1. Introduction and background
For more than 30 years, China's economy has grown at a spectacular rate. Since the early 1980s, GDP has doubled every seven years and exports have doubled every four years. China's dazzling export growth has put tremendous pressure on countries who import China's products, which, in turn, has led many countries to complain that China's exports are traded unfairly and causing injury to domestic competing industries.
Prior to China's accession to the WTO in 2001, members were not bound by WTO rules for how to treat imports from China. For instance, WTO members were not required to grant China most-favored-nation treatment and could simply unilaterally raise tariff rates applied against imports from China without violating any WTO rules. Despite this, policy flexibility countries did not generally opt to simply raise applied rates. Instead, countries chose to limit China's exports by resorting to policies of administered protection, and this most typically meant anti-dumping (AD) protection.
In Table 1 below, I document AD activity by the two largest ‘traditional’ AD users, the United States (US) and the European Union (EU). As seen, China accounted for 13–14% of AD activity and was the leading target of AD protection during the 1990s for both the US and the EU. AD actions against China grew sharply after China's WTO accession in 2001. China's share of US AD cases doubled from 13% in the 1990s to 27% in the 2000s. Similarly, China's share of EU AD cases rose from 14% in the 1990s to 29% in the 2000s. And, as was the case in the prior decade, China was the leading target of AD actions for both the US and the EU.
US and EU AD activityFootnote 1
The size of the anti-dumping duty (ADD) varies from case to case. There are four methods for computing the size of the AD margin and domestic industries generally opt for the method that produces the largest margin. For this reason, cases against China use the non-market economy (NME) methodology as that gives investigating authorities the maximum leeway to create large margins.Footnote 2 Broadly stated, under the NME methodology the investigating authority computes what it thinks it would cost to produce the subject product in the NME. But under NME rules, the authority computes the costs ignoring actual costs in the NME. Instead, the authority estimates a cost function and uses costs from a surrogate country to price inputs (e.g., wages in Thailand, price of electricity from Thailand).
NME methodology has a huge impact on AD margins. AD margins in US cases against China have averaged 154% as compared with an average of 49% for other targeted countries. AD margins for EU cases against China have averaged 61% versus an average of 31% for other targeted countries.Footnote 3
Given such large AD margins, one would think that ADDs computed using the NME methodology would be sufficient to eliminate any injurious effects of exports from China. Yet, US and EU domestic industries clamored for more. After several years of political lobbying, the US Department of Commerce (USDOC) reversed its policy against applying the countervailing duty (CVD) law to countries it deemed to be NMEs in March 2007.Footnote 4 That prohibition, first adopted by the USDOC in 1984 and in 1986 affirmed by the US Court of Appeals for the Federal Circuit in the Georgetown Steel case,Footnote 5 was based on the view that, given the highly distortionary nature of the role played by the governments of centrally planned economies with no discernible private sector, it would not be appropriate to apply the countervailing duty legislation to NMEs because a ‘non-market economy would in effect be subsidizing [itself]’.Footnote 6
The USDOC reconciled the Georgetown Steel case with its decision to apply CVDs against China by distinguishing the ‘Soviet-style economies of the 1980s’,Footnote 7 which were the subject of the Georgetown Steel case,Footnote 8 with present-day China. The USDOC considered that, while China should remain ‘an NME for purposes of the AD law’Footnote 9 in light of the continued perceived distortion in China's domestic sales prices and costs, and thus their inability to produce reliable ‘normal values’, China's economy had nevertheless changed in fundamental ways such that it was no longer considered to resemble the ‘traditional Soviet-style command economy’.Footnote 10
Pricing committees, or similar state agencies, administratively set nearly all prices in the Soviet-style economies of the 1980s. Moreover, prices were not fixed with any deference to the forces of supply and demand, but rather served as an accounting device between supplier and consumer enterprises. In contrast, although price controls and guidance remain on certain ‘essential’ goods and services in China, the PRC Government has eliminated price controls on most products; ‘market forces now determine the prices of more than 90 percent of products traded in China’.Footnote 11 (Emphasis added).
