Policy, Research, and External Affairs WORKING PAPERS Trade Policy and Industry Development Country Economics Department and Industry and Energy Department The World Bank October 1 991 WPS 776 Enforcement of Canadian "Unfair" Trade Laws The Case for Competition Policies as an Antidote for Protection Mark A. Dutz An "economywide" perspective - examining all allegations of "unfair" dumping and subsidization by comparing the impact on economic efficiency of existing foreign pricing practices with the impact of alternate forms of intervention - would provide a more rational way of responding to problems that trade laws now deal with. ThePolicy, Research, and ExtemalAffairs.Complex distributes PRE Working Papers todisserninatethefindings of work inprogress and to encourage the exchange of ideas among Bank staff and all others interested in development issues. These papers carry the names of the authors, reflect only their views, and should be used and cited accordingly. The findings, interpretations, ar.d conclusions are the authors' own. They should not be attributed to the World Bank, its Board of Diectors, its rnanagenent, or any of its member countries. Policy, Research, and External Affairs Trade Policy and Industry Development WPS 776 This paper-a joint product of the Industry Development Division, Ind:istry and Energy Department and the Trade Policy Division, Country Economics Department - is part of a larger effort in PRE to understand the economics of the emergence of "fairness" as a standard forregulating international trade, its implications for the continued openness of the international trading system, and its continued functioning as an important vehicle for development. It is a product of a research project on "Regulations Against Unfair Imports: Effects on Developing Countries" (RPO 675- 52). Copies are available free from the World Bank, 1818 H StreCt NW, Washington DC 20433. Please contact Nellie Artis, room NIO-013, extension 37947 (52 pages). October 1991. Canada was the first country to enact comprehensive laws could be adapted to deal with cross-border antitrust legislation (in 1889) and the first to institute pricing practices of foreign private enterprises, they an antidumping system (in 1904). Canada's original do not readily address pricing practices of foreign "unfair" trade legislation reflected a desire to prohibit governments. Under an "economywide" perspective, predatory dumping - pricing practices by foreign on the other hand, both private and public contentious exporters designed "to crush out the native Canadian pricing practices could be evaluated to determine industry." But the result of Canada's recent enforce- how, on balance, the national economic interest ment of unfair trade laws has been high levels of would best be served. All allegations of unfair protection for a few well-organized firms. Serious dumping and subsidies would be examined by predatory, anticompetitive concerns were probably comparing the impact of current pricing practices on not at issue in any of the cases in which antidumping economic efficiency with the impact of alternate duties were assessed. feasible forms of intervention so that the chosen policy action results in the largest possible net Canada's recent overhaul of its unfair trade economywide gains. legislation was not followed by any dramatic change in enforcement practice. If anything, the protection The concrete proposals Canada presented during bias of Canadian enforcement has increased. The bias the Free Trade Agreemer i negotiations to regulate against exports from developing countries has also Canada-U.S. cross-borde. pricing issues by competi- increased significantly in the years following imple- tion principles demonstrate that the competition mentation of the revised antidumping and policy alternative is workable. But as Canada's unfair countervailing duty legislation (known as SIMA). trade laws are administered currently, the economywide perspective fits in more readily than Comparing cases involving developed and competition policy. Relatively minor changes to developing countries suggests a protectionist bias existing laws - requiring, for instance, mandatory against the developing country bloc. This bias public interest hearings for each case considered by increased in the years following the implementation the Tribunal, to explicitly consider the economywide of SIMA: while the share of imports from developing impact of various forms of policy intervention - are countries fell to 11 percent, their share of unfair trade readily feasible. cases increased to 44 percent. Unfortunately, international standards, as Dutz argues that an approach based on competi- codified in the GATT and as practiced by Australia, tion policy principles or on an economywide perspec- the EC, and the United States, weigh against Canada tive, by focusing on the broader impact of policies, modifying its current standard. Canada - the first offers an economically more rational way to deal with country to institute an antidumping system - is now issues currently addressed by unfair trade remedies. constrained from adopting more sensible policies by While unfair trade laws aim to protect domestic the weight and momentum of the system it helpeu to cemp-;.itors, competition laws strive to protect the develop. competitive process. Although existing competition 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 the series is to get these findings out quickly, even if presentations are less than fully polished. The findings, inte pretations, and conclusions in these papers do not necessarily represent official Bank policy. Produced by the PRE Dissemination Center Contents I. Introduction I II. Regulations to curb discriminatory pricing practices 2 II.1. Competition policy regulations and an economywide perspective 3 (i) Price discrimination 4 (ii) Predatory pricing 5 (iii) An economywide perspective 6 II.2. Antidumping and countervailing duty regulations 7 (i) Overview of "unfair" trade investigations 8 (ii) Ncrmal values and subsidies determination 12 (iii) Price undertakings as an avenue for compromise 16 (iv) Injury determination and the public interest provision 19 III. "Unfair" trade enforcement experience in the 1980s 22 III.1. Review of recent enforcement practices 23 (i) Frequency and disposition of cases 23 (ii) Industry and country incidence 25 III.2. Selected "unfair" trade cases 28 (i) The Grain Corn case 28 (ii) The Korean Cars case 29 (iii) The Women's Footwear case 30 III.3. The Canadian proposal for a cross-border competition law 32 IV. Conclusion 35 References 37 Notes 47 I. Introduction Structural features of the Canadian economy -- the relatively small size of the domestic market, a high level of industrial concentration in many sectors, and a substantial and increasing dependence on international trade -- have exerted an important influence on public policy. Canada was the first country to enact comprehensive antitrust legislation (1889) and the first to inst-Lute an antidumping system (1904).' Demand for competition policy legislation arose largely from concerns about the trusts that were restricting supply and raising prices behind newly erected protective tariff barriers; the earliest legislation attempted to control monopoly power abuses, apparently without regard to economic efficiency. Canada's original "unfair" trade legislation reflected a desire to prohibit predatory dumping, or pricing practices by foreign exporters designed "to crush out the native Canadian industry" (Canadian minister of finance, cited in Goldman 1987, 96). Although the rhetoric of Canada's first "unfair" trade legislation was antipredation, the legislation did not explicitly require evidence of predation. In principle, competition laws constitute a domestic counterpart to "unfair" trade laws, especially in the area of price discrimination and predatory behavior. In application, however, the two sets of laws have diverged widely. As competition laws were being refined and structured to protect the competitive process, antidumping and countervailing duty laws were increasingly focused on protecting domestic producers from foreign competitive pressures, and the antipredation element was quickly forgotten. Canada's competition and "unfair" trade laws underwent major revision in the 1980s, opening up potential new avenues for compromise on protection law. The revised antidumping and countervailing duty legislation included a new public interest provision for incorporating broader economic principles into the "unfair" trade remedies process in an attempt to ensure that decisions correspond to Canada's national economic interests. Increased concern for applying competition principles to cross-border discriminatory pricing issues 1 also emerged during this period in negotiations for the Canada-U.S. Ftee Trade Agreement. Canada's objectives during the negotiations were to eliminate all border measures to trade between the two countries. However, Canada's position was that some forms of discipline were likely necessary for particular pricing practices; as part of an overall approach to "unfair" trade practices, Canada sought to replace existing antidumping laws with a competition law based system.2 Despite changes in the legislation, however, enforcement practices have retained their protectionist bias, providing "relief" for interests that would nor gain relief under a conceptual framework emphasizing competition principles or an economywide perspective. This paper reviews Canada's recent enforcement of "unfair" trade laws and analyzes it against the objectives and standards of these two alternative means of regulat\ing cross-border pricing issues. It argues that both offer a workable and a more economically rational way of serving the national interest and responding to the problems now dealt with by the trade laws, though existing competition laws do not address perceived "unfair" foreign pricing practices originating in the public sector. The economic impact of the two approaches is roughly equivalent, but the economywide perspective fits in more readily than competition policy with the current administration of Canada's "unfair" trade laws. Only minor changes to existing laws would be needed to ensure that an economywide perspective guided both antidumping and countervail procedures. Even though the relevant GATT regulations would allow such efficiency-enhancing changes, international standards for "unfair" trade actions, as practiced by tha United States, the European Community (EC), and Australia, remain an obstacle to modification of Canada's current standard into an economically more sensible one. II. Regulations to curb discriminatory pricing practices Since important sections of Canada's competition legislation are concerned with discriminatory pricing practices, an overview of these laws 2 provides a context for evaluating Canadian "unfair" trade remedies enforcement. Trade and competition regulations dealing with discriminatory pricing practices are structured in a roughly similar fashion, moving from a determination of the "offense," to an "injury" test, and culminating in the application of "remedies."13 Both sets of laws are concerned with the impact of pricing practices on the domestic economy, but each tends to protect different economic interests. Competition laws strive to protect the competitive process, "unfair" trade laws to protect domestic competitors. 