Policy, Research, and External Affairs WORKING PAPERS Trade Policy Country Ecomonics Department The World Bank December 1990 WPS 551 Australia's Antidumping Experience Gary Banks The Australian experience suggests that antidumping is, at heart. about safeguarding the interests of particular industries. As long as this is true, there will always be tension between antidumping policy and the broader interests of the national economy and the world trading systems. ThePolicy, Research, and Extemal Affairs Complex distibutes PRE Working Papers tndisseminate the f dins of work in progress and to encourage the exchange of ideas among Bank staff and all others interested in development issues. These papers carry the names of the authots, reflect only their views, and should be used and cited accordingly. The findings, interpretations, and conclusions are the authors' own. They should not be auributed to the World Bank, its Board of Directors, its management, or any of its member countries. Policy, Research, and External Affairs Trade Policy WPS 551 This paper - a product of the Trade- Policy Division, Country Economics Department - is part of a larger effort in PRE to understand the economics of the emergence of "faimess" as a standard for regulating intemational trade, its implications for the continued openness of the intemational trading system, and its continued functioning as an important vehicle for development. Copies are available free from the WorldBank. 1818 H StreetNW, Washington DC 20433. Pleasecontact Nellie Artis, room N10- 013, extension 37947 (58 pages. with figures and tables). One side of the debate on antidumping argues that dumping process in Australia. In particuiar it has tightened up on is not a problem in intemational trade - that it is a normal how the injury test is implemented and has given a second business practice that benefits the importing country's pass at "normal value" arithmetic. Several times its consumers and user industries - and that antidumping is assessments have diffcred from those of Customs - in a inherently protectionist. direction more sympathetic to the foreign importer. But The other side argues that an anti-dumping system has that may have more to do with the govenmment's current a legitimate role to play in maintaining a liberal trading attitude than with the institutional innovation itself. order, but that the process is being abused for protectionist The ADA operates within the same industry-specific ends. framework as the Customs Service but is more attuned to Gary Banks uses Australia's experience in the last the political environment in which technical decisions are decade to shine light on the issue. made. But that can cut both ways. Antidumping is a complex process with many rules The antidumping system retains a degree of adminis- that, depending on interpretation or minor changes, can trative or ministerial discretion that will always make it have important effects on the fortunes of home industries. vulnerable to the business and political cycles. That is true This means that the system and how it operates will remain whatever the country. The Australian government has at of abiding interest to import-competing industries. Lobby- least wrestled publicly with the issues to arrive at a more ing for rule changes or favorable interpretations will precise, objective position. continue as long as the expected retums from such lobbying But on the three touchstones of the antidumping exceed the costs. Whether or not antidumping serves as a process - defrnitions of normal values, material injury, and protectionist device in Australia, says Banks, industry sees causality - arbitrary judgments will always need to be it in that role. made at many points. These judgments will inevitably be It is difficult for industries to get the conventional colored by the political and economic climate of the day. border protection that was common 10 years ago. Under Compared with the early 1 980s, says Banks, that climate is trade liberalization, "protection" has become a discredited currently relatively "dry" - but that can change again. concept in Australia. But in the United States and else- The evidence from Australia's experience, says Banks, where, "faimess" is always popular - and antidumping suggests that dumping may be a problem in intemational (the very term) is seen as being about achieving faimess. trade but that antidumping presents even greater problems. It is the "low-track" route for getting protection against Tinkering with the procedures and criteria for taking imports. It takes place according to rules and procedures antidumping action can help reduce its protectionist that industry and specialist consultants soon master, away tendencies somewhat (though such changes are reversible). from the public glare. More "high-track" routes are more But it does not resolve the fundamental problem that costly and more likely to meet persuasive opposition. antidumping is at heart about safeguarding the interests of The demand for antidumping as a protectionist device particular industries. As long as this remains the objective, will continue and can be expected to rise when times "get there will always be tension between antidumping policy tough," says Banks. and the broader interests of the national economy and the Australia's new Anti-Dumping Authority (ADA) has world trading system. brought a fresh and more critical eye to the antidumping The PRE Working Paper Series disseminates the findings of work under way in the Bank's Policy, Research, and Extemal Affairs Complex. An objective of theseries is to get these findings out quickly, even if presentations are less than fully polished. The findings, interpretations, and conclusions in these papers do not necessarily represent official Bank policy. Produced by the PRE Dissemination Centcr CONTENTS I INTRODUCTION 1 II A BRIEF HISTORY 4 - Take off: the early 1980s 4 - The Gruen Review 8 - Legislative response 11 - Industry fights back 14 - The ADA review 15 m ANTI-DUMPING IN PRINCIPLE 16 - The economic case 16 - Systemic considerations: 'fairness' 22 Iv ANTI-DUMPING IN PRACIICE 28 - The pattem and extent of anti-dumping 28 - Rules and procedures 41 V CONCLUDING REMARKS 54 REFERENCE LIST 57 BOXES Box 1 The IAC - a unique institution 5 Box 2 Would dumping by any other name smell as bad? 23 Box 3 The cherries-in-brine case 45 FIGURES Figure 1 Take off in anti-dumping 6 Figure 2 The steps in the anti-dumping process 13 Figure 3 Australia's anti-dumping compared 29 Figure 4 The rise and fall of anti-dumping 30 Figure 5 Competitiveness and the demand for anti-dumping 32 Figure 6 Acceptance rate of anti-dumping petitions 34 TABLES Table 1 Duration of anti-dumping actions 19 Table 2 International comparisons of anti-dumping 29 Table 3 Anti-dumping stocks and flows 31 Table 4 Country incidence of Australian anti-dumping 36 Table 5 Incidence of anti-dumping by industry 38 Table 6 Dumping margins 40 Table 7 Methods used to determine normal value 44 Table 8 Anti-dumping measures applied 50 ACKNOWLEDGEMENTS In preparing this report the author has benefited greatly from discussions with officials from the Australian Customs Service, Anti-dumping Authority, Industries Assistance Commission, Department of Industry, Technology and Commerce and Department of Foreign Affairs and Trade. In particular, the author is very grateful for the assistance of the Australian Customs Service in providing statisdcal information used throughout the report Helpful comments on an earlier draft were received from officials in the first three organisations as well as from the OECD secretariat and the World Bank. It should be emphasised, however, that the views contained in the report are the responsibility of the author alone. 1. INTRODUCTION In 1923, Jacob Viner, the eminent American economist, published a book called Dumping: A Problem in International Trade. Viner defined dumping as 'sales for export at lower prices than those charged at the same time and under like circumstances to buyers for the domestic market', and his book was concerned with the circumstances in which dumping was likely to take place and be harmful or beneficial to the importing country. At the time when Viner was addressing the topic, there had been some experience of dumping in world trade, but very little experience with anti-dumping action. That situation has changed. Indeed, given the concerns now being raised in national and international forums, it is likely that if Professor Viner were re-writing his book today he would have substituted anti-dumping for dumping in its title. All industrial countries, and an increasing number of developing countries, have anti- dumping legislation. In most cases that legislation is based on internationally agreed rules and procedures under the General Agreement on Tariffs and Trade (GATT), rules which were themselves initially based on US legislation from the 1920s. Dkimping, unlike the government subsidisation of exports, is an activity initiated by private firms and is not 'illegal' in GATT terms. Article VI of the General Agreement - the original anti-dumping text, to which a Code was added in 1967 - states that 'dumping ... is to be condemned if it causes or threatens material injury' to an industry in the importing country. There is thus an ambivalence in GATT rules on (anti) dumping which does not arise in most other areas. That ambivalence reflects (a) recognition that dumping is a source of cheaper imports, which are normally beneficial to the importiag economy, and (b) an awareness that anti-dumping action has the potential to be an alternative means of protecting industries against (legitimate) world market competition. The GAIT rules on anti-dumping consist mainly of procedural and other constraints designed to ensure that any action is justified and seen to be so. (Though there is nothing in the GATT rules which requires anti-dumping action to be taken when the criteria are met.) T'here are essentially three hurdles which the GA t .1 greements place before a 2 government contemplating anti-dumping action: first, 'dumping' - sales at less than 'normal value'- must be found to have taken place; second, a domestic industry must be shown to have suffered 'material injury'; and third, that injury must be found to be attributable to the dumping in question. To each of these hurdles is appended a myriad of interpretative definitions and criteria - not least to do with the meaning of the key terms 'normal value', 'material injury' and the establishment of 'causality'. The Anti-dumping Code of 1967 was largely a 'ned at tying down these concepts and tightening procedures, out of concem that the looseness of the original GATT requirements was allowing anti-dumping to develop as a protectionist device. The pr,sent debate about anti-dumping can be divided into two. rFirst, there are those who argue that dumping is not a problem in intemational trade - that it is a normal business practice which benefits the importing country's consumers and user industries - and that anti-dumping is inherently protectionist. * S.econd, there are those who believe that an anti-dumping system has a legitimate role to play in maintaining a liberal trading order, but that the process is being abused for protectionist ends. Three sets of conclusions about reforms to anti-dumping have emerged from this dual debate: - tighten anti-dumping procedures to reduce scope for abuse - broaden their focus, to take into account wider economic interests - abandon anti-dumping altogether. The purpose of this paper is to use the Australian experience over the past decade - a period rich in anti-dumping activity under different procedures and institutions - to shine some light on the issues raised internationally. (Countervailing duty procedures, which are similar to those for anti-dumping, are not dealt with here.) It is not a comparative study, though at times international comparisons will be made. 3 The paper begins by briefly outlining the varied history of Australia's anti-dumping anrangements and activity since the late 1970s. It then looks at the Australian experience from the perspectives of the debate just referred to and ends with some conclusions relevant to that debate. 4 II. A BRIEF HISTORY Anti-dumping provisions were first contained in Australian legislation in 1901, at the time of federation. In 1906, measures were enacted to address 'predatory dumping'. but in 1921 the Customs Tariff (Industries Preservation) Act established broader procedures for the imposition of penalty duties on imports deemed to have been sold at prices lower than in their suppliers' home markets. This legislation was similar to that enacted in Canada and the United States (Dale, 1980). In the Australian system, Parliament delegates the power to take anti-dumping action to the Minister with responsibility for customs matters. Present legislation is a much amended form of the Customs Tariff (Anti-dumping) Act of 1975, which was introduced to implement the GATT Anti-dumping Code of 1967 and it displaced most preceding anti-dumping legislation. A distinctive feature of the Australian procedures until 1984, was provision for appeal to the Industries Assistance Commission ([AC) - a statutory authority of the Australi.an Govemment with the role of reporting on and advising government about industry assistance from an economy-wide perspective (box 1). The 1975 Act appears to have had little effect on the incidence or extent of anti-dumping actions, which remained steRdy at levels comparable to those in earlier years. In the words of Gruen (1986), 'prior to the 1980s the dumping arena in Australia was relatively quiet and noni-controversial' (p.8). Take-off: the early 1980s In the early 1980s, anti-dumping activity increased rapidly (see figure 1). This reflected a combination of factors (discussed in chapter IV) among which the global and national recessions, combined with a rising Australian dollar, and the increased difficulty of obtaining industry assistance through conventional means were important. Pressure mounted for the government to take tougher r stion against low priced imports. That pressure led to some significant legislative changes in the early 1980s. 5 Box 1: The IAC: A unique institution A sidelight to the anti-dumping story in Australia is the role played by the Industries Assistance Commission (AC) in the anti-dumping process. The IAC was created by Act of Parliament in 1973 to do something which at the time was novel in the Australian industry policy environment (and indeed that of any other country), namely: * to advise the government, in publicly available reports, about the effects on national economic welfare of any proposals to assist industry; and * to inform the general community (consumers, enterprises, the media) about the existence and economic effects of all measures in place. The IAC was carefully designed to ensure that its advice was impartial and of high quality. And the IAC has played an important role in paving the way for trade liberalisation and microeconomnic reform in Australia. In addition to being independent and open in its operations, a unique feature of the IAC was the set of legislative guidelines for its advice. These were: 'improving the efficiency with which the economy uses its resources; ensuring a consistent ind1istry policy; taking account of consumers and users of products affected by the Commission's proposals, and providing for public scrutiny of assistance measures' (Rattigan, 1988, p. 13). These economy-wide criteria for evaluating industry protection created a point of tension, however, in the IAC's handling of anti-dumping qiuestions and its involvement in anti- dumping has been progressively reduced over the ycr..s: * At first, the IAC conducted, as a matter of course, investigations into whether anti- dumping criteria were met. * In 1975 it lost this central role and instead became an appeal body on anti-dumping matters. * In 1984, the IAC Act was amended to provide that IAC judgments in such appeals were to be confined to 'the facts' - the Commission was specifically instructed to ignore economy-wide considerations. * Finally, in 1988 the IAC lost its 'court of appeal' function and was withdrawn altogether from any formal involvement in the anti-dumping process. The IACs disconnection from anti-dumping system underlines the fact that anti-dumping occupies a special place in industry protection policy. It has been and remains concemed with industry specific injury and takes no account of the effects on consumers and industry generally of remedying that injury through higher import duties. 