February 26, 2004
In 1984, the U.S. Department of Commerce (Commerce) made a policy decision that, as a matter of definition and the economic system that existed in non-market economy countries, U.S. countervailing duty (CVD) law does not apply to non-market economies. The legal and economic bases upon which Commerce made this policy decision have changed. Indeed, the WTO Uruguay Round negotiation resulted in a new Agreement on Subsidies and Countervailing Measures (SCM) that included a new and specific definition of subsidies. This SCM definition of subsidies has been implemented into U.S. law.
In order to change Commerce's policy, Commerce need only provide a reasoned basis and analysis for a change in policy. In view of the increasing importance of Chinese trade with the United States, and as Commerce prepares to consolidate all China cases into one office,[1] it is now an appropriate time for Commerce to reexamine its policy of not applying U.S. CVD law to non-market economies, particularly China.
Since 1984, Commerce has considered that, as a matter of law, countervailing duty law cannot be applied to exports from a Non-Market Economy (NME) country because subsidization is a market economy phenomenon and cannot exist in an NME.
Commerce first made this determination in 1984 in Carbon Steel Wire Rod from Czechoslovakia, 49 Fed. Reg. 19370 (May 7, 1984) (final negative CVD determination) and Carbon Steel Wire Rod from Poland, 49 Fed. Reg. 19374 (May 7, 1984) (final negative CVD determination). Following its decisions in the Wire Rod cases, Commerce rescinded initiations of CVD investigations on imports of potash from the Soviet Union and the German Democratic Republic. At that time, all four countries (Czechoslovakia; Poland; Soviet Union; German Democratic Republic) were considered non-market economies because each was characterized by central government control of prices and allocation of resources.
Commerce's NME classification was founded on an economic analysis that concluded that, in countries that relied on the existence of central planning to allocate resources and prices, markets did not exist. Commerce said:
We believe a subsidy (or bounty or grant) is definitionally any action that distorts or subverts the market process and results in a misallocation of resources, encouraging inefficient production and lessening world wealth.
In NMEs resources are not allocated by a market. With varying degrees of control, allocation is achieved by central planning. Without a market, it is obviously meaningless to look for a misallocation of resources caused by subsidies. There is no market process to distort or subvert. Resources may appear to be misallocated in an NME when compared to the standard of a market economy, but the resource misallocation results from central planning, not subsidies.
It is this fundamental distinction -- that in an NME system the government does not interfere in the market process, but supplants it -- that has led us to conclude that subsidies have no meaning outside the context of a market economy.[2]
Thus, Commerce believed that, without markets, there would be no way to quantify NME subsidies. Because of pervasive control of prices and resources, Commerce said it could not disaggregate government actions in such a way as to identify the exceptional action that is a subsidy.
Commerce's decision was contested and the U.S. Court of International Trade (CIT) reversed it.[3] However, upon appeal to the U.S. Court of Appeals for the Federal Circuit (CAFC), the CIT's decision was reversed and Commerce's decision was affirmed.[4] The CAFC accepted Commerce’s reasoning because it could not say that Commerce’s decision was “unreasonable, not in accordance with law, or an abuse of discretion” in view of the discretion accorded administrative agencies.[5] The CAFC also opined that subsidized unfair competition cannot exist in an NME. Quoting Commerce, the CAFC said:
[T]he nonmarket environment is riddled with distortions. Prices are set by central planners. “Losses” suffered by production and foreign trade enterprises are routinely covered by government transfers. Investment decisions are controlled by the state. Money and credit are allocated by the central planners. The wage bill is set by the Government. Access to foreign currency is restricted. Private ownership is limited to consumer goods.[6]
Based on this understanding of the way NMEs worked at that time, and specifically with reference to “export incentives,” the CAFC said that “even if one were to label these incentives as a 'subsidy,' in the loosest sense of the term, the governments of those nonmarket economies would in effect be subsidizing themselves.”[7]
Thus, the bases for Commerce's policy of not applying U.S. CVD law to NMEs may be summarized as follows:
1. Definitional: a subsidy is any government action that distorts markets;
2. Economic/Factual: production and investment and prices are all controlled by central planning which results in a market that is not rational, leaving prices and costs meaningless, and subsidies impossible, as subsidies are only meaningful in market economies; and
3. Practical: there are no benchmarks in an NME with which to quantify any subsidies.
It has been 20 years since the original Commerce decisions and the CAFC's affirmance of Commerce's policy. In that time, there have been many developments relevant to the rationales relied upon by Commerce and the CAFC. In fact, based on an analysis of current factors, Commerce has changed the status of the Czech Republic and Slovakia (successor states to Czechoslovakia), Poland and Russia from "non-market" to "market" economies. While some antidumping authorities have changed the status of China to a "market" economy, this paper does not suggest that Commerce take such a major step. However, this background is provided to illustrate that market conditions in NMEs do change.