11.1. Competition policy regulations and an economywide perspective The revised Competition Act of 1986 states that the aim of competition policy in Canada is to "maintain and encourage competition in Canada in order to promote the efficiency and adaptability of the Canadian economy, in order to expand opportunities for Canadian participation in world markets while at the same time recognizing the role of foreign competition in Canada, ... and in order to provide consumers with competitive prices and product choices" (section 1.1). Encouraging competition has as a primary goal the efficient use of resources. The role of foreign firms in providing domestic competition where it may otherwise be lacking in Canada's relatively small economy is clearly recognized. Canada is unusual among industrial economies in its use of sector-specific, trade liberalization measures as remedies for antitrust violations .* Provisions of the competition laws dealing with price discrimination and predatory pricing are the most closely related to issues arising in transborder pricing practices. The new act strengthens the role of the director of investigation and research of the Department of Consumer and Corporate Affairs, Bureau of Competition Policy, as the general competition watchdog in regulatory proceedings5. The director has the capacity to address the competitive implications of decisions of the Canadian international trade authorities on transborder pricing practices during the public hearing part of 3 the process. Enforcement of the competition laws is governed primarily by the criminal law prohibitions of section 50 (on illegal trade practices). (i) Price discrimination. The Competition Act distinguishes between vertical or intercustomer price discrimination and geographic price discrimination. The law on vertical price discrimination governs the relationship between suppliers and buyers and is designed to prevent large buyers from gaining advantage over smaller ones. Enforcement does not require an inquiry into economic consequences since the specific elements that define the prohibited conduct do not require measurement. As in the related laws (where the word "policy" is used), the discriminatory pricing must be part of a "practice of discriminating"; isolated ircidents of discrimination -- say to meet aggressive spot competition -- are permitted. The geographic price discrimination law applies to suppliers who use different pricing practices in different geographic markets and is designed to prevent a lessening of competition through the subsidizing of sales in one region from receipts earned in another. In principle, this law has the greatest overlap with existing antidumping laws, the main superficial difference being whether markets are delineated domestically or internationally. However, in contrast to antidumping laws, geographic price discriminatior. and predatory pricing laws contain an "effect-on-competition" test. To obtain a conviction, the government must show "monopolizing design" or "a visible lessening of competition." Remedies are premised on preeerving or enhancing competition, and punitive sanctions (imprisonment for two years and/or large fines) are imposed. In practice, very few cases have been brought under these two provisions of the law, and there have been no convictions for geographic price discrimination -- a practice that conforms well with the prevailing view in economics that price discrimination is usually innocuous and is beneficial when it promotes competition (Varian 1989). Given the criminal standard of proof that must be met, the authorities have been reluctant to prosecute when the evidence showed only the accused firm's ability to weaken target firms or 4 to cause them to compete less vigorously, rather than the ability to force withdrawal from the market.6 When price discrimination demonstratea a predatory purpose, it should be regulated on that basis. (ii) Predator, pricing. Predatory pricing is often referred to as "primary line" prico discrimination since the pricing practice is directed at economic agents at the same level of production or distribution. Although price discrimination is not a constituent element of predation, predatory pricing does have an anticipated intertemporal discriminatory dimension, with the predator temporarily lowering prices in the expectation of substantially raising prices in the long erxr. The law itself does not require any evidence of price discrimination. It proh4.bits a "policy" of selling at "unreasonably low" prices with the "intention, tendency or effect of substantially lessening competition" (Competition Act, section 50[l][c]). The law covers both practices to drive out existing competition (classical predation) and practices to preempt or deter entry (strategic predation). As with price discrimination, the predatory pricing must be more than a short-term reaction to changes in the economic environment (such as aggressive pricing by competitors). The frequency of sales and the length of time goods are sold at the questionable prices must also be considered. Exactly what is meant by an "unreasonably low" price has been left for the courts to decide. The argument that any price below cost is unreasonable has been rejected. Case history suggests that prices are not considered "unreasonably low" if they cover variable costs but not full costs. And even when price is below average variable cost, all the circumstances in the case must be considered before it is judged unreasonably low'. Thus in virtually all foreseeable cases, the enforced pricing standard is lower than full current production and financial costs. In bringing predatory pricing cases to trial, the Bureau of Competition Policy may use the criminal proh'bition section 50(l)(c) of the law or it may proceed under the civil-law-based "abuse of dominant position" provisions (sections 78 and 79). The civil procedure involves a weaker standard of proof, 5 but it carries the extra burden of demonstrating that the offender is in a position of dominance. Importantly, the anticompetitive pricing must again pass an "effect-on-competition" test; the law also explicitly recognizes that superior competitive perrormance is not anticompetitive.8 Canada's competition law therefore imposes a stringent limit on the power of the state to discipline private actions. While no judicial authority exists for determining what constitutes a "substantial lessening of competition," legal decisions have argued that a motive of self-preservation or minimization of losses does not constitute an intent to lessen competition. (iii) An economywide perspective. As an alternative to competition policy regulations, an economywide perspective also provides a workable and economically more rational way to evaluate cross-border pricing problems than current "unfair" trade mechanisms. Under an economywide perspective, allegations of both "unfair" dumping and subsidization would be examined by assessing the economic efficiency impact on all agents within the economy (including producers, intE -mediate users and consumers) of existing pricing practices versus alternate feasible forms of intervention. Competing claims as to how benefits and injury should be distributed between domestic interests would be evaluated so that the chosen policy action results in the largest possible net economywide gains. Both competition laws and the economywide perspective view cross-border pricing problems from the similar economic perspective of intervening to promote an efficient allocation of resources throughout the national economy. Implementation mechanisms of the economywide perspective are very different from those of the competition policy approach, however. Competition laws intervene in the conduct of business and the structure of markets, seek to uncover violations of the criminal and civil legal codes, and rely on imprisonment or fines as remedies; competition laws could be adapted to deal with cross-border pricing practices of foreign private enterprises, but do not readily address pricing practices of foreign governments. An economywide perspective, on the other hand, would evaluate a contentious private or public 6 cross-border pricing practice by comparing its long-run benefits and costs to the economy as a whole to determine how, on balance, the national economic interest would best be served. II.2. Antidumping and countervailing duty regulations The imposition of antidumping and countervailing duties in Canada is governed by the Special Import Measures Act (SIMA) of 1984.9 SIMA was designed to overhaul previous legislation and "to make Canada's