While the EU did not have a formal policy like the US, the EU's practice had long been not to allow simultaneous AD/CVD actions. As has happened with other policy changes (e.g., cumulation), the change in US policy effectively triggered a change in EU policy. In 2010, the EU changed its ‘unofficial prohibition’ and European industries also began to file simultaneous AD and CVD petitions against China.
The policy change altered the use of AD and CVD in both the US and the EU. As seen in Table 2 below, about 71% of US AD cases against China now involve a simultaneous CVD case. In the EU, about 19% of AD cases against China now involve a simultaneous CVD case.
Proportion of AD cases with related CVD case
Source: Bown (2016), Global Antidumping Database.
The decision to simultaneously use AD and CVD against NMEs is controversial and potentially WTO inconsistent because it may result in ‘double remedies’ (also sometimes referred to as ‘double counting’). Double remedy refers to when the simultaneous use of AD and CVD results in the same instance(s) of subsidization being offset twice. When this occurs, the simultaneous use of AD and CVD violates the Agreement on Subsidies and Countervailing Measures (SCM Agreement).
The double remedies’ issue was a core element of the recent WTO dispute United States – Definitive Anti-dumping and Countervailing Duties on Certain Products from China.Footnote 12 China argued that because it was designated an NME in the parallel AD investigations and thus subject to the US NME methodology for the calculation of normal values, any ADD calculated on the basis of that methodology would necessarily reflect any instances of subsidization, thereby rendering CVDs duplicative. The basis for this argument stems from the fact that under US law normal values based on surrogate countries are presumed to be unsubsidized, and thus not ‘depressed’ or affected in any way by subsidies granted by the government that is the target of the investigation. Where that is the case, the dumping margin on which the eventual duty is based would be higher than it would be if the normal value were depressed by the subsidies. This assumes, of course, that the subsidies countervailed in the parallel CVD investigation would have, in fact, depressed the normal value in the AD investigation pro rata (or ‘dollar for dollar’) if Chinese domestic costs/prices had actually been used to derive the normal value used in the comparison between export price and normal value.
2. Non-market economies AD rules in the GATT/WTO
The second interpretative note Ad Article VI(1) GATT 1947 (hereinafter: second Ad note) provides that:
It is recognized that, in the case of imports from a country which has a complete or substantially complete monopoly of its trade and where all domestic prices are fixed by the State, special difficulties may exist in determining price comparability for the purposes of paragraph 1, and in such cases importing contracting parties may find it necessary to take into account the possibility that a strict comparison with domestic prices in such a country may not always be appropriate.
This note gives GATT members carte blanche to determine normal value in AD investigations involving countries meeting these criteria.Footnote 13 In practice, members developed the surrogate or analogue country concept for this purpose, meaning that they would use prices or costs in a market economy country as the basis for normal value.
Furthermore, as neither this note nor any other provision in the GATT (or the WTO) provides a listing of countries which have a ‘complete or substantially complete monopoly of their trade and where all domestic prices are fixed by the State’ (colloquially and in various instances of WTO members’ domestic trade remedies legislation referred to as non-market economies), it was left to GATT members to unilaterally determine which countries met these requirements. As a matter of practice only non-GATT members such as China, Russia, Ukraine, and Viet Nam were treated as non-market economies and subjected to the surrogate country methodology, which means no country labeled an NME could challenge the designation.
Russia and Ukraine over time succeeded in being treated as market economies by the US and the EU. China and Viet Nam had to agree to special provisions in their Protocols of Accession to the WTO, effectively allowing other WTO members to continue to treat them as NMEs until 2016 and 2018, respectively, whether or not the conditions of the second Ad note are actually met.Footnote 14 Indeed, in the case of China, recognition among the negotiating members that its economy as of 2001 could not reasonably be characterized as the sort referred to in the Ad note is what prompted the inclusion of the special provision in the first place.