6 Figure 1: The anti-dumping take-off 200 Measures In placo 160 -.- 140 , 120 , 100~~~~~~~~~~1 100 ,,- 80 - Cases initiated 60- 00 4 , - " o/' Action taken 20 1979 1980 1981 1982 1983 1984 Source: Australian Customs Service. Legislative changes In 1982-83 there were a number of amendments to the Act which facilitated the use of anti-dumping action. These included: - the removal of some appeal provisions (relating to t; I imposition of cash securities) and the streamlining of others (access to the Federal Court), - the removal of section 14 of the Act, which obliged the Minister not to take action 'inconsistent with Australia's intemational obligations', - and, consistent with the Tokyo Round revisions to the Code, removal of the requirement for dumped imports to be the 'principal' cause of material injury. Then, following a change in government in 1983, the incoming Minister with responsibility for anti-dumping ordered a departmental review of the process, observing 7 that 'the Government is determined to ensure that the protection afforded industry is not undermined' (IAC 1984). The main areas of legislative change in 1984 related to: - Sales at a loss: The administration had adopted the practice of excluding prices charged by exporters in their domestic markets when those sales were made at a loss, for the purpose of determining normal values (of which more later). This was regularised by additions to the legislation (section 5(9)). - Material injury: An indicative list of elements for judging material injury was added to the legislation. - IAC reviews: The IAC was relieved of its statutory obligation, when reviewing anti- dumping mattxrs, to refer to its economy-wide policy guidelines. The amended IAC legisla.don provided that such reviews simply be confined to 'the facts'. - Monitoring: Administrative changes were also mad_. to expedite and facilitate the anti-dumping process. These changes were interpreted by a number of A ustralia's trading partners as making the anti-durnping process a more protectionist instrument. (GATT, 1984) Domestic conflict Within Australia, too, concerns were being voiced about the direction taken by the anti- dumping process. In a 1985 draft report on assistance to the chemicals and plastics industries, the IAC described ways in which anti-dumping procedures were imposing costs on user industries and the economy generally, noting the high incidence and country coverage of anti-dumping action in that sector. The IAC warned of the potential for the anti-dumping system to countermand the government's goal of reducing industry protection and recommended that any future decisions on anti-dumping for the industry should reflect national interest considerations. The capacity of anti-dumping action to impose costs on user industries was forcefully demonstrated a short time after by a case which triggered a new review of anti-dumping and esntually resulted in the revised scheme which operates i.day. That case involved 8 certain fertilisers - again part of the chemicals sector. The sequence of that dispute was as follows: * Following a long overdue review, anti-dumping actions on nitrogenous fertilisers were revoked in April 1984, benefiting the fanming sector. * In October 1985, the IAC recommended the termination of fertiliser subsidies to farmners - subsequently accepted by government. * But in December 1985, dumping duties were re-imposed on certain fertdlisers. * A severe political backlash forced the government to provide farmers with temporary payments to offset the cost burden caused by the imposition of anti-dumping duties, until reviews of the industry and the whole anti-dumping process could be conducted. The Gruen Review In February 1986, Fred Gruen, a Professor of Economics at the Australian National University, was appointed by the government to undertake that review, with terms of reference requiring him to report on: * whether, in the light of the GATT Code and other relevant agreements, improvements to the Act could be made and, if so, to recommend appropriate changes; * whether 'consideration should be given to the merits of including a National Interest provision as a condition for the purposes of decisions to impose duties...' (Gruen 1986, p.A3). That review had considerable influence on subsequent changes to anti-dumping procedures and on perceptions about how the scheme had operated in the past, and is therefore worth looking at in some detail. 9 Gruen, aided by a departmental secretariat, received some 130 submissions from industry associations, trade unions, private firms and government authorities, as well as from foreign governments. Meetings were held throughout the country. The Review was prepared against a tight deadline and released in April 1986. It focused on the operation of the anti-dumping process in Australia since 1980. Gruen was critical of a number of ways in which the scheme had operated during that period. In a summary to his report he began by observing that: Australia makes greater use of anti-dumping action than do other comparable countries. Such extensive use of anti-dumping action has the potential to frustrate the achievement of other govemment objectives in the industry, trade, competition and economic policy areas. (p.iii) He observed that it is 'normally in the importing country's overall economic interest to take the external trading environment as given and accept cheap imports, even if they are dumped or subsidised'. However, he also concluded that the principle of 'fairness' in international trade, which stands behind anti-dumping policies, is widely supported throughout the Australian community. And he noted that: [The] commitment to a speedy and readily available anti-dumping system underpins the existing fragile consensus between Govemrnment, industry and unions on the need for change towards a less assisted, more outward-looking, restructured industry. Gruen therefore recommended that the existing system be continued, subject to some substantial changes designed to: (a) reduce the discrepancy between the concept of 'unfair trading practices' as it is applied within Australia and as it is applied by Australia to its imports of goods; and (b) discourage too extensive use of the anti-dumping system as a more readily available system for restricting imports. (p.iv) Gruen's recommendations were largely based on his investigations into three key issues: the assessment of 'fair prices'; material injury and the causal link with dumped imports; and the question of a 'national interest' provision. 10 Assessment of 'fair prices' Gruen was critical of Australia's increased propensity, shared with the other main anti- dumping countries, to use cost estimates instead of market prices for determining normal values. In particular, he criticised the practice of rejecting exporters' local prices where they do not cover all costs - resulting in 'declared normal prices which can be substantially above the actual domestic market price'. He therefore recomniended the repeal of section 5(9) of the Act, concerning sales at a loss, and urged that where no domestic price data were available, representative prices for comparable goods sold to third markets should be used, with constructed values used 'only when there is no conceivable alternative'. Material injury Gruen observed that, while the tests for material injury in anti-dumping cases in Australia are equivalent to those used in assessing cases for emergency assistance (based on GALT Article XIX), the 'Customs Service is prepared to reach a positive preliminary finding in a dumping case on evidence of lesser injury than is the Department in assessing a prima facie case for emergency assistance'. He concluded: 'At present the injury test is operated on the basis that virtually any injury caused by dumping is unacceptable' (p.30). He recommended in his report that: - price effects by themselves should not be sufficient evidence of injury; - a 'substantial effort' should be made to allow for the influence of factors other than dumping in causing injury; and - injury must be seen in the context of the 'entire operations of the relevant establishment' (p.32). A 'national interest' provision As indicated previously, the Gruen review was precipitated by conflict between producers seeking action against (dumped) imports on the one side and the users of the products concerned on the other - and by general dissatisfaction with procedures which only took 11 the interests of the former group into account. The IAC had argued in its Chemicals report that anti-dumping action should only be taken when it is in the 'national interest'. The government had expressly asked Professor Gruen whether such a provision should be included in the legislation. Gruen decided, largely on 'practicality' grounds, that it should not: The addition of a 'national interest' clause must add to the uncertainty of the proceedings and to administrative complexity and would increase costs of investigation to both the participants and the Government. It would also expose the Minister - and the Department - to intensive lobbying on individual dumping decisions. (p.36) Legislative response to the Gruen Report In announcing its decision on the Gruen Report in October 1986, the government said that if Australia was to rationalise protection it was essential that manufacturers were not subject to injury from products imported at prices below those at which they are sold in the country of origin. The Government further promised that it was not prepared to provide Australian industry with 'a lesser safeguard against unfair competition' (emphasis added) than that provided by the US, Canada and the EC. However, it also implied that in the past the anti-dumping process had exceeded its charter and expressed a new determination that anti-dumping not be used as an alternative form of industry protection. The government's response was implemented through a package of legislation - involving amendments to existing legislation and a new Act - which made the following changes to the anti-dumping system. Creation of an Anti-dumping Authority This, one of the major changes, was not proposed or even discussed in the Gruen Report. The Anti-dumping Authority (ADA) was established by separate legislation as a new, independent authority with the functions of: - recommending to the Minister whether anti-dumping or countervailing duties should be imposed; and 12 - providing the Minister with advice on anti-dumping issues in general. The first of these roles (and implicitly the second) was previously assigned to the Australian Customs Service (ACS), with provision for review by the IAC. Apart from administering anti-dumping measures, the ACS's role is now confined to obtaining preliminary findings; after that it may be called upon to assist the ADA in obtaining information needed by that body. And, as noted previously, the IACs formal role in the anti-dumping process was terninated. (The current procedural steps are set out in figure 2) The Act provides for the responsible Minister to issue directions to the ADA, with the aim of ensuring that the Authority 'is guided by the current industry policy objectives of the governmentf in its interpretation of the legislation. The government made it clear that tests for demonstrating 'material injury' at-d causality would be tightened through such guidelines. Sales at a loss and constructed normal values The government, while rejecting Gruen's recommendation that section 5(9) be removed (relating to the invalidity of normal values based on domestic prices where sales are consistently made at a loss) altered the legislation to the effect that in this and other circumstances, constructed normal values should include a profit only in circumstances that were subsequently to be specified in regulations - with the presumption that otherwise a zero profit was to be imputed. National interest The government rejected the inclusicn of a national interest provision in the legislation for much the same reasons as Professor Gruen. ('Proceedings could become unnecessarily complex and protracted.') However, it made it explicit that the Minister had the power to take national interest criteria into account in exercising his discretion when considering ADA reports. 