In 1994, the United States enacted the Uruguay Round Agreements Act, implementing the results of the Uruguay Round negotiations establishing the World Trade Organization. Revised agreements, including the Agreement on Subsidies and Countervailing Measures (SCM Agreement ) were also adopted. The original GATT Articles (including VI, XVI, and XXIII) were not amended. One of the notable features of the new SCM Agreement was that, for the first time, an explicit and expansive definition of subsidies was agreed upon. Although the prior 1979 Subsidies Code, and GATT Articles VI (Antidumping and Countervailing Duties), XVI (Subsidies) and XXIII (nullification or impairment of benefits through government action) provided rules for subsidies discipline and application of countervailing duties, they did not provide a definition of a subsidy. The 1979 Subsidy Code used the terms “subsidy” and “subsidize” without elaboration. GATT Article XVI, paragraph 1, referred to “any subsidy” as including “any form of income or price support, which operates directly or indirectly to increase exports of any product from, or to reduce imports of any product into" a country. The definitional gap was filled by Article 1 (“Definition of a Subsidy”) of the WTO SCM Agreement, which states that a subsidy shall be deemed to exist if:
(a) (1) there is a financial contribution by a government or public body,[8]
or
(a) (2) there is any form of income or price support in the sense of Article XVI of GATT 1994;
and
(b) a benefit is thereby conferred.[9]
Article 2 of the SCM Agreement requires that, to be actionable, a subsidy must be given to a specific enterprise or industry or group of enterprises or industries.
Significantly, the SCM Agreement, unlike Commerce's 1984 working definition and GATT Article XVI, defines a subsidy based on what it is, instead of what it does. This approach is much more practical because a subsidy can be identified by its characteristics; it is not necessary to examine the effects of a subsidy in order to determine whether it is, in fact, a subsidy.
In view of the current WTO and U.S. statutory definition of a subsidy, Commerce's 1984 definition of a subsidy as being any action that distorts markets is outdated and no longer relevant. Absent the "economic effects" approach used by Commerce in 1984, its conclusion of non-application of CVD law to NMEs falls apart. While the result of a subsidy may be market distortion, that is not the standard by which to judge whether an action is a subsidy.
Neither the WTO SCM Agreement nor U.S. law provides for an NME exception from CVD law application or WTO discipline. Interestingly, the CIT's decision in Continental Steel noted that the CVD law specifically applies to “any country.”[10] Although Commerce recognized this, it considered that the nature of NMEs required an additional jurisdictional test to determine if NMEs could subsidize. However, the CIT said that a failure to meet this jurisdictional criteria, would amount to a per se exemption and be in conflict with the plain statement that the law covers any country.[11] The CAFC's decision did not comment on the CIT’s determination that CVD law applies to subsidies in any country. The CAFC reversed the CIT because it determined that Commerce's conclusion that NMEs could not confer a subsidy was not unreasonable, accepting Commerce’s definition of subsidy and characterization of non-market economies.[12]
Although the SCM Agreement's change of subsidy definition alone would be a basis for Commerce to change its policy, an examination of the evolution of economic and factual conditions is needed to determine if, in fact, subsidies exist in China and are measurable.
Since there is now a definition of a subsidy, the current economic circumstances in China need to be examined in relation to Commerce's 1984 decisions to see if subsidies can, and in fact do, exist in China. The economic situation in China today is far different than that that existed in the four NMEs that Commerce examined in 1984. China is in the process of a gradual liberalization of its markets, which was acknowledged in the Agreement on Market Access between China and the United States of November 15, 1999 (at p. 4). Now, it is not accurate to characterize China as having an economy totally directed by the State and one in which the State owns or controls all means of production. Currently, much of China’s GNP is produced by private enterprises with a declining share produced by State-controlled enterprises. In China today, there is joint venture production by foreign firms and increasing amounts of foreign direct investment.
1. Independent Studies of Chinese Economic System
a. IMF review
The transition of China from a state-directed economy was reviewed in a 1993 study published by the International Monetary Fund.[13] This study identified four phases of reform:
1. 1978-1984: Government policies placed greater emphasis on material incentives and allowed a larger role for the market. Farming was decentralized from the cooperative to the household level. China began to experiment with allowing State-owned enterprises to retain profits. Preferential policies were conferred on special economic zones to attract foreign investment, and act as laboratories for bolder market-oriented reforms.