3. Double remedies and US–AD and CVD (China)
The WTO Panel ruled in favor of the US with respect to the double remedies’ issue. China cited no less than eight different articles from the GATT and the SCM AgreementFootnote 15 in support of its contention that the USDOC's imposition of double remedies in the underlying investigations was impermissible; the Panel rejected the arguments made in connection with each. Hence, even though the Panel found that double remedies were likely to occur, the articles cited by China simply did not address the issue of double remedies.Footnote 16
The AB disagreed and considered that, when read in light of various other articles for context, Article 19.3 of the SCM Agreement, which requires that duties be collected ‘in appropriate amounts’,Footnote 17 forbids the imposition of concurrent duties which offset the same subsidization twice. In addition to finding the concept of the collection of duties constituting double remedies to be incompatible with the notion of ‘appropriateness’,Footnote 18 the AB went a step further and placed the burden of proof on the investigating authority to rebut the presumption that there is, in fact, double remedies in cases in which a parallel AD/CVD investigation makes use of an AD NME methodology. The AB considered that while it did not necessarily agree with the pro rata argument advanced by China, and the accompanying conclusion that CVDs in dual investigations using NME methodologies were entirely accounted for in the dumping investigations, double remedies were nevertheless ‘likely’ to occur in such dual investigations.Footnote 19 This likelihood seemed to be the impetus for the AB's shifting of the burden onto the investigating authorities.
4. The economics of double remedies
4.1 Pass-through
The fact that simultaneous imposition of ADD and CVD may lead to double remedies has been recognized by the GATT signatories from the outset. Article VI:5 GATT provides that ‘no product … shall be subject to both anti-dumping and countervailing duties to compensate for the same situation of dumping or export subsidization’. The fact that Article VI:5 GATT notes the impermissibility of double remedies in the context of export subsidies but not domestic subsidies was a key element of the US’ legal argument.
The following two examples clarify why the GATT distinguishes between export and domestic subsidies. First, consider an example where a producer receives an export subsidy of US$5/unit; she sells in her domestic market at US$100 and for export at US$95; the effect of the export subsidy is to lower the export price by US$5. Article VI:5 implies that the investigating authorities could either impose an ADD of US$5 or a CVD of US$5; however, it could not impose both because that would create a double remedy.
Article VI:5 can be criticized on the grounds that it abstracts from what the recipient of the subsidy actually does with the money and/or, more generally, that it ignores the fungibility of money. Formally speaking, Article VI:5 embodies two very specific assumptions about the pass-through of the export subsidy: (i) the export subsidy is completely passed-through to the export price and (ii) none of the export subsidy is passed-through to the home market price. Despite the questionable empirical validity of these pass-through assumptions, the explicit link between the act of exporting and the receipt of an export subsidy appears to be the basis for limiting the scope of Article VI:5 to export subsidies.
Now contrast this example with the following which involves a countervailable domestic subsidy of US$7. Suppose investigating authorities observe a domestic market price of US$93 and an export price of US$89. In addition, suppose investigating authorities could determine that without the domestic subsidy the domestic market price would have been US$100 and the export price would have been US$96. Under these circumstances, the proper trade remedy would involve both an ADD of US$4, which would eliminate the price differential, and a CVD of US$7, which would restore the prices to what they would be without the domestic subsidies. Like the previous example, this example embodies very specific (and again perhaps empirically invalid) assumptions about pass-through of the subsidy; notably, the domestic subsidy is assumed to symmetrically and completely pass-through to both the export price and the home market price.Footnote 20
These examples make it clear that the economic logic behind Article VI:5 GATT hinges on the pass-through effects of domestic and export subsidies being different. One type of subsidy (export subsidy) is presumed to affect the export price (after all, it is contingent on export) and have little impact on the normal value while the other type of subsidy (domestic subsidy) is believed to affect the export price and normal value comparably.
In practice, however, the economic effects of export and domestic subsidies are likely to be far murkier. Export subsidies may well affect home market prices. Similarly, it is not hard to imagine a situation where, for example, a domestic subsidy has a large effect on the home market price but a small impact on the export price. An enormous economics literature has examined pass-through both theoretically and empirically over the past two decades and there is overwhelming evidence that pass-through is neither complete nor symmetric.Footnote 21 Even weak forms of the ‘law of one price’, which follows from an assumption of symmetric pass-through, are rejected in empirical study after empirical study.Footnote 22 Economists have repeatedly found that pass-through will not typically be symmetric across destination markets. The robust empirical finding is that a cost shock will result in a price change of x% to one market but of y% to another market. Market structure, technology, upstream and downstream cost conditions, market share, the nature and duration of cost shocks, and product differentiation have all been found to affect pass-through.