13 Sunset clause The government introduced a three year sunset clause, 'consistent with its intention that anti-dumping and countervailing measures do not become a substitute form of ongoing assistance' (Jones 1988, p.2312). This applies to actions against commodities, regardless of the time profile of actions against the particular exporters involved. Figure 2: Steps in the anti-dumping process Applicatdon received by Customs |Application | | Application| rejeted rnma)dmu properly -inscomplete documented | AppIication A| Investigation | rejected 1 adu ) ititiated -no pruns facie case|w -pnnuA fade Mae PrelEminary <>Undertkng findin8g 2dy accepted by Customsaximu =ig , ~~~~~Application Appeal to terminated Finding positive ADA - either no dumping, provisonal for review no muterial injury mees if jusified or no causl link Fu _ fidng(l20 daysA a_te by ADA ~ alum , -nceig [ Application |r Appeal to | temninated | Finding l federal court * -ithe no dumping, 46- positive l I no mate"g iJryM I duties levied I | or no causal liHk& 14 IAC's role eliminated The IAC Act was amended to remove that body as a place of review on dumping matters. That function is now performed by the ADA itself. Lodging a complaint The new legislation contains specific provisions to make it clear that an anti-dumping complaint may be made by any person on behalf of the Australian industry affected (or likely to be affected). Statutory time limits For the first time, specific time limits were provided for the different stages of the anti- dumping process: 55 days for establishment of a prima facie case; 120 days to make a preliminary finding and, in the event of a positive preliminary finding, 120 days for a final finding. Industry fights back Manufacturing interests wasted little time in showing their dissatisfaction with the 1988 legislation. The Australian Chamber of Manufactures (ACM) soon formed an Anti- dumping Task Force drawn from a number of national industry associations. The ACM put its case in terms of what it saw as 'fewer safeguards' in the new anti-dumping system than existed overseas. The Task Force argued that there should be another review of the legislation and that, in the meantime, the government should consult with industry 'to ensure the current legislation operates effectively' (Stubbs 1989). The ACM further demanded that: Regulations and directions should be made iumediatey to:- Expunge any thought, as expressed in the Minister's second reading speech, that industry needs to be threatened with being 'knocked out' before anti-dumping measures are applied. 15 * Include a profit component [not less than 8 per cent] when constructed nonnal values are used for fair prices. * Provide interim action after 55 days [instead of the 180 days in legislation]. * Remove Ministerial discretion [to reject or amend an ADA recommendation on 'national interest' grounds] (Stubbs 1989). The last two demands have fallen on deaf ears thus far (although a review of the revised anti-dumping will be undertaken sometime after September 1991). The other demands, however, were considered in the context of a November 1988 report to the Minister by the Anti-dumping Authority. The ADA review of three key issues The Minister sought the ADA's advice about what administrative guidelines should be given on three aspects of anti-dumping procedure/criteria which had not been precisely spelt out in the legislation: (i) what constitutes 'material injury'; (ii) what is the appropriate 'extended period of time' beyond which domestic sales at a loss cannot be used for normal values; and (iii) when a profit element should be included in a constructed normal value. These matters are discussed in some detail in a later chapter. Suffice it here to note that the ADA recommendations brought greater scope for taking anti-dumping action than the Minister's statements had previously indicated would apply (especially in relation to the inclusion of an imputed profit in constructed normal values, and the injury test). The changes, reflected in regulations and legislation in October and December 1989, appear to have been accepted by the Australian Chamber of Manufactures for the time being. 16 111. ANTI-DUMPING IN PRINCIPLE A fundamental question is whether anti-dumping action is ever justifiable in a liberal trading order. A number of economists, as opposed to policy makers and lawyers (and just about everybody else), have argued that it is not; that anti-dumping is inherently protectionist and should be thought of in the same way as other forms of protection - as just another device for assisting domestic industry against competition from imports (see, for example, Stegemann 1980, Finger and Murray 1989). If we took the definition of protection as 'barriers to imports which assist a local industry to maintain a level of activity which it could not otherwise sustain', then anti-dumping action obviously could be classified as 'protectionist. On the other hand, if we think itt terms of an international trading system in which GATT has set the rules about what is and is not permissible, then the fact that anti-dumping is permitted would mean that it cannot be protectionist by definition. Such semantics do not get us very far. The real question is whether dumping is harmful to national economies - and the intemational economy - in the long term. If it is, the case for an anti-dumping system then depends on whether, :ch a system can improve the situation. This chapter addresses the first issue, the next chapter addresses the second. The question of what 'harm' dumping does, can be looked at from two perspectives- the (narrowly) economic and what might be called the 'systemic'. The economic case Considering first the dumping exporter's home country, it is sometimes suggested that dumping in itself imposes costs on the local economy. However, this confuses dumping with subsidised exports. The latter, if they involve net subsidies - and putting to one side 'strategic trade' issues - generally do impose costs by distorting the allocation of resources in the exporting country. However, dumping is quite different in nature from subsidised export pricing. It is done by a private firm as part of a profit maximising (or 17 loss minimising) strategy and, assuming that private firms know what is in their own best interests - and that their decisions are not distorted by govemment interventions - it will also be in the interests of the dumping firm's economy. (In practice, 'dumping' is commonly a side-effect of domestic interventions which serve to protect the exporter's domestic market from world competition, keeping home prices high - but the reai source of economic costs are the interventions, not dumping as such.) The question of costs is usually focused on the recipient country. Normally, lower priced imports benefit an economy - allowing consumption gains from lower prices, and the reallocation of resources into areas of comparative advantage. In other words, they increase the 'gains from trade'. The theoretical literature on dumping, which began with Viner, has identified two circumstances in which countries may not gain from lower priced imports: 'predatory' pricing and intermittent dumping. Predatory pricing One possibility, often raised by industry in support of anti-dumping measures, is that dumping may be part of a foreign supplier's strategy to drive local suppliers out of the market; having eliminated the competition, prices could then be raised to monopoly levels. For the foreign exporter, the long term monopolist gains more than make up for the short term dumping losses. For the recipient country, however. the consumers' short term gains from lower prices are eventually outweighed by the losses from higher prices in the long term than would have obtained if the domestic industry had remained in business. While this argument may have some superficial appeal, the conditions required for successful predation are stringent. For example, there must be only one foreign supplier (a world trading monopolist or sustainable cartel); otherwise any attempt to raise prices after the local competitor's demise will attract other import competition. And, in the same way, there would need to be good reasons why the locals themselves could not profitably re-enter the market once prices again reached levels at which domestic supply had been profitable before. And even where the appropriate conditions for successful predation exist, there would be other less costly strategies available to foreign firms than a drawn- out price war (such as collusion with, or acquisition of, the local supplier). 18 The theoretical esotericism of predatory dumping seems confinned by the rarity of such cases in practice. (Dale, 1980) In Australia, with all the dumping activity that has been officially detected, there appears to be no evidence that any of this was part of a predatory strategy. (Gruen comments at one point that 'There have been instances where exporters' pricing decisions in Australia appear to reflect a predatory element', but this is not documented.) The fact that Australia's imports of nearly all commodities come from a range of suppliers in different countries eliminates a fundamental condition for the rational pursuit of such a strategy. Moreover, a substantial number of anti-dumping actions themselves involve more than one supplier. For example, of the 47 anti-dumping actions that were brought to an end in 1988-89 (mostly under the new sunset clause), 37 involved suppliers from more than one foreign country. The view that predation is not seriously seen as the raison d'etre for anti-dumping action is also supported by the absence from Australia's anti-dumping laws (and from GATT rules) of any requirement to prove or disprove predatory intent. Intermittent dumping Of greater relevance is the argument that dumping may bring net losses to a country because of its internittent nature. Viner (1923) identified a number of circumstances in which this might occur. The argument is essentially that foreign companies may sometimes treat their export sales as a surplus disposal medium. It is suggested that for a small economy like Australia's, even a one-off dump could wipe out an industry, or major sections of it. Like the case of trade liberalisation, the net costs or benefits to an economy from intermittent dumping depend on a comparison of the economy-wide benefits from lower prices and the industry-specilic production losses. The difference is in the temporary nature of the price reductions under intermittent dumping. The more temporary the price reduction, the greater the possibility that adjustment costs could exceed the consumption 19 gains. However, it is also likely that any reasonably competitive domestic firm could weather such temporary phenomena. In practice it is difficult to devise a scheme which could overcome intermittent dumping. While GATT rules and Australian iegislation contain a prospective element ('causes or threatens material injury') it would be difficult in most circumstances for a domestic industry to know that a one-off shipment at dumped prices was happening in advance of it actually arriving. And the lags in getting to a preliminary finding would mean that the injury may well have already occurred before any action could be taken. It would in any case be very difficult for anti-dumping administrators to determine the likely duration of dumping. The Australian legislation (again consistent with the GAIT) does not require any evidence about the duration of dumping or the reason why it has occurred. If anything, it is geared toward finding and dealing with dumping of a more continuous kind. This is illustrated by table 1, which shows that the average anti-dumping action, prior to the introduction of the three year sunset clause, lasted around four and a half years - with the longest actions attaining a duration of close to ten y' tars. (Though, as discussed later, this may also have resulted from an inadequate review process.) Table 1: Duration of anti-dumping actions Number of Year actions terminated Average duration (years) Longest duration (years) 1984-85 13 4.8 7.2 1985-86 50 4.3 9.6 1986-87 60 3.6 8.3 1987-88 66 5.3 8.8 1988-89 43 5.2 9.7 Source: Austualian Customs Service. While there may be a theoretical case that in certain circumstances intermittent or temporary dumping could reduce the welfare of the importing country, there is no such case where dumping is a continuing phenomenon. Long term dumping brings gains to consumers and user industries which could be expected to outweigh any losses in 20 production within the import-competing industry. Given the relatively lengthy application of anti-dumping duties in Australia (even the three year sunset clause goes beyond the sort of temporary dumping that might impose net costs) it might have been expected that doubts would be raised in this country about whether anti-dumping action has been in the 'national interest'. Economy-wide effects Recall that the Gruen Review was triggered by an acute awareness that anti-dumping action against imported fertilisers, while undoubtedly benefiting the local fertiliser industry, was imposing substantial costs on the much larger and economically more important farm sector. But Australia's anti-dumping procedures, like those of other countries, exclude any detailed consideration of the effects that dumping and anti- dumping have on the broader community. Concern is focused solely on the question of injury to a particular industry. The larger question of whether the avoidance of industry- specific injury might give rise to a greater injury to the economy as a whole is not considered. The questioli of whether dumping reduces the national welfare of the importing country requires a wider investigation of effects than just on the industry directly involved. Yet the Australian legislation - in common with that of other countries and consistent with the explicit intentions of the GATT - makes no provision for such an economy-wide perspective. In a submission to the Gruen Review, the Chairman of the Industries Assistance Commission drew attention to this anomaly: The benefits of anti-dumping action accrue to import-competing producers (including their employees, and dependent input suppliers) while the direct costs are incurred by domestic user industries and consumers. To serve community-wide interests, as opposed to the interests of a particular group, it would be necessary for provision to be made to take account of the competing interests of these groups. (Cannichael 1986, p.14) 21 The Chairman of the IAC proposed two alternative changes to administrative arrangements which would facilitate this broader 'national interest' perspective: * Require Customs to make recommendations to the Minister based upon broadened criteria, with provision for appeal to the IAC on all grounds including consideration of the wider interests. * Require Customs to determine whether a prima facie case exists with reference to broadened criteria and refer such cases to an independent authority for assessment. (p.17) Gruen included the second proposal among four alternative ways of allowing for 'national interest' considerations in anti-dumping - which he included in case the government decided not to accept his recommendation against including national interest provisions at all. In the event, the government accepted his overriding recommendation and the 1988 legislative amendments made no explicit changes to broaden the criteria for taking anti-dumping action. What the government did do was create the Anti-dumping Authority, as described previously, which it saw as bringing 'a broader perspective to the determination of such issues as material injury and causal link'. However, there are no guidelines for such a broader perspective in the Anti-dumping Authority Act, and in practice the ADA confines its perspective to the same legislative criteria that guide the determinations of Customs. Within those constraints the ADA's judgments have on occasion differed from those of Customs, but not as a result of considering economy-wide effects. Nevertheless, there is nothing in the Act which would prevent the ADA from recommending to the Minister that anti-dumping action not be taken, even where there are positive findings of dulaping, injury and causation. However, according to ADA officials, this is only likely to arise in an 'extreme' situation and no cases have yet occurred. 