legislation more effective in protecting Canadian producers from dumped or subsidized imports" (Proceedings of the Standing Senate Committee on Banking and Commerce, January 28, 1984, as cited in Buchanan 1985, 2). SIMA replaced the formerly separate provisions of the Antidumping Act and the Countervailing Duty Regulations, introducing broadly parallel procedures. While this reduced the degree of ministerial discretion available in countervailing duty cases (Martin 1984, 4), it also meant that such cases were now on an equal footing with antidumping cases and therefore that countervailing duties could be applied more readily and broadly than before (Stegemann 1984, 44). The reforms followed the 1979 revision of the GATT Antidumping Code and the GATT Code of Subsidies and Countervailing Duties. While the review of Canada's existing legislation concluded that no changes were required for Canada to meet its GATT obligations, changes were deemed necessary to ensure domestic producers full recourse to the rights specified by the new GATT rules (Martin 1984, 2). Apparently, the government also felt some obligation to provide Canadian firms, which had cooperated with the government during the recent GATT negotiations, with a set of protective remedies similar to those of their foreign couniterparts, particularly since the United States and the EC had already implemented new legislation (Stegemann 1984, 37). The inclusion in SIMA, for the first time, of detailed legislation governing countervailing duties was related to the U.S. government's frequent use of such measures since the mid-seventies (Hart 1989b). 7 Another motivation behind SIMA was the desire to strer3then the government's ability to respond to the trade practices of other governments by ensuring that Canada had all the protectionist measures available to other nations to use as bargaining chips in future trade negotiations.10 This is an example of what may be termed the "worst common denominator" effect: Canada wanted to be sure to have a full arsenal of "unfair" trade remedies and a full complement of triggers for each to be in a good position for future bargaining over reductions. Given these motivations, it is not surprising that many discussants at a policy forum on SIMA concluded that the new legislation "showed a much more protectionist stance being taken by the government" (Stegemann 1984, 43). SIMA contains a two-track system for resolving domestic complaints of dumping and foreign government subsidies. The procedures are governed by detailed rules intended to insulate the process from direct political interference. The administrative determination of dumping or subsidy is made by the Assessment Programs Division of the Department of National Revenue, Customs, and Excise (hereafter referred to as the "Department"), while the determination of material injury to domestic production, and of causation, is made by an independent, quasi-judicial body, the Canadian International Trade Tribunal."' The responsibility for Canada's trade policy is split functionally between the minister of international trade and the minister of finance. (i) Overview of "unfair" trade investigations. A dumping or subsidy investigation typically begins in response to a complaint registered with the Department by a domestic producer or several producers."2 The Department has twenty-one days to determine whether the filing is "properly documented."" Once that is established the Department has thirty days to formally initiate an investigation or reject the complaint. To initiate an investigation, the Department must determine that the complainant represents a "major proportion of producers of like goods," 14 that there is evidence that goods have been 8 dumped or subsidized, and that reasonable grounds exist for believing that the domestic industry is threatened or has suffered maaterial injury.1" Because the preinvestigation filings are confidential, exporters and importers affected by a dumping complaint may be unaware of the complaint until a formal decision is made to initiate the investigation (Magnus 1989, 181). The only challenge to initiation of an investigation available to exporters, importers, foreign governments, or other interested parties concerns the Department's determination of injury. The advice of the Tribunal can be requested in such instances, but the Tribunal must restrict its investigation to the very preliminary information on injury received by the Department. (Before the SIMA legislation, not even this limited provision for challenge existed.) Following the formal initiation of an investigation, the Department generally has 180 days to complete its administrative determination of the dumping margin or the amount of the subsidy. This time limit, a new feature under SZMA, is more advantageous tc domestic complainants than to exporters or importers. Complainants have ample time prior to formal initiation to collect and submit information favorable to their position, while exporters and importers have comparatively very Little time, to collect information to argue their side of the issue. Within the first ninety days following the initiation of an investigation, the Department makes a preliminary determinati.on of the dumping margin or subsidy."' the preliminary determination is based on information supplied by questionnaires (subsequently verified) submitted to each exporter and importer known to have traded the goods under investigation. There are no provisions for holding hearings at which parties can present informaticn or views in an adversarial setting. While SIMA introduced a statutory right of access to all nonconfidential information (broadly similar to U.S. and EC procedure), in practice, no effective challenge is possible since confidential portions of submissions to the Department remain inaccessible." An 9 affirmative prel4minary determination results in the imposition of provisional duties -- the early deadline permits an early imposition of duties. Once the Department has issued a preliminary determination, the Trade Tribunal has 120 days to complete its inquiry on material injury and causation. The Tribunal's prokeedings under SIMA do not differ greatly from those under the former legislation, except that the Tribunal now :.s a public interest advisory role. For its injury inquiry, the Tribunal collects as much data a& it can for the four preceding years of the product under investigation; however, the analysis :s relatively simple and a model oi the market in the sense commonly understood by economists is not developed. The total time limit for "unfair" trade proceedings has been shortened considerably under SIMA, since the Department's final determination and the Tribunal's injury inquiry now begin concu rently (previously, the final determination began after the issuance of an injury finding). Following an affirmative preliminary determination, the Department has another ninety days to refine its initial estimate of dumping margins or subsidy levels and to issue a final determination of dumping or subsidy. During this second cycle, meetings are arranged upon request between the Department and affected exporters and importers to solicit additional information and to advise traders about the basis on which dumping margins were established. After an affirmative final determination by the Department, the Trade Tribunal, which has twenty-five to thirty days left to complete its own injury inquiry, enters the decisive phase of its deliberations with a formal hearing. The hearing is conducted in both public and confidential sessions as an aeversarial process between the opposing parties. Users of the product under investigation may also testify. In line with Canada's two-track system, however, the function of the Tribunal hearing is to allow a detailed probing only of evidence pertinent to the question of material injury; no discussion is allowed of the rate of duty or subsidy announced in the Department's final determination.1 The Trade Tribunal's finding of "injury," "partial injury," or "no injury" marks the end of the maximum 120-day period following the 10 Department's preliminary determination. The Tribunal's decision is final; no appeal or review of evidence is allowed. Thus the entire "unfair" trade remedies process, from receipt of a properly-documented complaint to the Tribunal's injury decision, usually occurs within 240 days or eight months. A finding of injury generally requires elimination of the full margin of dumping or level of subsidy determined by the Department: the Tribunal has no authority to require that duties cover only a portion of the dumping or subsidy margin.'9 Both provisional duties (on imports that entered between the time of the preliminary determination and the Tribunal finding) and final duties (on future imports) are assessed. In practice, exporters (and their counterpart Canadian importers, who by law must make the payments) avoid paying antidumping duties by raising their prices to normal values (discussed below); in the case of countervailing duties, the importer must pay duties equal to the assessed amount of subsidy (unless the exporting country government collects an offsetting export tax). Tha result is that domestic producers receive a level of protection equal to the full assessed dumping or subsidy margin. This relatively high level of protection stands in sharp contrast to the GATT's stated preference for measures that sliminate only material injury when the margin of injury is lower than the margin of dumping.2" Unless the Tribunal initiates a review and continues its affirmative injury finding, duties are rescinded after five years -- a new provision of the SIMA legislation. While this sunset clause ensures periodic review of definitive duties, it may actually increase the protectionist slant of enforcement by forestalling earlier reviews (as argued by R.Gotlieb and reported in Stegemann 1984, 45). On the other hand, any intereated party can make application to the Tribunal at any time following the injury