Of particular relevance is that specific provisions in US AD statute imply the calculated pass-through will be essentially the same as the pass-through assumed in Article VI:5 GATT. Specifically, the US’ NME methods result in a domestic subsidy having price pass-through effects similar to that which Article VI:5 GATT implicitly assumes occurs with an export subsidy. The US’ AD NME methods entirely eliminate the impact of domestic distortions (including any subsidization) on the normal value side; hence subsidies can only pass-through to the export price. Therefore, double remedies will occur under US procedures.Footnote 23
4.2 Methods for calculating normal value
The pass-through issue is complicated by the fact that the Anti-Dumping Agreement (ADA) provides four alternative methods for calculating the normal value in a dumping determination: the home market method, the constructed value method, the third market method, and the NME method.
Pass-through of a domestic subsidy may vary depending on the method used. Moreover, at the time when a firm sets its prices it does not know what method would be used if it were to later find itself subject to an AD investigation. This uncertainty further dims the prospect of empirically observing symmetric pass-through.
In broad terms, the dumping margin is calculated as
$$\matrix{ {ADD} & = & {P_{normal\_value} - P_{exp}^{US} \left( {C_{Home}\left( {s_1} \right),s_2} \right)} \cr} $$
where home market costs are denoted as C Home . The domestic subsidy may take the form of direct cash support by the government or subsidized loan terms (denoted by s 2). The subsidy can also affect the costs of production, say by offering the firm(s) lower electricity rates, lower input costs, etc. (denoted by s 1). This latter type of subsidy is referred to as the provision of goods or services at less than adequate remuneration (LTAR) and played a prominent role in US–AD and CVD (China).
The computation of
$P_{normal\_value}$
varies by the specific method used. Crucially, in each of the first three methods the normal value will depend on domestic costs while under the NME method the normal value depends on costs in a surrogate country.
Under the home market method for determining normal value, an investigating authority would compare the home market price with the export price. The examples in the preceding subsection were based on this method. Economic theory implies that both prices depend on the costs and therefore both prices should experience some pass-through of the subsidy.
Under the US’ constructed value approach, the normal value home market price is calculated using a cost-plus approach. The USDOC first determines how much of each factor of production is needed to produce the good and then values each input using information on prices in the home market. For example, if a case involved a product from Japan, the US would use Japanese home market prices for labor, the raw materials needed to produce the product, electricity, etc., of the producer concerned when constructing the cost of the product. In addition, the USDOC will add a markup for selling, general, and administrative expenses and profits. However, the constructed value method will not necessarily fully purge the subsidy distortion from either the normal value or the export prices because home market costs can be affected by the domestic subsidy, which in turn can affect both home and export prices. Moreover, one expects asymmetric pass-through with the constructed value approach. The reasoning is that the pass-through of the subsidy to the normal value is via a fixed-coefficient function, which imposes a specific relationship between the subsidized inputs and the price. The pass-through to the normal value will not depend on other factors that economic theory tells us affects pass-through (e.g., the elasticity of demand). By contrast, pass-through of the subsidy to the export price (
$P_{exp}^{US} $
) will depend on pass-through elasticity which depends on many factors such as market structure, the firm's share in the export market, technology, etc. One would not expect two such very different approaches to produce the same pass-through elasticity.
Under the third country method, the export price of the like product to an appropriate third country is used for the normal value. As with the home market method, standard microeconomic theory predicts that both prices – the export price to the third country and to the complaining market (
$P_{exp}^{US} $
) – will be affected by pass-through of the domestic subsidy via distorted input costs and/or direct subsidization.