22 Systemic considerations: 'fairness' No lesser economist than Jagdish Bhagwati has given his support to anti-dumping procedures on the following grounds: An economist is right to claim that, if foreign governments subsidize their exports, this is simply marvellous for his own country, which then gets cheaper goods and thus should unilaterally maintain a policy of free trade. He must, how-ver, recognize that the acceptance of this position will fuel demands for protection and impcril the possibility of maintaining the legitimacy, and hence the continuadon, of free trade. A free trade regime that does not rein in or seek to regulate artificial subventions will likely help trigger its own demise. An analogy that I used to illustrate this 'systemic' implication of the unilateralist position in conversing with Milton Friedman on his celebrated Free to Choose television series is perhaps apt: Would one be wise to receive stolen property simply because it is cheaper, or would one rather vote to prohibit such transactions because of their systemic consequences? This line of thought supports the cosmopolitan economist's position that the world trading ordcr ought to reflect the essence of the principle of free trade for all - for example, by permitting the appropriate use of countervailing duties and anti-dwnping actions to maintain fair, competitive trade. (Bhagwati 1988, p.35, emphasis added) This 'systemic' rationale stems from the notion of an international economic order in which countries' actions are interdependent in their effects on economic welfare. Thus if there is a widespread view that certain trade actions are unfair, even if they are beneficial to the community as a whole, then failure to reflect that view in the rules of the game may erode support for the game itself. It is reasonable to question, however, whether Bhagwati is right to lump both anti- dumping and countervailing duties into the same 'system preserving' bag. The notion of fairness is an amorphous thing, but government intervention to help a national exporter compete against a foreign firm would seem to have more claim to being seen as unfair than the case of a firm merely selling at lower cost in a foreign market than at home- especially if, as we have argued, its actions are not based on predatory intent. The theft analogy, while applicable to subsidisation, also seems inappropriate for dumping. A more apt question might be: would one be wise to accept goods that are 23 being given away...? It is not apparent that this question should also be answered in the negative. Box 2: Would 'dumping' by any other name smell as bad? Language can exert a powerful influence on perceptions about what is right or wrong and what needs to be done. In some cases the words we use can constrain a d0ate to the point of virtually preempting the conclusion. For example, trade policy discussion in recent years has made liberal use of analogies with military conflict. If trade is war, then it is only natural to defend one's country from the 'invader'. It was not by accident that Henry Ford III, in pushing for protection for the automobile industry in the early 1980s, chose to describe competition from Japan as 'an economic Pearl Harbor. So too for dumping. Imports that have been 'dumped' are undesirable by definition: implying something that is not only 'unfair' but offensive. It would take a fearless or foolish person (or an economist) to stand up and say that 'dumped imports are good for us' - especially when they cause 'injury' to a local industry. But to say that 'imports priced at less than what foreigners have to pay are good for us' sounds fine. Yet the two statements are equivalent. Conclusion: As long as low priced exports are called dumping they will be popularly regarded as unfair, their perpetrators deserving of every bit of punislhment that the importing country meters out. Fairness 'down under' In the Gruen Review some attention was given to the question of fairness which is said to underpin Australia's anti-dumping laws. One of Australia's largest retail outlets - a major importer - expressed 'complete agreement with the [anti-dumping] principles espoused by the GATT ...' And the research bureau of the Department of Industry, Technology and Commerce has argued that 'notions of fairness in the setting of export prices are deeply entrenched in the Australian community' (Gruen 1986, p.24). 24 Gruen observed that: In the domestic context both predatory pricing and price discrimination are banned under the Trade Practices Act 1974 (Sections 46 and 49). There would seem some logic in applying the same type of rules to foreign suppliers. (Gruen,1986, p.25) Nevertheless the 'fairness' justification for Australia's anti-dumping legislation as it stands is not as cut and dried as it may seem. For one thing, as already noted, the rules are not directed at predatory pricing as such. And domestic rules on price discrimination are not as readily translated into anti-dumping law as Gruen suggests. In the domestic environment, the obvious inequity is between national consumers who pay different amounts for the same product. As Dale (1980, p.35) has said, it may be argued that 'the ideal of equality of opportunity lies at the root of anti-price discrimination laws in the domestic market'. But dumping does not breach this ideal in the domestic market. Moreover, since the early 1980s Australia's anti-dumping procedures have also been targeted at more than the simple price discrimination to which national trade practices legislation takes exception. As Gruen observes: 'Selling goods at prices which do not recover full costs ... is not regarded as an unfair trading practice internally. It should not be treated as unfair in international trade' (p.38). In practice, the government appears to see no moral difficulty in Australian firms selling overseas at lower prices than at home. A booklet titled Exports Can Be Highly Profitable, Realistic Costing Shows How, prepared by the Department of Trade and Resources as a guide for exporters, contains the following advice: ... The principle of charging only the additional costs against additional sales is known as 'differential costing', and this is the realistic approach in costing exports. (DTR 1980) As the IAC notes in its submission to Gruen, similar advice for exporters has been given by the Department of Industry, Technology and Commerce. It is not known to what extent Australian exporters have heeded (or needed) this advice, but it is likely to be as common a business practice in this country as it is elsewhere. 25 Thus the question arises as to whether there may be some double standards about what constitutes 'fairness' in export pricing - depending on whether or not it is a competing domestic industry that suffers 'injury'. 'Dumping' as normal market behaviour In many cases, as Viner observed back in 1923, the appearance of dumping or differential pricing will simply reflect a more competitive export market than domestic market in the exporter's country. For example, whenever an exporter is able to benefit from protection in his home market which he does not (cannot) receive internationally, his export price - if he is capable of exporting profitably - must be below the domestic price. Given the relatively high prices which tariff protection provides for much of Australian manufacturing industry in domestic markets, Australia's own manufactured exports might be expected to be typically in this situation. A related case has been documented by the IAC in its report into assistance to the rice industry (IAC 1986). It found that domestic prices for rice ranged from 32 to 116 per cent greater than the industry's export prices over the period 1980-86, largely as a result of statutory marketing arrangements in the domestic market. In such cases, exporters need not be following any predatory or other strategically discriminatory practices; all they need do is be responding to the realities of the export market to be caught for dumping. It seems unfair that this should be labelled unfair. It is commonly argued, however, that a firm's low export prices in this situation are being cross-subsidised by the higher prices which protection or regulation permits in the domestic market. This raises some complex theoretical issues which it is not necessary to explore here. (See for example Dale, 1980, pp. 35-7.) The upshot of that analysis is that the conditions for such cross-subsidisation to be either a feasible or rational strategy are similarly stringent to those for predation. A point that is often missed is that an exporting firm will only derive surplus profits ('economic rent') from protection at home if it has monopoly privileges. It *; in any case doubtful that concern about possible cross- subsidisation is an important factor in the Australian anti-dumping process, given that the anti-dumping laws in practice require no examination of it. 26 The anti-dumping compact While it is questionable whether the Australian community generally regards export pricing at below normal value as unfair, there is little doubt that import-competing industries in Australia are strongly of that belief. And, coming back to the Bhagwati point, the government clearly believes that in the absence of an acceptable anti-dumping mechanism, the liberalisation of (other) barriers to imports in Australia would be politically much more difficult to achieve. As Gruen has put it: There is a fragile consensus in place favouring lower long term levels of assistance and a rationalised outward looking industry structure. Any suggestion that the Government is resiling from its strongly expressed commitment to an effective speedy anti- dumpinglcountervailing system could put at risk the acceptance by manufacturers and unions of the process of restructuring. (p.25, emphasis added) Indeed, there could be said to be an implicit compact between government and industry about the protection/anti-dumping tradeoff. In the case of the chemicals industry - which accounted for 90 per cent of all dumping duties collected in 1986-87 - this link was recently made explicit. The IAC has observed that the release of the govemment's decision on the Gruen Report coincided with the decision to reduce tariff levels for these industries. Senator Button ... noted the concem which the industries had expressed about the recommendation of Professor Gruen's report on anti-dumping measures. The Minister said he considered the combination of the decisions on the IAC's report on Chemicals and Plastics and on Professor Gruen's report would lead to a sound competitive future for these industries. (Ministerial Statement, cited in IAC (1987), emphasis added) This suggests that Australia's anti-dumping process has as much to do with perceptions of political feasibility as with broader and more high minded issues of 'fairness'. A question that might reasonably be asked is whether industry's acceptance of reductions in tariffs reflects its perception that the anti-dumping process is a substitute for other forms of protection. That would explain industry's active lobbying against the key Gruen recommendations. Those recommendations were, after all, designed to 'discourage too 27 extensive use of the anti-dumping system as a more readily available trade regulation system ...' (p.48). If the real raison d'tre of the anti-dumping system is a strategic one, it is also reasonable to question whether it may not end up endangering the broader liberalisation objective that it is meant to serve. Apart from possibly encouraging a diversion of protectionist pressures from more open and accountable forums like the IAC, there is the danger that this sort of 'reputable' protection could be expanded into a comprehensive system of provisions against any trade detmed to be 'unfair', as illustrated by Section 301 and Super 301 provisions of recent United States trade legislation. Once justified in that way, trade barriers become even more difficult to remove or reform than those justified in straightforward protectionist terms. 