2. 1984-1988: Reform in the urban industrial centers, following the success of decentralization of farming. This included the introduction of taxation of enterprises, reform of the wage system to establish a link between productivity and pay, opening of 14 major cities to foreign investment, and other market oriented reforms.
3. 1988-1991: Retrenchment. The prior reforms spurred demand and production, leading to double-digit inflation. Some earlier reforms were reversed under a “Rectification” program which stabilized prices but caused a sharp slowdown in the economy.
4. 1992-1993: (the date of publication) An end to the “rectification” program and a decision to accelerate the process of reform and opening up, establishing the goal of creating a “socialist market economy system.” The Chinese constitution was amended in 1992 to delete references to a planned economy and establish the new goal of creating a market system.
b. PriceWaterhouseCoopers review
The economic structure of China has continued to change since 1993. PriceWaterhousCoopers estimates two-thirds of China’s GDP is generated by the non-state sector, and around half contributed by domestic private enterprises.[14] This document notes that registered private businesses rose from 90,000 in 1998 to over two million in 2001.
2. Commerce's calculation of individual dumping margins for Chinese exporters
Commerce itself recognizes the Chinese market reforms in its antidumping investigations of Chinese products. While still considering China to be a non-market economy, Commerce has recognized in practice that state control is not all-pervasive. Since 1991, Commerce has allowed Chinese exporters to receive an “individual” rate of dumping duty if the exporter can demonstrate in law and fact that no level of government controls its export activities.[15] Since 1991, virtually all Chinese exporters have received individual rates, with the exception of companies who failed to provide accurate and verifiable responses to Commerce questionnaires. As facts available, these companies are given the China-wide rate, which is normally very high. The practice of calculating individual rates has become routine in Commerce's preliminary determinations of sales at less than fair value in cases concerning Chinese imports. In fact, the practice is seldom contested in final determinations. When it has been, Commerce has almost invariably concluded the subject Chinese company is free of government control of its export activities, and deserves an individual rate based on its own factors of production.
In making its decision as to whether individual companies are free of government direction and control, Commerce examines two sets of factors, neither of which existed in the NMEs described by Commerce in 1984. The first step is to examine three de jure factors:
1. whether there are restrictive stipulations associated with the exporter’s business and export licenses;
2. legislative enactments decentralizing control of companies; and
3. other formal measures by the Government decentralizing control of companies.
The second step is that, in addition to the legal status of an exporter, Commerce examines four de facto items:
1. whether the export prices are set by, or subject to the approval of, a governmental agency;
2. whether the respondent has the authority to negotiate and sign contracts and other agreements;
3. whether the respondent can retain the proceeds from its export sales and make independent decisions regarding the disposition of profits or financing of losses; and
4. whether the respondent has autonomy from the government regarding the selection of management.
This set of criteria was followed in a recently completed investigation of Chinese exports.[16] In this particular case, of the four respondents receiving individual rates, U.S. firms wholly or partially owned three respondents. The fourth was an employee-owned enterprise.
None of de jure and de facto factors Commerce now routinely examines were foreseen in 1984.
The practice of granting individual rates to individual Chinese exporters has become normal, with almost all respondents asking for individual rates receiving them. The very frequency of Commerce determinations that separate rates are appropriate indicates how much China differs from the typical NME conditions that Commerce examined in 1984.
3. Legal changes in China
On December 22, 2003, the Communist Party formally tabled an amendment to the Chinese Constitution to provide "private property obtained legally shall not be violated."[17] The members of the Standing Committee of the National People's Congress (NPC) passed the draft amendments to the Constitution in Beijing on December 27 indicating formal adoption will be in March 2004.[18]
C. China's Accession to the WTO
In the negotiations with China for accession to the WTO, the U.S. government negotiated strongly, and successfully, to impose disciplines on Chinese subsidies. These conditions for WTO membership became part of the Protocol of Accession for China. China agreed to eliminate all export subsidies.[19] China also agreed that WTO Member Authorities could use non-Chinese benchmarks for subsidy quantification if Chinese benchmarks were not available or could not be adjusted.[20]
The WTO Accession Protocol for China (Annex 5A) listed 24 domestic subsidy programs which China did not agree to terminate or phase out. However, some members of the Working Party on Accession of China to the WTO considered the list incomplete.[21] In particular, they felt that some subsidies, such as “policy” loans by State owned banks, forgiveness of debt, and the selective use of “below-market” interest rates should have been notified.[22] There was also reference to unnotified tax subsidies, and subsidies provided by sub-national governments. In Annex 5B to the Protocol, China listed three export subsidies to be phased out, but some members of the Working Party also considered this list incomplete.[23]
It is notable, when reviewing the U.S.-China Accession Agreement and the WTO Accession Protocol and Report of the Working Party, that the Members of the WTO, including the United States, developed and approved accession documents which identify Chinese domestic and export subsidies and provide alternate methods of subsidy measurement. There is no indication that any of the Members involved in the accession process, including China itself, believed that subsidies do not or could not exist in the present Chinese economy. It would be strangely inconsistent for the United States to negotiate disciplines for Chinese subsidies while, at the same time, believing Chinese subsidies could not exist.