determination to have the finding reviewed. Thus the remedies under the "unfair" trade laws increase market prices and effectively eliminate the foreign competition. And, as argued below, the remedies may even promote collusion. In contrast, comparable remedies under criminal and civil provisions of the competition laws are designed 11 specifically to alter anticompetitive behavior. The punitive sanctions (in practice, large fines) are designed to affect the profits of the offender (and offended, if recovery of damages is included) rather than to affect market prices, while remedial orders prohibiting the abusive behavior are structured to preserve or enhance competition. (ii) Normal values and subsidies determination. An understanding of how the "margin of dumping" and the "amount of subsidy" are determined is crucial in evaluating the impact of the "unfair" trade laws on national economic interests. Since exporters (in the case of a finding of injury due to dumping) generally will raise their prices fully to the level of normal value to avoid paying duties, the way normal values are calculated directly affects the ultimate level of protection received by domestic industries whose complaints are successful; similarly, in the case of a finding of injury due to subsidization, the way the amount of subsidy is calculated directly affects the domestic sales price of the imported goods. Although the new rules governing pricing determination are similar in substance to their predecessors, they are set forth in much more detail. SIMA defines the "dumping margin" as the difference between normal value and export price, expressed as a percentage of normal value. In general, establishing the export price is relatively straightforward: it is the exporter's ex factory (fob) sale price. Determining normal value requires more analysis since the objective is to calculate a representative sales price for similar goods in the exporting country's home market. Under Canadian law, the primary method for determining normal value is the home-country selling prlce approach,2' or the price charged in the home market for the product when sold to customers at the same trade level as the importer. This price is adjusted for such differences between home-market sales and sales to the Canadian importer as trade-level quantity, quality, transport, and terms of payment. If the exporter's sales are primarily for export, other sellers in the home country may be substituted for the exporter. A new provision of the SIMA legislation requires that prices be screened for 12 full cost recovery: home market sales at prices that do not provide for the recovery of full costs plus profits must be excluded (Buchanan 1985, 17).22 An exporter that is more efficient than Canadian competitors may be subject to antidumping duties even if the home-country selling price exceeds the export price for economically justifiable reasons (justifiable in the sense that the exports are not sold below cost but earn a profit). This situation can occur if markets are segmented and the exporter maximizes worldwide profits through international price discrimination; such behavior is welfare-improving to the extent that it results in higher sales. Where a "sufficient numb2t" of sales cannot be identified using the primary method rules, normal vli'.e 4s to be determined either by using a home- country export price approach or a constructed-value approach (SIMA, s.19), or if prices deviate from competitive market ones because of government- monopolized trade or price controls, by using a third-country selling price approach or a third-country constructed-value approach (SIMA, s.20). What constitutes a "sufficient number" of sales is left to the discretion of the Department; prior to SIMA, this determination was governed by explicit rules. Under the export price method, the price of similar goods sold by the exporter to importers in a country other than Canada is used, but only if the goods are not dumped in that market as well. The new restriction requiring the exclusion of all sales below full cost plus profit also applies for such export sales. However, the Department has rarely determined normal value according to this method (Magnus 1989, 196). The most common secondary method, the constructed-value approach, is a fully distributed or stand-alone costing approach. The constructed value is calculated from the exporter's actual total costs of production including both fixed and variable costs, plus an amount for profits (the exporter's weighted average profit rate on sales of like goods, or close substitutes thereof, if calculable, or an 8 percent rate otherwise, applied to the net-of-profit constructed cost). In contrast to the former legislation, SIMA provides more detailed directions for the determination of constructed cost, for example the 13 requirement that all costs directly attributable to the design and engineering of the goods be included. For establishing normal values in the special case of state-controlled economies, where export trade is monopolized by the government or domestic prices are controlled, the Department has the discretion of using either a method based on sales (the primary selling price approach but in a designated surrogate, third country) or a method based on costs of production (the constructed-value approach in a third country). If these methods are "unsatisfactory", the price of like goods exported to Canada from a surrogate country can also be used. Clearly, if labor and other production costs are lower in the state-controlled economy than in the surrogate country, assessed normal value will be significantly higher than the price paid by the Canadian importer. If sufficient information is not availeble for any of these secondary methods, SIMA specifies that normal vzalue is to be determined by ministerial discretion. As under the primary method, even a profitable exporter may be subject to duties under the two home-country based secondary methods of calculating normal value. Demand characteristics may lead a profit-maximizing exporter to charge a higher price in the third country than in Canada, but the export price approach would still result in a finding of dumping. And under all of these approaches, the excl'sion of sales below cost will result in overestimation of the economically-relevant costs. Under Canada's competition laws, the accepted methodology for calculating costs in order to determine whether a price is "unreasonably low" differs sharply from these approaches. Competition policy principles would not automatically view even a price below average variable cost as sufficient proof that a price is "unreasonably low"; the state of competition in import and export markets would also have to be explicitly considered. Competition law also requires that pricing behavior be evaluated as part of a "practice" or "policy" to avoid interfering with isolated pricing behavior undertaken for legitimate reasons. This is in sharp contrast to the constructed-value 14 approach, which may result in the finding of a dumping margin sufficiently large to restrict imports simply because of an exporter's failure to adjust export prices to new exchange rates without any lag -- even though this is clearly an artifact of short-run changes in exchange rates and depends on the time at which prices are measured. Analysis of data obtained form the International Economic Relations Division of the Department of Finance (IER) highlights the particularly distortionary methodology generally used for state-controlled economies. The level of economic development of surrogate countries usually has been far higher than that of the exporting country, leading to unreasonably high normal value figures and inflated dumping margins. Positive margins of dumping have been found in all investigations of goods exported from state-controlled economies.25 In the period before SIMA, for example, when the method of determining normal value was left to ministerial specification, all six investigations initiated against China used the third-country selling price approach, with Denmark, West Germany, Japan, Malaysia, the United Kingdom, and the United States as the designated surrogate countries.24 Five of the six cases resulted in definitive duties. During the SIMA period, four of the five investigations ir.itiated against Poland used the third-country selling price approach, with Brazil, France, West Germany and the United Kingdom as the surrogate countries; the fifth investigation used the third-country constructed-value approach, with Portugal as the surrogate country. Three of the five cases resulted in definitive duties. A comparison of the methods used for determining normal value during the 19P08 'n the pre-SIMA period (1980-84) and under SIMA (1985-89) reveals that the constructed-value approach has become more common under the new legislation (table 1). (Data on dumping investigations initiated during the 1980s, disaggregated by exporting country, were available from IER for over four-fifths of all industry-country pairs.) The home-country selling price approach, on its own, was used in roughly 20 percent of instances under both former and current legislation. Use of ministerial discretion as a means of 15 determining normal value has apparently diminished under SIMA. The constructed-value approach, on the other hand, alone or in combination with other methods, has been used more frequently under SIMA (the frequency of use of the constructed value approach applied on its own has almost doubled). In addition, the requirement under SIMA that all sales, even under the home- country selling price and third-country approaches, be screened for full cost recovery -- a procedural requirement that is not reflected in the classification by method -- marks an important difference between the two periods. In effect, this requirement means that the methodology of the constructed-value approach has become pervasive during the SIMA period. The method used for determining dumping probably has an impact