28 IV. ANTI-DUMPING IN PRACTICE Putting to one side the more fundamental issues about the role of an anti-dumping system, there is increasing concem intemationally that anti-dumping is not being used in the spirit of the GAIT rules - that it is operating in practice as a defacto protectionist device, rather than solely as a means of preventing injury from imports priced below levels in the country of origin. (Hindley, 1988; Bhagwati, 1988) The fact that Australia - along with Canada, the EC and the United States - has been a much greater user of anti-dumping action than other comparable countries, has sometimes been interpreted as indicating that anti-dumping processes in Australia (and the other countries) are biased in favour of such action. This is the issue with which this chapter is largely concerned. It is approached from two (overlapping) perspectives: (a) trends in anti-dumping activity in Australia, and (b) how the rules and procedures operate. The pattern and extent of anti-dumping There is no denying that Australia has been one of the key players in the anti-dumping 'game'. Between 1980-81 and 1988-89 Australia took 218 actions, almost as many as the United States and considerably more than Canada and (possibly) the EC - all countries with much larger import volumes than Australia (table 2). At its peak in 1984, Australia's outstanding anti-dumping actions were equal to one-third of the total actions outstanding as declared by members of the GAIT Anti-dumping Code. However, as is also apparent from the table and from figure 1, if relative anti-dumping activity is in itself taken as an indicator of the possible misuse of a country's anti- dumping procedures, then Australia's record looks a lot better in 1988-89 than it did five years before. 29 Table 2: International comparisons of anti-dumping actionsa Year ended 30 June Australia USA Canada ECb 1981 23 5 15 13 1982 26 48 8 22 1983 57 10 37 33 1984 41 33 13 23 1985 18 28 17 22 1986 27 25 27 18 1987 4 40 10 18 1988 7 22 23 9 1989 15 29 5 19 Total 218 240 155 177 * Duties and price undertakings under anti-dwmping procedures. b Excludes anti-dumping actions taken against non-members of the Code and by individual EC countries on a bilateral basis. Sourcer Australian Customs Service; GATT. Figure 3: Australia's 'stock' of anti-dumping compared = 200 Australia USA 180 A 0 ~~~~~~~~~~~~~~Canada o 140- -1984 1989 Source: GAiT. Committee on Anti-Dumping Practices. Reports for 1984 and 1989. 30 Rise and fall In fact, as shown in figure 4, Australia has less anti-dumping actions on the books today than at any time in the past decade. The rise in the 'stock' of anti-dumping up to 1985 reflected an increase in the number of actions in the early 1980s, discussed previously, combined with only a very low rate of revocation of previous anti-dumping measures. After 1985, the situation was reversed: the fall in the stock of anti-dumping actions reflected a sudden increase in the revocation of old measures, combined with a decline in the number of new cases. (See also table 3.) The most straightforward explanation for the rise in anti-dumping activity in the early 1980s is that there was simply a sudden increase in dumping at that time. Gruen has observed that 'with the world economic recession in the early 1980s, basic commodities became available at bargain basement prices' (p.8). At such times, as indicated earlier, it is normal business practice to cut prices and reduce profit margins, as long as marginal variable costs are covered. This would normally apply to both domestic and foreign markets and need not lead to dumping in the traditional 'price disc-iniination' sense. Figure 4: The rise and fall of anti-dumping M200 X180 Ces outstanding z 160 140 120 - 100 / 60. 40. A / , ctions taken 20- _----_ 0 ~ 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 Year ended 30 June 31 Table 3: Anti-dumping stocks and flows Year ended New actions Old actions Outstanding actions 30 June taken revoked at 30 June 1979 25 0 29 1980 28 1 56 1981 23 1 78 1982 26 0 104 1983 57 2 157 1984 41 17 183 1985 18 13 188 1986 27 50 165 1987 4 60 109 1988 7 66 50 1989 15 43 22 Source. Australian Customs Service Coincidentally, however, it was just prior to this time that Customs began to interpret the GAIT provision that the 'normal value' for an export good should be taken as the comparable domestic price 'in the ordinary course of trade' to mean not at a loss. Redefined in this way, it would not be unusual for the incidence of 'dumping' to rise during a global recession. And, with domestic markets depressed, 'injury' is more likely to be found. The increased demand for anti-dumping action in Australia can similarly be explained not only as a reaction to the possible increase in dumping (as originally defined) but also by the competitive pressures placed on Australian industries in an economic enviromnent characterised by declining intemational competitiveness in depressed domestic markets. As shown in figure 5, the rise and faUl of anti-dumping cases was broadly correlated with movements in the international competitiveness of Australian manufacturing industry. In general the second half of the 1980s has also been characterised by a more buoyant domestic economy than the early 1980s. Similarly, the sudden increase in the revocation of anti-dumping measures which began in the mid- 1980s - well before the introduction of the three year sunset clause - may be attributable to the sharp depreciation of the Australian dollar, which greatly improved the intemational competitiveness of Australian manufacturers, combined with a more active review process by Customs to weed out anti-dumping which was no longer justified. 32 This in turn may well have been prompted by the new scrutiny to which the anti-dumping process was being subjected around the time of the Gruen Review. Figure 5: Competitiveness and the demand for anti-dumping e 120 -100 5 115 A - 590 - 110 80z 105 - 70 .ro Competntiveness 60 100 60 '5 95 50 g 90 40 85 30 80 20 Anti-dumping cases initiated 75 - 10 70 . 0 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 Year ended June Source: Treasury Budget Statements 1989-90 and Australian Customs Service. The translation of domestic competitive difficulties into anti-dumping action rather than demand for conventional assistance could have reflected perceptions that the need to pursue such assistance through the IAC process would be too difficult for most industries -except the politically most powerful ones, like textiles and clothing and motor vehicle producers. Gruen makes the following telling observation about the anti-dumping surge in the early 1980s: Increasing import competition and the Government's clear policy of avoiding raising taiiff levels or taking more drastic protectionist measures placed the dumping administration under pressure in the 1980s. Phrases such as 'the first line of defence' came to be applied to the system. (p.8) 33 Acceptance rate for anti-dumping petitions One possible indicator of how tight or loose the Australian procedures have been is the proportion of anti-dumping petitions which actually get through. Ideally we would want to know the proportion of applications received (the very beginning of the process, as set out above in figure 2) that result in duties or undertakings. Data are not available for the number of petitions which do not make it to the formal initiation stage, although one anti- dumping official's informed guesstimate is that the proportion of 'non-starters' would have been around one-third in recent years. In the absence of data on applications received, we must use the statistics on cases formally initiated as an indicator of the 'demand' for anti-dumping action. The proportion of cases formally initiated which has resulted in such action is shown in figure 6. In the 1970s it averaged 32 per cent, but rose to 46 per cent in the 1980s. The overall rise mainly reflected a sharp increase in the early 1980s, at the same time as the increase in cases (in other words, as the number of cases increased so too did the proportion of them with a positive outcome). This could be seen as supply responding to demand. On the other hand, it may simply reflect the rise in actual dumping through this period , so that there were more well founded petitions than before. However, as already noted, the dumping ground rules changed in the early 1980s, making it easier to obtain (supply) a positive finding of dumping. Figure 6 also shows a sharp rise in the acceptance rate in 1988-89, the year in which the current system began (in September 1988). In practice, much of this comprises actions carried over under the previous system and, according to anti-dumping officials, reflects greater stringency in the screening process before formal initiation of cases. Of the 19 cases actually initiated in 1988-89, however, only 5 (26 per cent) have led to positive anti-dumping action, the rest being terminated 34 Figure 6: Acceptance rate of anti-dumping petitions' % 80 70 - 60 - 40 - 30- 20- 10 0 ~ 1976 1980 1985 1989 Year ending 30 June a Ratio of gazenals and undertakings to final action taken. Source. Austalian Customs Service and GAIT Duration of anti-dumping action Returning to a theme in the previous chapter, one test of the protectionist effect of anti- dumping procedures is the length of time that anti-dumping actions remain in force. Long term 'dumping' is not an economic threat, nor can it be motivated by the sort of strategic considerations which people might regard as unfair. In addition, the longer the duration of anti-dumping action, the greater the possibility that it may cease to be justified according to anti-dumping criteria. Table 1 above suggests that on this basis the Australian scheme has not performed very well, al least up until the tidying up that began in 1985. It is of some interest that of the 245 cases that were revoked or released between 1985 and 1989, only 11 (or 4.5 per cent) have been the subject of renewed investigations. That has been explained as a consequence of the continuing low value of the Australian dollar. Although the 35 depreciation of the dollar should have made no difference to findings of dumping, it is likely to have made injury more difficult to prove. Country coverage and incidence As a general proposition it would seem that, for any given commodity, the more countries against which anti-dumping action is taken, the more of a 'protectionist' role played ' the anti-dumping process - in the sense of protecting an industry from, and depriving an economy of, prevailing world market prices. Between July 1983 and June 1989, there were some 270 dumping measures revoked or released, comprising about 115 commodities. For 50 per cent of these commodities, anti- dumping actions involved a single supplier, 10 per cent of commodities had two dumping suppliers; and 40 per cent had three or more suppliers subject to anti-dumping penalties. For nearly 20 per cent of commodities involved, actions had been taken against six or more suppliers. That the anti-dumping net is in practice catching much normal international trade is also illustrated by the mixed catch of countries in particular cases - from the least developed, to centrally planned and highly industrialised. For example, consider the following lineup of countries against which anti-dumping duties were imposed on imports of electric motors: India, Brazil, Taiwan, China, Czechoslovakia, Romania, Yugoslavia, Poland, East Germany, West Germany, the Netherlands and the United Kingdom - and this action lasted for over eight years. (Other examples can be found in table 6 below on dumping margins.) While all of these suppliers were undoubtedly found to have been 'dumping' according to Australia's legislation - and consistent with GATT rules - it is reasonable to conclude that ill such cases the dumped prices were in fact 'the world price'. Such cases seem highly unlikely to be of an intermittent or temporary nature, and predation is clearly out of the question. The incidence of Australia's anti-dumping action varies considerably by country group (table 4). Relative to their shares in Australia's impoits, the hardest hit regions are the 36 'Asian Tigers' (mainly Korea and Taiwan) and, especially, China and the Eastern bloc - the last accounting for 9 per cent of anti-dumping but only 0.2 per cent of imports. China and the Tigers are also the world's lowest cost suppliers of many light manufactures. However, another top competitor, Japan, had the lowest incidence of anti-dumping relative to its import share. The relatively very high incidence of anti-dumping actions involving centrally planned economies undoubtedly reflects the fact that, for these countries, domestic selling prices are not used to determine normal values. Instead, normal values are based on export prices from market economies, the choice of which will obviously be critical to the dumping assessment. It is difficult to imagine how such a decision could be made on any objective grounds and, even if it could, there remains the problem that whenever the centrally planned economy is the lowest cost producer, its export price must inevitably be found to be below the (third country) 'normal value'. This is clearly relevant to China, for example, and that country has been justifiably critical of its treatment under Australia's anti-dumping laws. Table 4: Country incidence of Australian anti-dumping Proportion of Share of anti-dnmping actionsa importsb North America 12 24.3 NewZealand 5 4.1 Westen Europe 34 26.4 Japan 10 23.1 Asian Tigersc 16 8.0 China 6 1.2 Other developing 7 12.3 Eastem bloc 9 0.2 a Based on 249 actions revoked or released between July 1983 and June 1989. b In 1985. C Hc...g Kong. Korea, Singapore and Taiwan. Sources: Australian Customs Service and IMF, Direction of Trade. 37 Industry incidence The pattern of anti-dumping cases (fonnally initiated) across industries is quite uneven, as shown in table 5. If industries simply saw anti-dumping as an alternative form of assistance, then it could be expected that the industries with the greatest need for assistance would be the predominant claimants. While the table reveals no clear pattern overall, it is apparent that the least competitive sectors such as textiles, clothing and footwear and passenger motor vehicles are, in fact, the lowest users of the anti-dumping process. That may reflect the fact that these industries were already receiving very high protection and largely through import quotas, which serve to insulate domestic industries from import-price competition. (The only items of clothing that have been subject to dumping duties are those for which quotas do not apply.) The predominant users of anti-dumping action have been the chemicals industries and producers of 'other machinery and equipment'. The former is a major input into industrial processes (recalling the fertiliser case) and has received relatively low assistance through the IAC process. Price effects of anti-dumping action The overall price effects of anti-dumping action are very difficult to evaluate from available information. Unlike the tariff, actual duties collected are next to useless as an indicator of the average 'wedge' between foreign and domestic prices. That is partly because a significant proportion of anti-dumping actions (50 per cent in 1988-89) take the form of price undertakings, which are not reflected in customs revenue. Moreover, anti- dumping duties themselves operate on a sliding scale, depending on the gap between the import price and a designated 'floor price' and, as described later, it is in the exporters' interests to price up to the floor level and appropriate the revenue that would otherwise have been paid in duties. Table 5: Incidence of anti-dumping cases by industrya Assistance rateC Industry 83-84b 84-85 85-86 86-87 87-88 88-89 6 year total Share of total Nominal Effective no. no. no. no. no. no. no. % % % Foodandbeverges 4 3 3 1 - 2 13 4.6 8 5 Textiles 1 - 2 1 - 1 5 1.8 24 71 Clothing and footwear I 1 - - - - 2 0.7 70 182 w Wood, wood products and furniture I - - - - - I OA 13 17 Paps adpapn products 3 2 - - - 5 10 3.5 12 15 Chemical and petroleum products 24 21 21 12 9 4 91 32.3 4 13 Non-metallicmineralproducts I - - - 2 - 3 1.1 3 4 Basicmetalproducts 9 9 8 3 8 - 37 13.1 6 9 Fabricatedmetalprducts 6 3 1 4 - - 14 5.0 15 22 Transport eqwpment - - 2 - - - 2 0.7 27 51 Other machinery and equipment 17 26 7 17 6 8 81 28.7 16 22 Miscellaneous manufactuing 4 3 10 2 4 0 23 8.1 18 26 Total 71 68 54 40 29 20 282 100.0 a Cases formaUy initiated. b Extrapolated from data including subsidy countervailing cases c The nominal rate of assistance is based on tariffs and other assistance for each industrys output, whereas the effective rate is a measure of net assistance, taking into account tariffs and assistance on inputs as well. Source: Australian Customs Service. 