D. Practical Constraints: lack of benchmarks to measure subsidies
In 1984, Commerce believed that there could be no benchmarks in an NME with which to quantify any subsidies, due to the absence of markets. However, as evidenced by its choice of factors to examine in determining whether individual Chinese firms are free of government direction and control in their export activities, and its decisions to grant separate rates, Commerce has moved far away from its 1984 views of non-market economies.
Of course, there still exists a need to quantify any subsidy in order to countervail it. A common benchmark in CVD investigations in measuring preferential loans or identifying the discount rate for grants is the market rate of interest. The International Monetary Fund publishes three interests rates for China in its International Financial Statistics: the Bank Rate, Deposit Rate, and Lending Rate. A variety of papers and studies document the mixed nature of the economy in China today.
Of course, practical difficulties could arise in the course of investigating Chinese subsidies, but Commerce cannot be sure of practical difficulties until it actually conducts CVD investigations of Chinese imports. One would not expect to find perfect markets in China, given that remnants of state control of economic activities are still in place. However, Commerce has never said that the market must be perfect in order to determine subsidy benchmarks.[24] In any event, the current economic structure in China bears no relation to the economic structure that existed in 1984. If, in investigating a Chinese subsidy, Commerce encounters practical difficulties in finding a benchmark, the default conclusion surely should not be that a subsidy does not exist. As noted above, in the WTO Protocol of Accession, China agreed that Members may use external benchmarks, even if only as a last resort.
Commerce has applied the current policy of non-application of U.S. CVD law to NMEs continuously since 1984. Commerce's policy, however, is not based on or required by statute, regulations or legislative history. As reviewed above, there is good reason for Commerce to change its policy of non-application of U.S. CVD law to NMEs, particularly China. And a change in policy by Commerce supported by a reasoned basis would likely be upheld, if challenged, by Commerce's reviewing courts.
In reviewing agency interpretations of law where the statute and legislative history are not clear and conclusive, courts normally accord deference to the agency. In Rust v. Sullivan, the U.S. Supreme Court said:
When we find, as we do here, that the legislative history is ambiguous and unenlightening on the matters with respect to which the regulations deal, we customarily defer to the expertise of the agency.[25]
Significantly, the Supreme Court said that this deference also extends to an agency's departure from a prior policy when the change is accompanied by a reasoned analysis. The Court stated:
This Court has rejected the argument that an agency's interpretation "is not entitled to deference because it represents a sharp break with prior interpretations" of the statute in question. * * * In Chevron,[26] we held that a revised interpretation deserves deference because "[a]n initial agency interpretation is not instantly carved in stone" and "the agency, to engage in informed rulemaking, must consider varying interpretations and the wisdom of its policy on a continuing basis." * * * An agency is not required to '"establish rules of conduct to last forever,'" * * * but rather "must be given ample latitude to 'adapt [its] rules and policies to the demands of changing circumstances.'"[27]
The principles enunciated in Rust v. Sullivan actually support a change in agency interpretations when such interpretations no longer represent the path of wisdom and changing circumstances demand adaptation. Thus, applying Rust v. Sullivan here, the following points are most apposite as to why a prudential change in policy would be upheld.
· Commerce's initial policy interpretation was "not instantly carved in stone.”
· It is incumbent on Commerce to "consider varying interpretations and the wisdom of its policy on a continuing basis.”
· Commerce must be afforded "ample latitude to adapt its rules and policies to the demands of changing circumstances."
· Courts will accord deference to a change in policy by Commerce even if it represents a sharp break from prior longstanding policy, as long as Commerce provides a reasoned analysis for its change in policy.