on the type of exporting industries that are found to be dumping and so on the domestic industries most likely to launch effective complaints. Given the predominant use of the constructed-value methodology, exporting firms most likely to be found to have high assessed normal values and therefore positive dumping margins are those with significant fix2d capital costs of production, firms with lumpy capital requirements that routinely face excess capacity during the business cycle, and multiproduct firms that measure profitability on groupings rather than on specific products.25 Canadian law provides less direction in its treatment of the "amount of subsidy", and the meaning of "subsidy" is not limited by any precise definition in SIMA.26 The SIMA regulations (FTA revision 1989) provide rules for determining subsidy levels for grants, loans at preferential rates, income tax credits, refunds and exemptions, deferral of income taxes, over-refund of indirect taxes, and government procurement practices. The Department's view on "generally available" (and therefore not countervailable programs) versus "targeted" (and hence countervailable programs) is set out in an appendix to the recent Graln Corn case. Again, ministerial discretion is to be used when information is deficient. (iii) Price undertakings as an avenue for compromise. Foreign exporters or governments facing actions under the "unfair" trade laws may propose an 16 undertaking before the Department makes its preliminary determination; this option was not available under the former legislation. In the case of a dumping investigation, exporters may offer price undertakings, in essence commitments to raise the export price or to cease dumping. In the case of a subsidy investigation, foreign governments may offer a price undertaking -- to limit or remove the subsidy -- or a quantitative undertaking -- to limit the volume of exports. These short-cut agreements, if accepted by the Department, effectively terminate the investigation at an early stage. Undertakings are then subject to a three-year sunset clause; they must be reviewed prior to the expiration of this three year period. Although settlement by undertaking avoids the additional costs of a full-length proceeding,27 the decision has to be made by the exporter or foreign government before the preliminary determination is issued. In choosing this option, therefore, the accused party is required to restrict exports before knowing the outcome of either the preliminary "unfair" pricing investigation or the detailed injury investigation. Furthermore, any party involved in the investigation may veto the Department's acceptance of an undertaking within thirty days, in which case the investigation continues. Undertakings therefore do not undermine the interests of the complainants. The intent of the undertaking provision seems to have been to expedite the process and reduce costs (Martin 1984, 4). The years of experience with EC policy undertakings was crucial in the decision to include undertakings in SIMA. In Canada, undertakings were intended as a major innovation for regular use (Stegemann 1990, 27-9), a contrast to U.S. policy, which views them as an exceptional alternative to the full-length procedure. Under the EC system, there is also a revenue reason for seeking undertakings, since exporters (in the case of dumping investigations) can collect the revenues from price increases only when settling by price undertaking (customs authorities collect the revenues if EC duties are imposed). Under the Canadian system, exporters charged with dumping can routinely recoup the difference (since Canadian duties are not imposed on sales for which the price has been raised to the 17 assessed normal value). Apparently, the undertakings option was limited to the period before the Trade Tribunal phase of the investigation to ensure that the prozess would be significantly shorter than the full-length procedure and to avoid having "a bureaucrat undermining something that is goieg on before a court" (Stegemann 1984, 38). Two aspects particular to undertakings reflect competition policy concerns. One is that undertakings allow the application of measures that eliminate only the material injury whereas an affirmative injury finding in a full-length proceeding requires elimination of the full margin of dumping." This provision allows for compromise on the amount of the price increase required of exporters and, in principle, allows them greater market access. In practice, only five of the ten price undertakings accepted under SIMA during 1985-89 were intended to raise prices only enough to eliminate material injury rather to eliminate the full dumping margin.29 A second aspect of undertakings, one that has troubled Canadian competition authorities, is their potential to facilitate collusion (McDonald 1989, 63). Since the complainants provide information to the Department on acceptable price levels and are again consulted on prices before final acceptance of an undertaking, exporters could cooperate and exchange information and use the undertaking procedure to arrange international price cartels or market sharing. Undertakings substitute a bureaucratic process of trade management by government negotiation for a quasi-judicial Tribunal process open to the public. Two requirements offer some safeguards against collusive pricing agreements: an undertaking must not increase the price by more than the assessed dumping margin, and each exporter must give individual price undertakings. Unfortunately, this last provision "does not preclude discussions with...an association representing a group of exporters during the period leading up to the offer of an undertaking" (1987 edition of the SIMA Officers Manual, as quoted in Stegemann 1990, 61), though such discussions should be strictly for the purposes of explaining the process and Department requirements. However, different exporters can find common ground in such 18 meetings, and SIMA does not prohibit communication among suppliers. The rapid reaction of several exporters in submitting their proposals for price undertakings in the 1987 carbon steel welded pipe case suggests that exporters may well have communicated with each other (Stegemann 1990, 63). While the Canadian government is aware of the potential anticompetition effect of undertakings, the concern about facilitating collusion does not (but should) extend more generally to the antidumping mechanism itself, which has a price-raising impact similar in effect to that of collusive behavior. (iv) Injury determination and the public interest provision. The Trade Tribunal must determine whether the dumping or subsidizing found in the Department's investigation has caused or is likely to cause material injury to the Canadian industry producing "like goods." Because the procedures involved in this process -- determining like goods, defining the domestic industry, and determining material injury and causation -- all affect the competitiveness of the domestic economy, the Tribunal's role under SIMA includes a new public interest advisory function. The Tribunal's first task is to determine the class of like goods, defined as products that are identical in all respects to the product(s) under consideration or products that have "uses" and other characteristics closely resembling those of the product in question. In practice, "uses" has come to be determined by a "functional similarity" test based on market criteria: whether the goods can be substituted for each other and whether the same consumers are being sought. Apparently, the Tribunal is prepared to consider cross-elasticities -- to consider dissimilar products that compete with each other functionally as like goods. This trend is a desirable departure from the previous "identical/similar characteristics" definition based on physical features.'° In defining the domestic industry, Canadian law refers to "production in Canada" and "producers constituting a major proportion of total domestic production." Injury from the imports must be to domestic production for domestic consumption (the goods with which imports compete) rather than to 19 domestic production including exports. SIMA provides no guidance on what constitutes a "major" proportion of domestic production. In practice, the Tribunal has adopted a very loose standard that does not require the complainant producers to constitute more than 50 percent of the industry; at least eight cases in the IER database3' fall below this threshold, with the smallest share being as low as 35 percent. SIMA does not define the term "material injury," and whether material injury has occurred is left to the Trade Tribunal to determine.32 Certain key criteria have been applied in practice, however* A comparison of the criteria used by the Tribunal before SIMA (1980-84) and under SIMA (1985-89), based on a sample of over four-fifths of all investigations that resulted in the imposition of definitive duties, shows a strong similarity in emphasis (table 2). "Price suppression" was the most frequently cited injurious effect of investigated imports (25 to 30 percent of investigations), followed closely by "decline in market share" (22 percent of investigations). These were followed by "profitability," "output decline," "employment," and "growth". A wider variety of "other" injurious effects wes cited during the SIMA period than before. In general, many of the criteria were cited concurrently in a single investigation; in only nine cases out of 42 during the pre-SIMA period, and in one case out of 25 during the SIMA period was the application of only one criterion apparently deemed sufficient (either "price suppression", "decline in market share" or "future injury"). On the crucial question of causation, the Canadian procedures are considered to be "the most probative of all the GATT jurisdictions" (Magnus 1989, 217).33 Several factors are typically examined in determining whether a causal link exists between the "unfairly" priced imports and the assessed material injury: the actual and potential volume of "unfairly" priced imports in Canada, any significant increase in these imports, their effect on domestic prices or on production in Canada, and evidence of margin erosion, price suppression, or any significant undercutting of the prices of Canadian goods. Often central to the causation inquiry is the question of whether any loss of 20 market share by the domestic industry can be directly attributed to the "unfairly" priced imports. Findings of "no injury" have been issued in cases where the relevant cause (a declining market, lack of an established reputation, startup and production difficulties, or other factors) was not related to "unfairly" priced imports and in cases where the imports were found to be a cause of the assessed injury but not an important one.34 A comparison of ;hese procedures on injury and causality with injury standards under Canada's competition laws shows an important difference. All the factors typically examined under the "unfair" trade law focus on injury to domestic producers, whereas comPatition law practices focus on preventing injury to the competitive procoss -- pricing behavior generally is prohibited only if it lessens competition or eliminates a competitor. The "unfair" trade legislation does not preclude a finding of injury in the case where the importing country's industry is less efficient (less productive) than the exporting country's. A major new provision of Canadian "unfair" trade legislation (section 45) -- and one unmatched by U.S. or EC35 provisions -- empowers the Trade Tribunal to examine the impact of a positive injury finding and of asesessed duties on the public interest. Should the Tribunal find the collection of such duties to be contrary to the public interest, it must submit a report to the minister of finance advising what alternative level of duty would best serve the public interest. The law sets no lower limit, and the minister has the discretion to choose the final level or even to eliminate the duty completely. The concept garnered some support from developing countries in Geneva and was seen as a possible useful first step in having economywide concerns introduced into antidumping and countervail proceedings. The public interest provision was motivated by concerns that the previous legislation focused on domestic producer interests without adequately considering the effect of duties on downstream users and consumers. While the national interest provision was viewed as a safeguard for user-industries and consumer interests36 (a preference for low-priced goods), section 45 contains 21 no criteria for guiding the Tribunal's determination of what is in the public interest or which cases qualify for consideration. (The Tribunal is not required to consider the public interest in each case.) Neither does it contain any recommendations for subsequent action by the minister of finance once a report is received. The current legislation does not direct the Tribunal to adopt the type of methodology implicit in an economywide perspective, that is, it does not require that each antidumping and countervail investigation reaching the Tribunal be assessed by comparing its long-run benefits and costs to the economy as a whole to determine how the national economic interest would best be served. In the absence of other guidance, the Tribunal has reasoned that SIMA, like all Canadian legislation, was enacted in the interest of the public good, and that section 45, as a specific provision within the statute, "is to be applied on an exceptional basis, as for instance when the relief provided producers causes substantial and possibly unnecessary burden to users (downstream producers) and consumers of the product" (Canadian Import Tribunal 1987, 2). In practice, only three public interest hearings have been held sitce the enactement of SIMA,37 and only one (the Grain Corn case) resulted in a report to the minister. Despite considerable speculation that a public interest inquiry might be launched in the recent Women's Footwear case, this never happened. (Both these cases are examined in greater detail later in this paper.) It seems likely, in light of the Tribunal's stated intention to apply the public interest provision only on an exceptional basis, that the provision will be invoked primarily in response to organized pressure from consumer, intermediate producer, and user groups to protect their own interests. III. "Unfair" trade enforcement experience in the 1980s While an examination of the legislation can tell us something about the intentions of policy makers, only enforcement experience can tell us about the economic impact of the legislation. A broad comparison of enforcement practice 22 in the 1980s before and after enactment of SIMA indicates that the protectionist bias of Canadian practice has changed little, if at all. On balance, enforcement practice may even have become slightly more protectionist. An examination of selected "unfair" trade cases Allustrates the extent to which application of current legislation, including the public interest clause, diverges from the economically more rational practices that would prevail under a competition policy or an economywide perspective. And a discussion of the cross-border competition law proposal put forward by Canada during negotiations for the Free Trade Agreement with the United States reveals that more economically rational alternatives to current practices are workable. III.). Review of recent enforcement practices (1) Frequency and disposition of cases. The number of "unfair" trade investigationis or cases initiated during the 1980s gives a rough idea of how frequently these measures were used. A given case is directed at one or more exporters in one or more ccuntries; an industry may raise successive complaints over the years (or within a year) for similar products exported from different countries, leading to a succession of separate cases over that period.38 The focus on cases initiated (rather than on the number of countries with exporters accused of discriminatory pricing) is appropriate since the investigation of each case is triggered by the complaints of a particular domestic "industry." This makes the information particularly relevant for an assessment of the degree of protection provided through '"unfair" trade remedies practices since the number of cases corresponds to the number of instances that specific domestic industries potentially receive protection. Even if no dumping or injury is found, the number of cases initiated is indicative of the level of uncertainty created since exporters may respond to an increase in cases by raising prices themselves to avoid being challenged. 23 Of the 145 investigations initiated during the 1980s, 138 concerned dumping and 14 concerned subsidies (some cases involved both).39 While 82 cases were initiated in 1980-84, and 63 cases in 1985-89, it is not possible to conclude from the number of cases alone that there was any dramatic change in enforcement practice in the SIMA period (table 3). Changes in the macroeconomic environment probably explain much of the difference between the two periods. A lower number of cases would be expected during the post- recession upswing in Canada thas roughly coincided with the SIMA period than during the previous period, when cases were probably motivated by depressed- demand adjustment difficulties. One change in enforcement practice that doee seem related to the SIMA legislation is the increased use of countervailing duty remedies: from 6 percent (five of eighty-two cases) before SIMA to 14 percent (nine of sixty-three cases) under SIMA. The disposition of cases in the two periods also provides litte evidence of any marked change in enforcement practice. The proportion of cases resulting in "effective" price increases -- whether through the imposition of duties or through price undertakings (which are also formal agreements to raise export prices) -- went from 65 percent of the cases initiated in 1980-84 to 60 percent during the SIMA period. This finding suggests a relatively high but constant level of protectionism in practice. A comparison of the level of "effective" price increases as a percentage of assessed normal value in the two periods suggests that the level of protection provided to domestic industries by "unfair" trade remedies has actually increased under SIMA (figure 1). "Effective" price increases were derived from information on ninety-five country-specific observations for 1980-84 and seventy-five observations for 1985-89 for each country in wnich exporters were required to pay duties or had negotiated undertakings.40 A comparison of the two distributions shows that "effective" price increases resulted in higher absolute levels of protection, on average, in the SIMA period. Fifteen instances of price increases higher than 40 percent of normal value (or an equivalent duty rate of more than 67 percent) occurred in the 24 pre-SIMA period, and twenty-six in the SIMA period. Alternatively, given a decision to raise prices, the probability that the ptUce increase would be 40 percent or greater was 16 percent in the pre-SIMA period and a significantly higher 35 percent in the more recent years. The duration of the remedies imposed also affects the degree of protection provided by "effective" price 4ncreases. Changes in the stock of outstanding remedies provides an indication of their duration (table 4). (The stock of cases under enforcement for each year was calculated by taking the stock outstanding from the previous year and adding new remedies imposed and subtracting old remedies revoked.) The increase in the outstanding stock of "effective" price increases during 1980-84 was due to the large number of new remedies imposed in 1980-83 compared to the smaller number of existing measures revoked. This increase was no doubt largely attributable to the increased demand for protection by domestic industry during the recession of the early 1980s. The outstanding stock of remedies stabilized during 