39 Statistics on dumping margins are not kept systematically, but are indicated in the reports on particular cases and in reports to the GATT. Lloyd (1977) had access to statistics for the early 1970s, which showed that average margins across industries ranged from 10 to 43 per cent, and were in most cases significantly greater than the relevant average nominal rates of assistance. Table 6 shows the dumping margins for cases on which action was taken in 1984-85 and 1988-89. As in the early 1970s, the dumping margins vary considerably from one product to another, and also for the different suppliers of a given product. (For example, margins on fluorescent lamps ranged from 3 per cent for Japan to 50 per cent for Canada.) In most cases the dumping margins are considerably larger than the actual protective tariff. No significant differences are discernible in the 1988-89 data, compared to that of five years earlier, except that the average dumping margin is if anything higher. Gruen (1986) notes that at the end of 1985 (when anti-dumping action was near its peak) only two per cent of statistical import keys and two per cent by value of imports were affected (p. 5). While this suggests that anti-dumping's economy-wide price effects may not be important, such statistics are deceptive. The IAC has argued that the very existence of anti-dumping procedures can lead to pre-emptive price increases, sometimes in situations where anti-dumping action would not have occurred. One witness to an inquiry stated that ... overseas suppliers are generally not prepared to offer their best export price to Australia for fear of ... being charged with dunping. (Cited in Cannichael, 1986, p. 11) To the extent that this system has such an intimidating effect, the statistics relating to formal anti-dumping action may considerably understate its potential effect on domestic price formation. 40 Table 6: Dumping margins Average dumping Product Country Margina Tariffb A. PERIOD 1984-8S Outboard motors USA 27.1 P 10 Steel pipe and tube Singapore 17.5 10 Taiwan 31.0 Thailand 10.4 Korea 26.9 South Africa 64.0 15 Alloy steel chain Sweden 3.0 P 20 Fluorescent lamps Taiwan 44.0 P 10 Thailand 28.0 P Korea 15.0 P Philippines 18.0 P Hungary 15.0 P Japan 3.0 P 15 West Germany 6.0 P Canada 50.0 P East Germany 10.0 P Polystyrene France 5.8 30 Canada 5.9 West Germany 5.3 Ceiling sweep fans Hong Kong 32.0 20 Pasta Italy 23.0 10 Phosphoric acid Japan 25.0 20 Diagnostic reagent strips West Germany 79.0 15 Pdmalic anhydride Brazil 50.0 P 20 Israel 6.0 P Taiwan 11.0 P Dextrose monohydrate Italy 10.0 P 10 Singapore 10.0 P 5 PVC Italy 7.0 P 25 Domestic microwave Japan 12.0 P 25 Waxed cotton motorcycle garments Korea 22.5 P 40+ Stearic acid Malaysia 16.3 P 5 Di-oxtyl phtholate Korea 26.8 P 15 Vinyl floor tiles Norway 20.6 P 20 Film laminate USA 10.0 P 22.5 B. PERIOD 1988.89 Levamisole Hydrochloride Switzerland 9.0 P flee Singapore 27.0 P China 32.5 P Hong Kong 36.5 P Belgium 20.0 P West Gennany 18.5 P 41 Table 6: (Continued) Average dumping Product Country Margina TariffP Coloured pencils Brazil 43.0 P 8 Hungary 29.0 P 13 Poland 44.0 P Multityred rollers Czechoslovakia 56.0 P 24 Fork lift trucks Japan 14.0 P 21 Cement clinkers Katea 55.0 P fiee Electric motorparts Taiwwa 32.5 P 13 Outboard motors Belgiun 39.0 P 10 Japal 25.5 P USA 28.0 P A dunping margin is the difference between the 'nomal value' and the fob export price, expressed as a percentage of the nomnal value. P denotes provisional finding. b For each product, the prefeential rate for developing country exporters is 5 per centage points below the general rate. Source: Australian Customs Service, Semi-annual report of Anti-dumping actions, prepared for GA1T, various issues. Rules and procedures A necessary complement to looking at the outcomes of anti-dumping cases is an understanding of how the rules and procedures work. It is here that the international debate is mainly focused - there is concern to tighten up the rules and procedures to prevent or limit the extent to which anti-dumping is used as a protectionist, instead of 'fair trading', device. In Australia, as in other countries, the rules and procedures have changed considerably over the past decade. Three aspects which are central to the anti-dumping system are the criteria relating to 'normal value', 'material injury' and causality between the two. To paraphrase the GATT requirement: 'products introduced into the commerce of another country at less than normal value is to be condemned if it causes or threatens material injury to an industry'. We look at those in turn. 42 Normal value GAlTs Article VI states, in the rather difficult language for which the text of the General Agreement is famous: For the purposes of this Article, a product is to be considered as being introduced into the commerce of an importing country at less than its normal value, if the price of the product exported from one country to another (a) is less than the comparable price, in the ordinary course of trade, for the like product when destined for consumption in the exporting country, or, (b) in the absence of such domestic price, is less than either (i) the highest comparable price for the like product for export to any third country in the ordinary course of trade, or (ii) the cost of production of the product in the country of origin plus a reasonable addition for selling cost and profit. Due allowance shall be made in each case for differences in conditions and terms of sale, for differences in taxation, and for other differences affecting price comparability. Even to the uninitiated it should be apparent that this definition is full of linguistic loopholes. For example, what exactly does the 'ordinary course of trade', the 'like product', the 'cost of production' and the 'reasonable addition' refer to? The Anti- dumping Code is not very helpful. (It tells us that the 'term "like product" shall be interpreted to mean a product which is identical, ie. alike in all respects to the product under consideration, or in the absence of such a product, another product which, although not alike in all respects, has characteristics closely resembling those of the product under consideration'.) It is thus in each country's national legislation and administrative practice that we must seek an interpretation of such terms. The Australian legislation (the 1975 Act) was similarly vague on the above aspects of 'normal value' to GATT Article VI and the Code. However, as already noted, the term 'ordinary course of trade' began to be administratively interpreted in the early 1980s as 43 excluding sales at a level which does not cover all costs (fixed and variable). This was given legislative backing in the 1984 changes to the Act which essentially put it that, if the selling price in the exporters home market is - for an 'extended period of time' - and for a 'substantial quantity of goods' - less than what Customs judges to be the relevnit costs of production, then, if 'it is likely' that the seller won't be able to make up the difference subsequently, that price shall not be taken to have occurred in the 'ordinary course of trade'. These changes (shared by the other leading anti-dumping countries) opened up a Pandora's Box of additional possibilities for finding dumping. They mean that the essential price discriminatory requirement for dumping need no longer apply. And export prices need not be lower than domestic selling prices for dumping to be found. It is difficult to find any justification in economics or equity for this new rule. The ADA has suggested that if an exporter is selling domestically at a loss for a lengthy period then he 'is up to something funny; he is not behaving as does an ordinary business' (ADA, 1984, p. 23). But it would not be out of the ordinary for a manufacturer not to be covering all of hisfLxed costs (depreciation on plant and equipment). In depressed market conditions this might indeed be the rational profit maximising strategy, at least up to the point when assets needed to be replaced. Pricing all sales below variable costs, however, would be unusual (irrational) and it is hard to think of any plausible circumstances in which such a strategy would be pursued (especially given the unlikelihood of achieving long term predatory gains). In any case, as noted earlier, the longer the period involved, the larger are the likely net benefits to the importing country from low prices. Given the lack of basis for introducing this Section 5(9) rule, it is unfortunate that Customs have interpreted it to mean that local selling prices can be used as normal values only if it is proven that they are not below cost. Most firms keep production and financial records for their internal planning and domestic tax purposes, not for the convenience of other countries' anti-dumping administrators. It is inevitable, therefore, that such proof will often be lacking - even where reasonable grounds exist for accepting that sales have 44 not been made at a loss. And this is what has in fact happened. (In a recent report, the ADA has interpreted the proof requirement in the Act in the opposite way to the ACS; namely, that local selling prices are to be used unless there is proof that they are below cost.) The legislation provides a hierarchy of techniques for deterrining normal values in the absence of a usable domestic selling price. They are shown in descending order of priority in table 7. What this reveals is that in only 42 per cent of cases in the period 1980-85 were local selling prices directly used, while 23 per cent of normal values were 'constructed' or based on third country producers' prices, and 36 per cent were based on 'available and relevant information'. Discussions with anti-dumping officials reveal that in practice the 'available and relevant information' that has typically been used is the local selling price anyway - an indirect procedure necessitated by the proof requirement. Table 7: Methods used to determine normal value, 198045 Number Reason for not using method of times Method (1) used (1) Domestic price in country of export 73 (2) Constructed from exporters cost information No domestic sales 14 Domestic sales at a loss 15 (3) Assessed from third country information Non-market domestic pricing 11 (4) Assessed using 'available relevant and reliable Insufficient information to use 63 information' methods (1) or (2) Total 176 Source: Based on Gruen (1986). Table 3. p. 27. Information is unavailable on the proportion of positive dumping findings derived from each method. There is the suspicion, however, that techniques other than local selling price are more likely to yield normal values above export prices. Gruen's view was that 'it is desirable to reduce the number of "constructed normal values" to the irreducible minimum' (p.29). He noted that 'there are technical problems in constructing prices (cost allocation, most economic producer or average producer, depreciation policy, etc.), which give considerable variation in possible results' (p.28). 45 As indicated previously, the imputed profit issue has been a central concern since the Gruen Review. Gruen had argued that if his recommendation to repeal S5(9) on sales at a loss was not accepted, 'some minor move towards a more realistic normal value could be obtained by not including a "reasonable profit" in constructed prices' (p.29). The government appeared to accept this approach, but the ADA recommended, on grounds consistent with the logic of the Act, that this should only apply to S5(9); constructed values in other circumstances (such as when there are no domestic sales) should include an imputed profit. This interpretation was accepted. Box 3: The cherries-in-brine case The Austrlan experience has been that it is possible for quite different findings to be made about anti-dumping matters on the basis of the same 'facts'. This is illustrated by the case of 'cherries in brine or other preservative, from Italy'. This case was sent .o the IAC for review in Mareh 1985, following a request by a producer of glace cherries that the Minister revoke anti-dumping duties that had recently been applied to imports of cherries in brine. The Chairman of the IAC has described some points at issue: While the Act requires that like goods be compared with like, this is easier said than done. Even for this product - which may appear to present little difficulty in comparing Australian imports with like goods in the exporting country - four different grades of cherries are commercially recognised in Italy, the country about which the dumping complaint was made. Customs' judgment was to recognise only three grades. In its subsequent review, the Commission found that if four grades had been used the extent of dumping found would have been much less. This was true also of the valuation of reusable drums in which cherries are packaged. Customs considered that the drums should be valued as new whilst Commission investigations suggested that, in view of their physical condition, the drums were only worth half that much. The exercise of different judgments in this case can be seen to have produced widely divergent results in terms of the penalty which could be imposed on the users of brined cherries (Carmichael, 1986, pp. 2-3). 