that without subsidies there would not have been the same development pace of LCA products of Airbus.

4. Conclusions

1

It is a mixed blessing for the WTO dispute-settlement mechanism that the Airbus and Boeing cases have been brought. If there ever was a constellation ill-suited for being ‘battled out in court’, it was this dispute between two close trading partners and military allies, sharing for the moment the lead with regard to large civil aircrafts. While this is not apparent when reading the Airbus Report – whereas in the sister Boeing Report⁵⁶ the strong involvement of the military is obvious, as the granting authorities are the Pentagon and NASA – it is clear that the European public authorities granting the subsidies in question did so not the least in order to avoid a complete dependency on US aerospace technology that would have had the potential to be crippling on many levels. Regardless of whether these vital interest connotations would have sufficed to reach the threshold necessary for an invocation of Article XXI GATT, they were (and are) real nonetheless. Certainly, the United States has a history of active avoidance of having independent adjudicators look at issues that are of security relevance, and the European Union has never pushed too hard in this respect, as the last-minute compromise in the Helms–Burton dispute illustrates.⁵⁷

Because of the economic significance of this dispute, the strong position of Boeing as a lobbyist, and the rather obvious violation of WTO and EU state-aid law by the Europeans, low priority was given to the arguments advanced by those who preferred a noncontentious negotiated settlement. The result is, of course, far from satisfactory for the US interests pushing that agenda, given that both the European Union and the United States were held to be not in conformity with their obligations under the SCM. Now a dispute that is utterly limited to two trade superpowers and military allies will feed another generation of lawyers’ children (maybe even cover their expensive graduate education …) and keep academics busy. While these results may allow us to rejoice, they would seem not to suffice to tip the balance in favor of an overall positive assessment: a good opportunity to not use the multilateral mechanism was missed.

That aside, and apart from some of the desiderata mentioned above, this is a well-crafted Report, in particular given the time constraints: for the United States, speedy justice in the WTO is of constitutional importance, as is illustrated by its threat to propose the application of (regular) consensus rule in the COOL case; given the completely different profile of the Airbus case, the right timing was very important.⁵⁸

The dispute DS316 is far from over: the European Union, on 5 December 2011,⁵⁹ notified the DSB that it had taken appropriate steps to bring its measures fully into conformity with its WTO obligations:

[T]he European Union took note of all elements of the DSB's recommendations and rulings, including, in particular, the … guidance on the way in which subsidies and adverse effects expire, dissipate, terminate or are otherwise removed or withdrawn. In undertaking this review, we consulted, among others, independent experts in: financial economics; investor behaviour; financial and cost auditing, accounting and controlling; product engineering; and Large Civil Aircraft (LCA) fleet management. We have also closely monitored and assessed LCA product and market developments … following the period covered by the Panel's review.

As a result of this review, the European Union has adopted a course of action that addresses all forms of adverse effects, all categories of subsidies, and all models of Airbus aircraft covered by the DSB's recommendations and rulings.

Specifically, in bringing its measures into conformity with its WTO obligations, the European Union has addressed all categories of subsidy covered by the DSB's recommendations and rulings: Member State Financing (MSF) loans, capital contributions, infrastructure support and regional aid. Amongst others, the European Union has secured repayment of MSF loans and terminated MSF agreements, increased fees and lease payments on infrastructure support to accord with market principles, and ensured that capital contributions and regional aid subsidies have, in the Appellate Body's words, ‘come to an end’ and are no longer capable of causing adverse effects. Additionally, the course of action adopted by the European Union affects Airbus’ A300, A310, A320, A330, A340 and A380 aircraft, as well as derivatives thereof, as implicated by the DSB's recommendations and rulings. Finally, as a result of these steps and other intervening market events, the European Union has addressed the forms of adverse effects covered by the DSB's rulings.

In the meantime, the United States had requested, on 9 December 2011, authorization to take countermeasures pursuant to Article 22 DSU and Article 7.9 SCM. Due to a sequencing arrangement, the matter is now before the original Panel: it is interesting, though, that the United States quoted para. 1264 of the Appellate Body Report:

Without the subsidies, Airbus would not have existed under these scenarios and there would be no Airbus aircraft on the market. None of the sales that the subsidized Airbus made would have occurred. As Boeing (or the other US manufacturer envisaged by the Panel) would be the only supplier(s) of LCA, it (or they) would have made the sales instead.

That seems to indicate a willingness on the side of the complainant to give a broad reading to the European Union's obligation to remove the adverse effects of the subsidy: the line of reasoning would be that as long as there is Airbus, the harmful effects of the subsidies can be felt. It will be interesting to see how this argumentation plays out: the allegedly wrongful birth of Airbus did not hinder the Appellate Body to accept in United States – Measures Affecting Trade in Large Civil Aircraft (Second Complaint) (DS 353) that the European Union was capable of suffering, pursuant to Article 5(c) SCM, serious prejudice, as a consequence of the lost sales that Airbus had to digest as a consequence of US subsidies.