1985-89 but did not fall from its 1983 level. While it is too early to tell decisively whether the five-year sunset clause for antidumping and countervailing duties under SIMA is becoming a floor and having the effect of extending the average period of protection, available evidence suggests such a tendency. Of the measures initiated since mid-1984 under SIMA, only nine were reviewed before the end of 1989 (and only two of these were rescinded). Given the greater importance of undertakings in 1985-89 and their more frequent mandatory review (a three-year sunset clause), a more pronounced fall in the stock of cases could have been expected. But of the five undertakings that required a review before the end of 1989, only one has been rescinded. (ii) Industry and country incidence. The frequency with which particular domestic industries have used "unfair" trade remedies as a vehicle for seeking higher, less threatening prices from foreign exporters has been relatively constant across both periods (table 5). By far the largest number of complaints during both periods came from the primary metals group, which 25 accounted for over one-fifth of all complaints. Omitting the very diverse "miscellaneous manufacturing" group, the other industrial groups that were major users of "unfair" trade legislation during both periods were the metal fabricating, electrical products, chemical and petroleum, and food and beverages industries. A striking similarity among the range of products subject to complaints is that, generally, very few Canadian firms produce them. Given the narrow definition of the relevant product in most cases, Canadian "industry" is usually concentrated in the hands of a single or very few producers. The products are also internationally comparable and physically easy to trade. Taking the metals group as a standard, another common feature across the many targeted products is that barriers to entry are relatively low (it has been fairly straightforward, for example, for newly industrializing countries to enter the basic metals production industries). Conversely, barriers to exit are relatively high. Many of Canada's countervailing duty cases involve the food and beverages industry, mostly agricultural iports; of Canada's fourteen subsidy investigations during the 1980s, seven i lved agricultural products. In the small rLumber of dumping investigations involving crops, the domestic industry is, in each case, organized under either a provincially-sanctioned marketing board or a producer co-op (Lexenomics 1990, 3.4). The geographic incidence of "unfair" trade investigations initiated during the 1980s also demonstrates some clear patterns (table 6 and 7). Comparing cases involving developed and developing countries suggests a protectionist bias against the developing country bloc: while the share of imports from developing countries was 13 percent at the mid-point of the 1980- 84 period, its share of "unfair" trade cases was twice as large, at 26 percent (table 6). More disturbing for developing countries, this bias increased in 1985-89: while their share of imports fell slightly, to 11 percent, their share of "unfair" trade cases increased dramatically to 44 percent. Exporters from the Asian Tigers bloc (in particular from Taiwan), Eastern Europe, 26 Brazil, and Mexico were much more frequent targets of complaints in 1985-89 than in 1980-84. Without exception, the share of "unfair" trace cases initiated against all the developing country groups in table 6 exceeded their respective import share by a wide margin in the later period. Undoubtedly, a major reason that Eastern European countries and China have such high shares of "unfair" trade cases relative to their Canadian import shares is the particularly discriminatory methodology used to determine normal values for state-controlled economies. Given the overwhelming importance of the United States in Canada's trade flows, it is not surprising that U.S. exporters were subject to a large number of the complaints by Canadian industry against developed countries. However, the U.S. share of 'unfair" trade complaints in both periods (15 and 23 percent) is dwarfed by the U.S. share in Canada's imports, which was about 70 percent in both periods. Part of the reason for the discrepancy between "unfair" trade case shares and import shares is the importance of the Canada- U.S. Automotive Products Agreement, which created an integrated North American market in automotive products in which Canadian producers do not have incentives to bring complaints against their U.S. counterparts. The most-discriminated against countries -- those subject to more than five investigations during either five-year period, of which more than two- thirds resulted in "effective" price increases -- were Brazil, China, and Korea for the 1980-84 period, and Brazil and Taiwan for the 1985-89 period (table 7). All are within the developing country bloc. Brazil is included in both five-year periods, with "effective" price increases resulting for all six cases initiated against it in the earlier period and five of the seven cases in the later period. Brazil also was subject to the greatest number of subsidy cases (shown in brackets in table 7) durirg 1985-89; the other major target of subsidy investigations was the EC, which had four cases initiated against it during the 1980s. The most frequent user of price undertakings (shown in parentheses in table 7) was the United States, with five instances, followed by Japan, with two instances. 27 111.2. Selected "unfair" trade cases A more detailed examination of several "unfair" trade cases offers some insights into the forces involved in these investigations and the principles applied in their resolution. Of particular interest is the way in which the public interest provision figured in two of these cases. (i) The Grain Corn case. On May 12, 1986, the Department received a formal complaint from the Ontario Corn Producers Association alleging injurious U.S. subsidization of grain corn (excluding seed corn, sweet corn, and popping corn) exported to Canada. The Department initiated an investigation on July 2, 1986 and reached a final determination on February 2, 1987, that imports of grain corn from the United States were benefiting from a subsidy of US$0.849 per bushel (roughly CAN$1.10 at the time). On March 6, 1987, the Trade Tribunal announced a finding of material injury to the production in Canada of like goods. This case is particularly notable for two reasons. First, it represented only the second time that the United States had been subject to a formal subsidy complaint." Second, this case was the sole instance in which an application of Canada's public interest provision led to a reduction in the level of duties imposed. During the material injury inquiry, the Trade Tribunal received representations from several Canadian corn users (the feed industry, hog and poultry producers, industrial corn users, distillers and brewers) claiming that imposition of the duty would not be in the public interest; several hundred individuals also sent letters (Herman 1987, 417). The Tribunal examined the public interest issues and advised the minister of finance on October 20, 1987, that a duty higher than CAN$0.30 per bushel would not be in the public interest. The Tribunal argued that a duty above that level would provide little additional relief from injury to Canadian corn producers while creating a major irritant for corn users. Thus, the Tribunal argued that it had ample scope In this Instance to form an opinion pursuant to section 45 "without having to evaluate competing claims as to how injury and benefits 28 should be distributed between domestic interests...or without making use of new and unfamiliar welfare economics-based methodologies" (Canadian Import Tribunal 1987, 5-6). The minister, after receiving the public interest report, met with and received submissions from various interest groups, including the original complainants and corn users. On February 4, 1988, the minister announced that the duty was being reduced to CAN$0.46 per bushel and that the Tribunal would be asked to reconsider the public interest issue in roughly eighteen months; the review, dated December 29, 1989, recommended no further modification. As this case highlights, Canadian experience with its public interest provision has been very limited and disappointing from a national economic efficiency perspective. The provision has not been used as a mechanism to promote an economywide perspective: the domestic producers received as much benefit as was realistically possible from government protection, while alternative schemes that might better serve the national economic interest were not considered. Domestic producers gained at the expense of consumers. What was atypical in this case, however, was that users who would be hurt by higher domestic prices were able to organize themselves effectively to request a public interest hearing. And the hopeful sign, from an economywide perspective, was that this case demonstrated that consumers do have a forum in which their interests can be heard -- and that this process can result in policy changes. (ii) The Korean Cars case. On July 15, 1987, the Department initiated an investigation in response to complaints submitted by General Motors (Canada) and Ford (Canada) alleging dumping of cars produced by the Hyundai Motor Company of Korea. The Department's final determination of February 19, 1988 found dumping margins ranging from 6.5 percent to 61.9 percent, for an overall weighted average of 26.3 percent. The Trade Tribunal, however, ruled on March 23, 1989, that the dumping had not caused nor was likely to cause material injury to production of like goods in Canada, and so no duties were levied. 