46 Material injury The central issues here have been described by the IAC as follows. Anti-dumping cases always start with a local firm or industry complaining that imports are causing injury. Typically, evidence is produced of some decline in, say, prfitability, sales or employment. However, considerable judgment is required to determine in a practical sense what injury means. Does the loss of one sale constitute injury? Does, say, a 5 per cent loss in sales carry more weight in determining whether injury is occurring than a 5 per cent drop in prices or employment? Even where injury is judged to exist, it is common for investigations to find that injury is attributable to a range of factors - changing technologies, consumer tastes and general economic conditions - and dumping may, or may not, be numbered amongst these. And even where dumping is involved in such cases, it is often not possible to determine what share of total injury is attributable to it alone. Before anti-dumping action is taken it is necessary to establish that there is injury and that there is a causal link between dumping and material injury. In the Commission's experience, unraveUling an often confusing web of cause and effect is a precarious task. When the exercise is conducted in an administrative framewoik which places most emphasis on the situation of a domestic producer, there must be a tendency for outcomes to favour these producers - even where factors other than dumping may be contributing substantially to their problems. (Carmichael, 1986, pp.5-6) The GATT Anti-dumping Code, which as already noted was motivated mainly by concern to reduce alleged US protectionist use of anti-dumping, stiffened the language on injury by requiring that dumped imports were required to be the 'principal' cause of injury. In the Tokyo Round revisions of 1979 this was deleted, so that the GATT injury test remains today that dumped goods 'cause' 'material injury'. This, of course, begs the question of what constitutes injury and what is 'material'. These questions have also received considerable attention in Australia since the Gruen Review. Like the GATT Code, the Australian legislation provides a 'non-exhaustive' list of factors to be taken into account in determining injury - including loss of market share, employment or production declines, price 'suppression', profit declines and under- utilisation of capacity. In practice, the experience through the early 1980s had been that 47 most industries fairly readily passed the injury test. The Administration appears to have been more concerned with establishing that an industry had been injured, by reference to one or more of the criteria just listed, than the question of whether the imports in question had caused that injury. As one researcher has observed, on the basis of an examination of anti-dumping reports through the early 1980s: The Department's view seems to be that if dumping is occuning and there is evidence of injury then there is a case for the imposition of dumping duties. This seems to be the case even when injury is not appreciable or when there are other factors significant to the local industry's plight. (McGinnis 1985, p.38) And Gruen noted in his report that he had been 'provided with a substantial number of examples where Australia has applied anti-dumping actions which, on the face of it, did not produce "material injury" (or threatened injury) or where there was no causal link between injury and imports' (p.31). An approach used in Australia (and other countries) has beei. to 'cumulate' the market effects of dumped imports from different sources where imports from a single source are too small to have much effect on their own. For example, the Singapore High Commission in Australia drew Gruen's attention 'to three cases in 1980 when Singapce's share of the Australian market was 0.01% (vertical filing cabinets), 0.05% (synthetic organic dyes), 0.26% (correcting fluid)' (p.3 1). (Digressing for a moment, it seems curious that Singapore, a virtually free trading country, should have been found guilty of dumping in the first place. Arbitrage - the moving about of goods and services to benefit from price differentials - would normally ensure dhat there was little scope in a free trade country for selling on world markets at lower prices than at home.) There was an active debate in the GAIT a few vears ago about whether the practice of cumulation was consistent with the Code. Like many interpretative debates about the intent of (intentionally) ambiguous GATT 'law' it appears to have led nowhere. However, there is no doubt that such an approach is on firm legal ground in Australia, as 48 it has been tested in the Federal Court. (Although that need not make it any more rational from an economist's perspective.) Gruen recommended three changes to the tests for injury 'to place emphasis on the need for material injury and a causal link': * price suppression must be backed up by other factors such as 'seriously reduced profits' * a 'substantial effort should be made to allow for the influence of factors other than dumping in causing injury' * injury should be significant, not just for the specific product, but 'in the context of the entire operations of the relevant establishment'. The government appeared to accept the broad thrust of these recommendations and allocated injury/causal link questions as an important function of the new Anti-dumping Authority. In the Second Reading Speech for the 1988 ADA Bill, the Minister remarked: Assessments of material injury and the causal link must be vigorous and anti-dumping measures should not be used as a de facto fonm of protection: they have to be seen as a set of measures to discourage unacceptable short-term trends to knock out an industry ... The Authority will take into account the influence of factors other than dumping in causing injury .. (Jones 1988, p.2311) In a reference to the ADA shortly after its formation, the government sought the Authority's advice on what regulations - formal ministerial directions might be issued to put into effect the Govemrnent's intention that ... in determining material injury ... (a) the reduction of prices caused by dumping should substantially reduce ... profits on the Australian production ... and (b) the market share of dumped imports should not be insubstantial ... As indicated earlier, industry bodies objected strenuously to this dual test for injury and also to the Minister's statement that industry needs to be threatened with being 'knocked out' before anti-dumping measures are applied (Stubbs 1989). 49 The ADA's report, while providing a masterly exposition of the issues, was in the end unable to tie the concept of 'material injury' down much better in operational terms than previously. The Authority concludes that 'material' should be considered in terms of its opposite: thus not immaterial, insubstantial or insignificant; greater than that likely to occur in the normal ebb and flow of business. (ADA 1989, p.1) On the question of the causal link, the ADA stated its interpretation of the government's position as follows: The government expects that material injury, or the threat thereof, will only rarely be taken as proven when the Australian industry producing like goods has not suffered, or is not threatened with, a 'material' diminution of profits or when the dumped imports do not hold (or threaten to hold) a sufficient share of the market to cause or threaten 'material' injury. However, the Authority did not exclude the possibility that material injury might (in 'rare cases') be found where neither of those conditions held. Nevertheless the current procedures relating to injury appear to be considerably tighter in their application than under the previous administrative arTangements. This is illustrated by a simple statistical comparison: of 22 anti-dumping cases terminated in ! 984-85, only five (or 23 per cent) were because of failure to pass the injury or causality test; in 1988- 89, the first recorded year in which the new regime applied, this had risen to 70 per cent. Anti-dumping measures The GATT Code provides that where dumped goods are found to have caused material injury to an industry, the government is entitled (though not required) to impose duties which (a) 'must not exceed the margin of dumpi.g ...' and (b) should be 'less than the margin, if such lesser duty would be adequate to remove the injury' (Article 8). Alternatively, the government may accept 'voluntary' price undertakings of the same magnitude from the supplier. 50 Australian practice has been that most cases have involved duties, rather than price undertakings (table 8). In conuist to other countries, the duties are not fixed, but are calculated for each shipment so as to raise the import price to the declared nonnal value or (lower) non-injurious level. Table 8: Anti-dumping measures applied Year D)uties Undertakings 1979-80 27 1 1980-81 21 2 1981-82 18 8 1982-83 46 11 1983-84 36 5 1984-85 15 3 1985-86 21 6 1986-87 3 1 1987-88 6 1 1988-89 10 5 Source: Australian Customs Service The IAC has made the following observation about this approach: Many exporters react to the imposition of anti-dumping dutes by increasing the fob price of the imported good until it is equal to the 'floor price' established by the normal value, thereby avoiding the imposition of anti-dumping duties. Where this happens, Australia loses income which it would not lose if the reduction in imports were brought about by an increase in tariffs' (Carmichael, 1986, p.10). While this implicit criticism is justified if the objective is to protect the local industry from import competition in the lowest cost way, the Australian approach is nevertheless consistent with the GAITs anti-dumping objective of achieving 'fair' import prices. Foreign exporters are given every chance of setting their prices 'fairly', at least cost to them. (This explains the low incidence of formal undertakings.) In this respect, the Australian scheme is more flexible and 'exporter friendly' than that of the EC, for example (see Norall 1986). Moreover, in Australia anti-dumping procedures have also not been used as a lever for negotiatir. g separate voluntary restraint agreements, as appears to have occurred in the 51 United States. (Finger and Murray, 1989) Price undertakings must be made in the context of the anti-dumping procedures and require the prior determination of the dumping margin, on which they must be based. VS here large dumping margins are found, the administration, in line with GATT requirements, has often determined a lower floor price which in its judgment is sufficient to prevent injury (the so-called 'non-injurious free on board' value or NIFOB). The Act provides no guidance as to how this might be done and judgments must inevitably be made on the basis of local manufacturers' evidence and advice. NIFOBs are therefore, as Gruen suggests, more likely to err on the high side. Gruen estimates that NIFOBs accounted for about 10 per cent of duties in place in 1986 and 40 per cent of undertakings (p.29). As both Gruen and the IAC point out, NIFOBs do not compensate for excessive estimates of dumping margins - often based on constructed, or third party, prices. This is illustrated in the Commission's anti-dumping report on Dining Candles from the Peoples' Republic of China. Customs had determined that the dumping margins assessed using Malaysian normal values were higher than necessary to redress injury suffered by the Australian industry. A single nifob value, up to 42 per cent below the three assessed normal values for Malaysia, was established as the basis for dumping duties. Had normal values initially been assessed for the 'preferred' third country, Hong Kong (as was done in a subsequent review) these normal values would have been significantly lower than the nifob value initially used. (Carmichael 1986, p.7) Once anti-dumping measures have been imposed, there is a statutory requirement that such actions be reviewed and monitored. In practice, since import prices will typically have been increased to the designated floor levels, a review is faced with making the difficult judgment as to how prices would move if the floor level were removed. This could account for the apparent reluctance of Customs to engage in reviews as programmed, referred to in the IAC submission, and there is little doubt that many of the active anti-dumping measures in force through the eighties were no longer justified. A number of examples have been given from IAC inquiries. Further evidence is the sudden rash of revocations from 1985 and the relatively small number of renewed petitions, referred to previously. 