The non-market economy method is the final approach. The ADA gives importing countries significant discretion in the calculation of normal value of products exported from NMEs. As with the constructed value method, the USDOC gathers information on the company's ‘factors of production’ – the physical quantities of all the inputs used in producing the merchandise. However, in the NME method the USDOC values those inputs on the basis of prices in a surrogate country. The normal value is therefore a cost-based normal value derived from company-specific factors of production combined with surrogate country prices of those factors. The USDOC only chooses surrogate costs which are deemed to be fairly traded. Thus, under the US’ NME procedures a non-market economy's domestic subsidization program should have no effect on the constructed normal value. However, subtle nuances in the precise constructed costs and specific input pricing data might mean the NME constructed value is not fully purged of the domestic subsidies. Even so, there can be no doubt that the US approach results in a considerably muted (if any) impact of the subsidies on the normal value. By contrast, NME's domestic subsidization programs will have a more direct effect on the export price (
$P_{exp}^{US} $
) via distorted input costs and direct subsidization.
4.3 Pass-through and normal value
The double remedies problem hinges on whether the impact of a domestic subsidy (which is being offset by a CVD) is also partially or fully captured by the ADD. For the home market and third country methods, if the subsidized firm passes through the subsidy symmetrically to all measured prices, then double counting will not occur and the imposition of concurrent AD and CVD duties will not offset the subsidization twice. For these methods, double remedies do not depend on full pass-through but symmetric pass-through.
This, however, is not the case for the NME method. The USDOC's NME approach means the main avenue, and in all likelihood the only avenue, for the subsidy to influence the dumping margin is the export price. Therefore, double remedies are inevitable to some extent unless a given export subsidy has had zero impact on the export price – any difference between the NME constructed normal value and the export price is already corrected by the ADD, at least to the extent that there has been pass-through to the export price.
The objective of the US’ surrogate country methodology is to calculate the price of the product as if it would have been produced and/or sold unsubsidized by a producer in a market economy; i.e., under normal market conditions without government interference, the resulting normal value would normally be an unsubsidized price. The process of constructing the normal value for a NME should therefore purge all subsidies. Given that the subsidy has already been taken into account when calculating the dumping margin, a concurrent CVD will generate double remedies. USDOC NME procedures therefore virtually guarantee that double remedies will occur.
Article VI:5 GATT disallows simultaneous use of AD and CVD for export subsidies based on a presumption that the export subsidy will only affect the export price – when in fact money is fungible within the firm and hence it is possible an export subsidy could affect the home market price too. It is logical to believe the economic basis for Article VI:5 GATT applies equally to the US NME methodology because domestic subsidies are essentially purged from the first term of the AD duty, and are allowed to fully pass-through to the second term.
The implicit assumption of Article VI:5 GATT is the attenuated ability of an export subsidy to pass-through to the normal value. Likewise, by selecting a ‘market economy’ surrogate country, the logic of the US NME methodology is designed to result in little or no ability for domestic subsidies in the NME to pass-through to the normal value. Moreover, the US’ countervailing duty statute required the CVD duty to be equal to the full value of the calculated benefit (i.e., prohibited any adjustment in light of pass-through or other considerations).Footnote 24
Therefore, the economic logic behind Article VI:5 GATT and the economic implications of US NME methods are the same. Given the particulars of the US procedure, Article VI:5 GATT could have been the basis for a strong definitive finding that the US practice was inconsistent. The AB instead opted for a more cautious approach. The AB concluded that ‘double remedies were “likely” to occur in cases where NME methodology is used to calculate the margin of dumping’ and permitted the US to pursue implementation by focusing on how much double counting occurs.Footnote 25 The AB seemed to understand that if an investigating authority were to make adjustments to account for such double counting, then the two remedies could be used simultaneously. Depending on the specific procedures, such adjustments may resolve the double remedies’ issue.
Nevertheless, by focusing on measurement, the AB may have inadvertently opened the door to additional double remedy challenges. What if affected countries can measure and document the extent of double remedies under other methods for calculating normal value? It seems improbable that the WTO would allow double remedies under some methods for calculating normal value but not under the NME method.
4.4 Examples of double remedy
The AB report does not clearly address (i) the pass-through implications of Article VI:5 GATT and (ii) the applicability of that logic for the current dispute, especially given the specifics of US AD and CVD rules. The AB determination has resulted in the dispute being viewed as a measurement issue. Even the US acknowledges that measurement is extremely messy and complicated.Footnote 26 Going forward, it will be very difficult for future Panels to ascertain whether the measurement was in fact done correctly.