In reviewing the history underlying Commerce's policy decision not to apply U.S. countervailing duty law to NMEs, together with subsequent developments in international rules and changes in the structure of the Chinese economy, it is readily apparent that Commerce's 1984 decision to, in effect, exempt NMEs (particularly China) from the application of U.S. CVD law would not be a prudent or realistic approach today.
Indeed, the world has changed so much since 1984 that most of the countries then classified by Commerce as "non-market" have since become "market" economy countries to which Commerce applies countervailing duty law as a matter of course.
Accordingly, Commerce should immediately reconsider its current policy of not applying U.S. CVD law to NMEs (particularly China) with an eye to bringing it into accordance with the statute, international agreements, agency practice in granting individual rates to NME producers whose export prices are not controlled and changing market conditions (e.g., in China). Such a change in policy, supported by a reasoned analysis, would almost certainly be accorded deference by the courts.
[1] Prepared Testimony of Assistant Secretary James J. Jochum before the Senate Committee on Governmental Affairs Subcommittee on Oversight of Government Management, the Federal Workforce, and the District of Columbia, December 9, 2003, page 13; available at http://govt-aff.senate.gov/_files/120903jochum.pdf.
[2] Carbon Steel Wire Rod from Czechoslovakia, 49 Fed. Reg. 19370, 19371 (May 7, 1984) (final negative CVD determination) (emphasis added); Carbon Steel Wire Rod from Poland, 49 Fed. Reg. 19374, 19375 (May 7, 1984) (final negative CVD determination) (emphasis added).
[3] Continental Steel Corp. v. United States, 614 F. Supp. 548 (CIT 1985).
[4] Georgetown Steel Corp. v. United States, 801 F.2d 1308 (Fed. Cir. 1986).
[5] Georgetown Steel Corp., 801 F.2d at 1318, citing Chevron U.S.A. Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837 (1984).
[6] Georgetown Steel Corp., 801 F.2d at 1315.
[7] Georgetown Steel Corp., 801 F.2d at 1316.
[8] This includes any direct transfer of funds; fiscal incentives such as tax credits; provision of goods (other than general infrastructure) or purchase of goods; payments to a funding mechanism or direction of a private body to carry out what would normally be government functions. SCM Agreement, art. 1.1(a)(1)(i)-(iv).
[9] SCM Agreement, art. 1.1.
[10] Continental Steel Corp. v. United States, 614 F. Supp. 548, 550 (CIT 1985).
[11] Continental Steel Corp, 614 F. Supp. at 550 (emphasis in original).
[12] Georgetown Steel Corp., 801 F.2d at 1318.
[13] Michael W. Bell, Hoe Ed Khor, and Kalpana Kochar, China at the Threshold of a Market Economy (IMF, 1993).
[14] Allan Zhang, Hidden Dragon: Unleashing China's Private Sector (PriceWaterhouseCoopers, 2003); available at http://www.pwcglobal.com/extweb/newcolth.nsf/docid/3D15C57A6D220BB985256CF6007B9607.
[15] See Sparklers from the PRC, 56 Fed. Reg. 20588 (May 6, 1991) (final LTFV determination).
[16] Preliminary Determination of Sales at Less than Fair Value, Certain Malleable Iron Pipe Fittings from the PRC, 68 Fed. Reg. 33911, 33913-33915 (June 6, 2003). The preliminary determination of separate rates was unchallenged and unchanged in the Final Determination, 68 Fed. Reg. 61395 (October 28, 2003).
[17] See Richard Spencer, China amends constitution to protect private property, The Age, Dec. 24, 2003; available at http://www.theage.com.au/articles/2003/12/23/1071941725876.html.
[18] See Website of the Embassy of the People's Republic of China in the United States of America, http://www.china-embassy.org/eng/gyzg/t57116.htm.
[19] Protocol of Accession of the Peoples Republic of China, WT/L/432 (23 November 2001) at Item 10.3.
[20] Protocol of Accession of the Peoples Republic of China, WT/L/432 (23 November 2001) at Item 15.
[21] Report of the Working Party on the Accession of China, WT/MIN(01)/3 (10 November 2001) at para. 173.
[22] Id.
[23] Id. at para. 166.
[24] In the 1984 Wire Rod cases, Commerce said that few modern economies are purely market driven. Carbon Steel Wire Rod from Czechoslovakia, 49 Fed. Reg. at 19371; Carbon Steel Wire Rod from Poland, 49 Fed. Reg. at 19375.
[25] Rust v. Sullivan, 500 U.S. 173, 186 (1991).
[26] Chevron U.S.A. Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837 (1984).
[27] Rust v. Sullivan, 500 U.S. at 186-87 (citations omitted).