52 Long term anti-dumping measures can have two possible effects: * the exporter's best price could become progressively lower than the floor price, as costs and prices of certain goods decline over time (electronics are a good example), thus increasing the protective effect; * alternatively, world-wide inflation could eventually make the floor price redundant as a restriction on foreign imports. Which of these has happened in the Australian case is a matter for speculation. Transparency The GATT requires that the basis for anti-dumping decisions be known to the parties involved and that there be avenues of appeal. The procedures themselves, as summarised in figure 2, are intended to ensure that the parties have adequate opportunity to put their case and scrutinise decisions. Australian procedures have not been challenged in the GATT for failing to meet the transparency requirements of Article VI and the Code. Nevertheless, the Chairman of the IAC levelled the following criticism at the system in his submission to the Gruen Review: The present administrative system is relatively closed. While meetings of parties are convened by Customs the rationale for decisions is generally not made public. Many of the facts on which decisions are based remain confidential and the rates of cash securities and dumping duties imposed on imports are in many instances also confidential. There is not even a guarantee that the application of anti-dumping action to a particular class of good will be made public. The creation of a more open system of administration is a pre-requisite for well informed decision making. (Carmichael 1986, p. 15) These charges are partly explicable in the context of the difference between the economy- wide information and public scrutiny which characterises the IiC system (discussed in box 1) and the more limited transparency that the law requires of the anti-dumping process. Where decisions are based by law on considerations relating only to a particular domestic industry and the foreign exporter, it is perhaps inevitable that these interests are the best catered for. 53 Even within that system, however, Australia's anti-dumping procedures and laws could be made considerably more transparent Gruen commented that The Act itself shows all the signs of having had amendments grafted on top of amendments. Unrelated clauses sit side by side and related clauses are widely separated (p.39). He suggested the rewriting of the Act, with explicit definitions of a number of key terms. The fact is that anti-dumping is in practice not just a contest between the foreign exporter and the local producer. Domestic consumers and other industries are vitally affected - even if their interests cannot be directly taken into account. It is important in terms of providing at least some external scrutiny of the system that as much information as possible is made generally available - not only on particular cases, but also on the larger picture. Under the present arrangements some steps have been made in this direction. The ACS annual report now contains tables showing trends in the composition, outcome and incidence of anti-dumping (and countervailing) activity, including by industry. And the ADA reports generally provide a coherent discussion of the issues in its inquiries and how decisions were arrived at. (As already indicated, its report on material injury and other matters is particularly informative.) One of Gruen's recommendations which has not been acted on was that all normal values and NIFOBs should be published. In principle the Act allows this, subject to confidentiality requirements. In practice the parties immediately involved have vetoed it on that basis - as it is alleged that publication would provide competitors with commercially relevant information about production costs. Prior to 1984, appeal procedures involved the IAC and permitted that body to provide some information on the broader effects on the Australian economy of taking anti- dumping action. The 1984 changes ruled out IAC consideration of economy-wide effects and since 1988 the IAC has been displaced from the system by the ADA. The Federal Court provides a higher level of appeal - but again oily in relation to 'the facts'. 54 V. CONCLUDING REMARKS Anti-dumping in Australia has been through a major cycle over the past decade, eloquently depicted in figure 4 above. The analysis in this report suggests thlat that cycle has had more to do with domestic economic conditions and the degree of permissiveness in anti-dumping rules and their interpretation than with dumping as such. In the past few years, more buoyant domestic economic conditions and the diminished Australian dollar have reduced the demand for anti-dumping, while legislative and other changes to the system have simultaneously reduced the supply. Compared with our own past and other countries' present, Australia is much less of an anti-dumping activist than it was. The question is whether this will continue. The foregoing analysis raises a number of basic issues which have a bearing on that. For one thing, it is clear that the anti-dumping system retains a degree of administrative or ministerial discretion which will always make it vulnerable to the business cycle and the political cycle. That is not just an Australian characteristic: it is common to the nature of the anti-dumping process wherever it may be found. The Australian government has at least wrestled fairly publicly with it to obtain some greater precision and objectivity. And in some areas improvements have been made. But on the three touchstones of the anti- dumping process - normal values, material injury and causality - there are many points at which arbitrary judgments will always need to be made, as long as there is an anti- dumping system. It is inevitable that those judgments will be coloured by the political and economic climate of the day. Compared to the early 1980s, the present climate is relatively 'dry'- but that can change again. The fact that anti-dumping is a complex process with many rules which, depending on interpretation or minor changes, can have important differential effects on the fortunes of Australian industries, means that the system and how it operates will remain of abiding interest to import-competing industries. Lobbying for rule changes, or favourable interpretations, will continue as long as the expected return from such lobbying exceeds the cost. Recent events in Australia suggest that that continues to be the case. 55 Regardless of whether anti-dumping has served as a protectionist device in Australia, it is clear that industry sees it in that role. The Gruen Report and the recent ADA review of key concepts provide sufficient evidence of that. For example, industry argued in the latter review that market conditions in Australia should be taken into account in constructing f..reign exporters' normal values. And on the national interest issue, industry's position could not have been more unequivocal: 'Dumping should be considered as a corporate and not a country activity and national interest should play no part' (Stubbs 1989). Rent-seeking efforts in the anti-dumping arena are also influenced by relative costs and benefits. It is now quite difficult for industries to obtain the sort of conventional border protection that was common a decade ago. Policy has been set in a liberalising direction; that is now widely understood and accepted. 'Protection' has become a discredited concept in Australia. But, as in the United States and elsewhere, 'fairness' will always be popular. And 'anti-dumping' by the very term is seen to be about achieving fairness. Moreover, anti-dumping has the attraction of being a 'low track' route for obtaining protection against imports, as Finger, Hall and Nelson (1982) have explained in the American context. It takes place according to rules and procedures which industry and specialist consultants soon master, away from the public glare. The IAC route, by contrast, is 'high track', costly to claimants and more likely to meet persuasive opposition. Thus the demand for anti-dumping as a protectionist device will continue and it can be expected to rise again when times 'get tough'. In the past, the supply of anti-dumping has to some extent been responsive to that demand. Whether that occurs in the future hinges in part on the new Anti-dumping Authority. It seems clear that the Anti-Dumping Authority has brought a fresh and more critical eye to the anti-dumping process in Australia and in particular has tightened up on how the injury est is implemented in practice, as well as providing a second careful pass at the normal value arithmetic. On a number of occasions its assessments have differed from those of Customs, and generally in a direction that is more sympathetic to the foreign exporter. However, it could be argued that these positive developments have more to do 56 with the government's current attitude than with the institutional innovation as such. The ADA essentially operates within the same industry-specific framework, following the same criteria, as the Customs Service. If there is a difference in perspective, it results from the fact that, in the words of a representative of the ADA, the Authority 'stands in the Minister's shoes'. It is thus more attuned to the political environment in which 'technical' decisions are made. But that can cut both ways. There is ample evidence from the Australian experience to suggest that, while dumping may sometimes be a problem in international trade, anti-dumping presents greater problems. Tinkering with the procedures and criteria for taking anti-dumping action can help reduce its protectionist tendencies to some extent (although such changes are reversible). But it does not resolve the fundamental problem that anti-dumping is at bottom about safeguarding the interests of particular industries. As long as this remains the objective, there will always be tension between anti-dumping policy and the broader interests of the national economy and the world tading system. 57 REFERENCES ADA (Anti-Dumping Authority) (1989), Inquiry into Material Injury, Profit in Normal Values and Extended Period of Time, Report No.4, AGPS, Canberra, March. Bhagwati, J. (1988), Protectionism, The Ohlin Lectures, MIT Press. Carmichael, W.B. (1986), Review of the Customs Tariff (Anti-Dumping) Act: Submission by Chairman, Industries Assistance Commission, Canberra, February. Dale, R. (1980) And-dumping law in a liberal trade order, MacMillan for the Trade Policy Research Centre, London. Dam, K.W. (1970), The GATT: Law and International Economic Organisation, University of Chicago Press. DTR (Department of Trade and Resources) (1980), Exports can be Profitable, Realistic Costing Shows How, AGPS, Canberra. Finger, J.M. and Murray, T. (1989), Policing Unfair Imports: The United States Example, Mimeo, April 11. Finger, J.M., Hall, I.K. and Nelson, D.R., (1982), 'The Political Economy of Administered Protection',.American Economic Review, June. GATT (1984), Committee on Anti-dumping Practices, Minutes of Meetings, ADPIM/12, 25 July and ADP/M/13, 19 December. Gruen, F.H. (1986), Review of the Customs Tariff (Anti-Dumping) Act 1975: Report, AGPS, Canberra, March. Hindley, B. (1988), 'Dumping and the Far East Trade of the European Community', The World Economy, December. 58 IAC (Industries Assistance Commission) (1984), Annual Report 1983-84, AGPS, Canberra. - (1986), The Rice Industry -Interim Report, No. 396, AGPS, Canberra, 15 October. - (1987), Annual Report, 1986-87, AGPS, Canberra. Jones, B. (1988), Second Reading Speech on the Anti-Dumping Authority Bill, House of Representatives, Weekly Hansard No.7,26-28 April. Lloyd, P. (1977), Anti-dumping Actions and the GATT System, Thamies Essay No.9, TPRC, London. McGinnis, J. (1985), Dumping Theory and Anti-dumping Practice, Seminar Paper No.6/85, Department of Economics, Monash University. Norall, C. (1986), 'New Trends in Anti-dumping Practice in Brussels', The World Economy, April. Rattigan, Godfrey A. (1988), Domestic Transparency Procedures, A report prepared for the UNCTAD Secretariat, Geneva, UNCTAD/MD/MISC.23, 22 April. Stegemann, Klaus (1980), The Efficiency Rationale of Anti-dumping Policy and Other Measures of Contingency Protection', Discussion Paper No. 387, Institute for Economic Research, Queens University, Canada. Stubbs, K. (1989), 'Industry needs a strong anti-dumping safety net', Topic Paper 6, Australian Chamber of Manufactures, Melbourne. Viner, J. (1923), Dumping: A Problem in International Trade, University of Chicago Press (reprinted 1966 by August M Kelly, New York). -(1926), A Memorandum on Dumping, League of Nations, Geneva. PRE Working Pager Series Contact fl1 Author 1 for paper WPS526 Rural Poverty in India, 1973-86 Nanak Kakwani October 1990 B. Rosa Kalinidhi Subbarao 33751 WPS527 Voluntary Choices in Concerted Ishac Diwan October 1990 S. King-Watson Deals: Mechanics and Attributes Ken Kletzer 31047 of the Menu Approach WPS528 Monetary Policy Instruments for Gerard Caprio, Jr. October 1990 W. Pitayatona- Developing Countries Patrick Honohan karn 37666 WPS529 The Sectoral Structure of Poverty Monika Huppi October 1990 C. Spooner During An Adjustment Period: Martin Ravallion 30464 Evidence for Indonesia in the Mid-1 980s WPS530 The Menu Approach to Developing Asli Demirguc-Kunt November 1990 S. King-Watson Country External Debt: An Analysis Ishac Diwan 31047 of Commercial Banks' Choice Behavior WPS531 The World Bank's Role in Shaping Fred T. Sai November 1990 L. Bruce-Diggs Third World Population Policy Lauren A. Chester 31091 WPS532 Privatization in Turkey Sven B. Kjellstrom November 1990 T. Nana-Sinkan 36026 WPS533 Government Revenue from Alberto Giovannini November 1990 A. Bhalla Financial Repression Martha de Melo 37699 WPS534 Risk Facing U.S. Commercial Banks Menahem Prywes November 1990 J. Israel 31285 WPS535 Shared Investment in General Eliakim Katz November 1990 V. Charles Training: The Role of Information Adrian Ziderman 33651 WPS536 The Link Between Pwverty and Maurice Schiff November 1990 S. Fallon Malnutrition: A Household Theoretic Alberto VaIdes 37947 Approach WPS537 Commodity Exports and Real Income Arvind Panagariya November 1990 S. Fallon in Africa Maurice Schiff 37947 WPS538 Agricultural Reform in Developing Avishay Braverman November 1990 C. Spooner Countries: Reflections for Eastern J. Luis Guasch 30464 Europe WPS539 Maternal Health in Jamaica: Health Chris Naschak Feifer November 1990 L. Bruce-Diggs Needs, Services, and Utilization 31091 PRE Working Paper Series Contact v~ AW~r IQLQfael WPS540 Venture Capital Operations and Silvia Sagari Their Potential Role in LDC Markets Gabriela Guidotti WPS541 Pricing Average Price Options for the Stijn Claessens November 1990 S. King-Watson 1990 Mexican and Venezuelan Sweder van Wijnbergen 31047 Recapture Clauses WPS542 The Metals Price Boom of 1987-89: Boum-Jong Choe November 1990 S. Lipscomb The Role of Supply Disruptions and 33718 Stock Changes WPS543 Development Assistance Gone Donald B. Keesing November 1990 S. Fallon Wrong: Why Support Services Have Andrew Singer 37947 Failed to Expand Exports WPS544 How Support Services Can Expand Donald B. Keesing November 1990 S. Fallon Manufactured Exports: New Andrew Singer 37947 Methods of Assistance WPS545 Health and Development: What Nancy Birdsall November 1990 L. Mitchell Can Research Contribute? 38589 WPS546 The Transition to Export-Led Growth Stephan Haggard November 1990 E. Khine in South Korea, 1954-66 Byung-Kook Kim 39361 Chung-in Moon WPS547 Does High Technology Matter? An Andrea Boltho November 1990 M. Hileman Application to United States Regional Robert King 31284 Growth WPS548 Depos,t Insurance in Developing Samuel H. Talley November 1990 M. Pomeroy Countries Ignacio Mas 37666 WPS549 Intertemporal Substitution in a Patricio Arrau Decemb3, 1990 S. King-Watson Monetary Framework: Evidence 31047 from Chile and Mexico WPS550 Firms' Responses to Relative Price John L. Newman December 1990 A. Murphy Changes in C6te d'lvoire: The Victor Lavy 33750 Implications for Export Subsidies Raoul Salomon and Devaluations Philippe de Vreyer WPS551 Australia's Antidumping Experience Gary Banks December 1990 N. Artis 37947 WPS552 Selected World Bank Poverty Nancy Gillespie December 1990 M. Abiera Studies: A Summary of Approaches, 31262 Coverage, and Findings