I now present two examples of the implementation approach proposed by the AB. In the first example, the subsidies are of the form of lower (LTAR) input costs. In the second example, I consider the situation where market benchmark prices (used in the CVD calculations) differ from the surrogate country prices (used in the AD calculations).
Example 1 – subsidized inputs
I begin with a simple example where the subsidies only affect the ‘materials, labor, and electricity’ costs of the firm (see Table 3).
AD duty calculation
To keep the example as simple as possible, I assume other factors in the USDOC's factors of production analysis (overhead, SG&A, and profit) are not affected by the subsidies. In this example, I assume that (i) the material inputs are subsidized by US$150 less than their surrogate cost (US$300 versus US$450), (ii) labor costs are subsidized by US$5 (US$40 versus US$45), and (iii) electricity costs are subsidized by US$5 (US$20 versus US$25). As a result, the cost of the firms’ ‘material, labor, and electricity’ (or MLE) is US$160 less than their comparable surrogate values (US$360 versus US$520).
Double remedy– subsidized inputs
I assume the Chinese firm's home market price is US$535 and the export price is US$500.Footnote 27 If China were considered a market economy and the home market method were used to compute the normal value, then the dumping amount would be US$35 (US$535 less US$500), implying an AD margin of 7%.Footnote 28
Under US NME rules, however, the home market prices for materials, labor, and electricity would be replaced by those in the surrogate country. For instance, the USDOC might use Thai costs. In the example, the NME normal value is US$695, which implies a dumping amount of US$195 (or 39%). Of the dumping amount obtained using NME methods, 32% is due to using the NME methodology (39% less 7%).
CVD duty calculation
Under US CVD rules, the subsidy benefit is based on a USDOC ‘market benchmark’ valuation. The USDOC attempts to ascertain what the market prices were at the time the Chinese firms purchased the inputs. Given that it was investigating both AD and CVD allegations, it would seem sensible (and simplest) if the USDOC just used the prices in the surrogate country (which were determined during the AD part of the investigation). However, that is not what the USDOC does. Instead, it creates what it believes is a representative market price. This market benchmark price could be the average of prices in ‘comparable’ countries, be based on another country's price, or based on reliable and verifiable market information. In some cases, the market benchmark price is the same as the surrogate country price. Often it is not.
To keep this first example as simple as possible, I assume that the market benchmark is the same as the surrogate cost. This means that the USDOC would determine that the subsidy benefit is US$160 (US$695 less US$535), implying a CVD of 32%. This is precisely the dumping amount that is due to NME methods. In other words, in this example the CVD is exactly equal to the dumping amount attributable to NME methodology. The full amount of the CVD is a double remedy.
If there is any doubt that double remedies have a large effect, the normal value calculation is US$695. This in fact would be the destination price with the NME AD duty levied. However, if the unadjusted CVD is also levied the destination price is US$855.
Example 2 – subsidized inputs combined with mismatch of market benchmarks and surrogate country valuations
I again assume that the subsidy only affects material, labor, and electricity costs (see the far right column in Table 3). The additional complication in this example is due to the difference between the market benchmark and surrogate country valuations.
AD duty calculation
The AD duty calculation in this second example is the same as in the previous example. Due to the subsidy, the firms’ MLE is US$160 less than the surrogate values (US$360 versus US$520).
I again assume the Chinese firm's home market price is US$535 and the export price is US$500. If China were considered a market economy and the home market method were used to compute normal value, then the dumping amount would be US$35 (US$535 less US$500), implying an AD margin of 7%.
As in the previous example under US NME rules, the home market prices for materials, labor, and electricity would be replaced by those in the surrogate country. In our example, the NME normal value is US$695, which implies a dumping amount of US$195 (or 39%). Once again, of the dumping amount obtained using NME methods 32% is due to using the NME methodology (39% less 7%).
CVD duty calculation
I now consider a complication stemming from the US’ use of ‘market benchmarks’ in its CVD analysis. I now allow the market benchmark valuation to deviate from the surrogate country values. Given that the USDOC basis for determining the ‘market benchmark’ is entirely different from the basis for picking a surrogate country, this mismatch often occurs. For example, while the surrogate country might value the raw materials at US$520 (used for the AD calculation), the market benchmark might value those same raw materials at US$575 (used for the CVD calculation).
In this example, the subsidies are determined to have produced a benefit of US$215 (as compared to US$160 in the first example). As a result, the CVD margin is larger (43%) than the dumping amount. Despite this mismatch, double remedies occur. The CVD margin can be decomposed into the part that is due to the use of NME methods when computing the dumping amount (32%) and the part that is due to the mismatch between the surrogate country and benchmark valuation (11%).
In this example, the AD remedy alone is sufficient to make the destination price equal to the normal value (US$695). When the unadjusted CVD is also levied, the destination price rises to US$910.
4.5 Caveats
The intention of the US NME method and the US CVD is the same – to purge the effects of government involvement on prices. The AB could have used Article 19.3 of the SCM Agreement to definitively find the US practice inconsistent. Given the US’ NME AD methodology, I believe that some form of double remedies are automatic. The real issue is not if double remedies will occur under the NME method but rather how much occurs.
There are three possible qualifications to this conclusion. First, one might be concerned that surrogate country producers are also benefitting from subsidies. While this might be a valid concern for NME methods used by other WTO members, I do not believe it is relevant for the US. According to USDOC procedures, the surrogate country is chosen because (i) it is comparable with the NME and (ii) because it is unsubsidized. Under USDOC procedures, a country with subsidization in the relevant input market would not be selected as a surrogate country.
Second, one might be concerned that the Chinese producers have such a dominant role on the world market of the product concerned that their (subsidized) prices have depressed prices on the world market or, at least, in the surrogate country. Again, I believe this concern is generally not applicable to the US situation given USDOC procedures which construct normal value using surrogate costs. I stress, however, that this conclusion hinges on the fact that the US uses the factors of production methodology. While the NME's subsidy does not affect surrogate country values because of the US’ factors of production approach, it could affect NME surrogate country normal values used by other WTO members. For example, the EU will choose a surrogate country and then normally use the prices in that country. In such cases, the concern about the possibility of China's subsidized prices depressing prices in the surrogate country might be legitimate.
Third, there is a concern that the profit margin used by the USDOC might be affected by the subsidy. That is, the subsidies might allow the export prices charged by the Chinese firms to be lower than they would otherwise be, which in turn might lower the profit rate in the surrogate country. Thus, even though surrogate values are used, the ADD might not fully capture the effects of the subsidy. This is a valid concern; however, it was not part of the US’ argument to the AB.
5. Concluding comments
The double remedies’ issue is not inherent to all simultaneous AD/CVD actions but rather is crucially tied to the use of the NME AD methodology along with CVD. China and Viet Nam Protocols of Accession to the WTO seemingly require WTO members to treat them as market economies effective December 2016 and December 2018, respectively. If so, the double remedy issue may soon be a footnote to WTO dispute resolution history. Yet, all signs point to the US and the EU resisting the change in market economy status.
In mid-December 2016, China requested consultations with the US and the EU regarding market economy status.Footnote 29 While the political value of foot-dragging might be part of the motivation of the US and the EU, there is reason to believe this is not just obstruction on the part of the US and EU. Rather, Bown and Mavroidis (2013) argue that the AB could rule against China in a dispute.Footnote 30 For instance, in EC–Fasteners the AB offered views broadening the sense of what might constitute government control statingFootnote 31
These situations may include: (i) the existence of corporate and structural links between the exporters, such as common control, shareholding and management; (ii) the existence of corporate and structural links between the State and the exporters, such as common control, shareholding and management; and (iii) control or material influence by the State in respect of pricing and output.Footnote 32
And later, when discussing the criteria for state control, the AB noted
evidence of State control or instruction of, or material influence on, the behaviour of certain exporters in respect of pricing and output. These criteria could show that in the absence of formal structural links between the State and specific exporters, the State in fact determines and materially influences prices and output.Footnote 33
The AB's language suggests that market economy status will not be automatically granted to China and Viet Nam and hence that the double remedies issue may remain an important issue for many years. If so, as double remedies continue to be applied China will continue to argue that it is suffering from unfair application of AD and CVD rules.