29 This case is notable primarily because of the active intervention of the director of investigation and research, in his role as the general Canadian competition watchdog, in two written submissions and in a formal representation at the Trade Tribunal's public hearing on competition and trade practices in the automotive industry. The director argued that Hyundai was a positive competitive influence, providing consumers with more competitive prices and greater product choices. He stressed the importance of such competitive influence and of not weakening or nullifying it by inappropriate application of trade restrictions. The director also noted that the Tribunal, in past cases, had taken into account the GATT directive that injuries caused by other factors must not be attributed to the dumped imports. In this connection, he cited growing world overcapacity for passenger car production, the displacement of Canadian production by "captive" imports, the use of incentives to eliminate excessive inventories, and the shift of consumer preferences toward smaller cars (Canadian Import Tribunal 1988a, 10-4). The director's participation apparently had an important impact: the Tribunal's summary rationale stressed that "Hyundai's imports 'Wre but one of many factors operating in a very dynamic market" and that "given the intense competition from other participants...price suppression cannot solely be attributed to the dumping.... As to the future, ... Hyundai will be but one of many participants in an increasingly competitive and globalized environment" (Canadian Import Tribunal 1988b, 28-9). (iii) The Women's Footwear case. In the first half of 1989, several meetings were held between the Shoe Manufacturers' Association of Canada and the Department to discuss complaints of dumping and subsidies of women's footwear. Following the receipt of an initial complaint lacking sufficient detail and a subsequent revised complaint, the Department initiated an investigation on August 25, 1989 of alleged dumping of women's footwear from Brazil, China, Taiwan, Poland, Romania, and Yugoslavia, and the subsidization of women's footwear from Brazil. In its final determination of April 3, 1990, the Department found weighted average dumping margins of 26.1 percent for 30 Brazil, 47.3 percent for China, 38.7 percent for Poland, 20 percent for Romania, 27.5 percent for Taiwan, and 26.2 percent for Yugoslavia; the amount of subsidy was calculated at between 3.5 percent and 17.4 percent. One month later, the Trade Tribunal announced a finding of material injury in all cases. The final outcome of this recent case raises two important issues. One concerns the substitution of trade remedies for other, less politically- acceptable forms of protection. Since the late 1960s, the domestic footwear industry had been facing growing import competition, especially from low-cost countries. In 1977, footwear imports represented 56 percent of the apparent Canadian market, an increase of 16 percent over 1973 (Canada 1987, 2). That same year, after a finding of injury by the Antidumping Tribunal, which explained the increased foreign penetration as a consequence of a competitive advantage arising from lower labor costs, a three-year global quota on nonrubber footwear was introduced. Women's footwear quotas actually remained in effect much longer, until November 1988. Three months later, the footwear industry initiated formal efforts to replace the expired quotas with trade remedies. Continued protection of footwear is particularly distressing in light of the findings of an empirical study that removing all footwear quotas for the 1978-83 period would have resulted in an average annual net increase in welfare of $8.58 million (in constant 1978 Canadian dollars), and total cumulative social adjustment costs of $0.84 million compared with cumulative gains in consumer surplus of $41.38 million (Moroz and Salembier, 1985). This case also demonstrates how infrequently, under current legislation, more rational principles from an economywide perspective (such as a broader public interest inquiry and active intervention, under the Competition Act, of the Director of Investigation and Research) are incorporated into enforcement practice. The Tribunal, in issuing its reasons for the injury decision, invited submissions from parties who wished to conduct a public interest inquiry. The Tribunal received only one submission, from China, which it did not find convincing, and announced a decision not to conduct a public interest inquiry. 31 Why was there no domestic response to this issue? Perhaps the parties felt that they would be unable to affect the result, believing that the Tribunal had bent over backwards to see the facts from the domestic industry's perspective. Or perhaps they had other, more urgent concerns. Whatever their reasons, the case highlights the dependence of a public interest inquiry on an exceptionally strong lobby of users, intermediate producers, and consumers. Similarly, participation by the Director of Investigations and Research is likely to be ccntingent on several factors or constraints, including an evaluation of the economic stakes involved. In this case, it may be that the decision not to participate was based on a comparison of the competitive status of the industry and its importance in the overall economy against the availability of financial resources and the relative expertise of staff in the Department of Consumer and Corporate Affairs. Important (though readily feasible) changes in the existing legislation would likely be required before principles in line with an economywide perspective could be consistently applied in Canada. 111.3. The Canadian proposal for a cross-border competition law In negotiations for the Free Trade Agreement (FTA), Canada and the United States had implicitly decided that multilateral channels and procedures were not, at that time, the preferred route for meeting their objectives. One proposal put forward by Canada called for each country to replace existing antidumping laws with compatible competition policy procedures. This proposal was not adopted, but an examination of why it was rejected may help in assessing what scope exists for a greater emphasis on competition-based principles in the enforcement of "unfair" trade policies. A little background on recent trade disputes between Canada and the United States is useful for an understanding of Canada's position in the FTA negotiations. Canada, with its much smaller economy, is far more dependent on international trade than is the United States. While many Canadian firms export a substantial proportion of their production to the United States, very 32 few U.S. firms are in a similar position relative to Canada. There is further assymetry between Canada and the United States in their ability under the GATT codes to take countermeasures to subsidies. Small export-dependent economies have much less scope to engage in subsidy practices without challenge and less scope to counter others' subsidy practices than larger, less export-dependent economies.'1 That Canadian exports to the United States were more likely to be countervailed than U.S. exports to Canada was an important factor in persuading business leaders and government officials in Canada to pursue bilateral negotiations for fundamental reforms. During the FTA negotiations in 1985-87, Canada raised the issue of whether participation in a free trade zone should involve not only elimination of tariffs but also coordination of competition procedures". Since the type of price discrimination motivated by trade barriers at the border would no longer be possible, firms involved in bilateral trade would follow the types of pricing strategies that are currently regulated by domestic competition laws. The standards governing pricing strategies would then be based on injury to competition rather than injury to competitors. Canada proposed that, following the removal of tariffs, a set of compatible domestic competition laws should regulate the pricing strategies of firms involved in bilateral trade. Because the competition laws in both countries had attained a relatively similar stage of development, it was deemed feasible to deal with undesirable private cross-border pricing practices using price discrimination and predatory pricing laws. The laws on price discrimination would require elimination of several differences in substance. Common rules would be needed to reduce jurisdictional conflicts and to facilitate the gathering of evidence, and special rules would need to be developed for handling third- country price discrimination, since this was not covered in existing national laws. On the issue of remedies, a remaining obstacle was the divergence between the treble-damages rules of U.S. antitrust laws and the relatively weaker Canadian remedies. However, the group of trade and competition 33 officials that examined the proposal concluded that, at a technical level, all these deficiencies could be resolved. A stalemate over reform of the subsidies-countervailing duty issue seems to have been the main reason the proposal was not included in the final agreement. In this area, Canadian objectives could not be met, given prevailing political imperatives in the United States. The devised solution included in the FTA, however, did recognize that "unfair" trade laws should be applied in a bilateral fashion: dumping and subsidy disputes between Canada and the United States are now subject to bilateral review, and the decisions of a bilateral panel are final and binding." It is interesting to speculate whether a country may have more success in a multilateral context in pushing new rules based on competition principles. In the EC, cross-border sales within the free trade zone are treated as domestic sales, exempt from member nations' antidumping and countervailing duty laws; EC competition laws regulate the cross-border pricing behavior of member country firms. Apparently, introducing such changes was possible in the case of the EC member countries because they shared the joint goal of a common market and because member countries believed that their individual rights were protected by the existence of two supranational institutions, the European Commission and the European Court of Justice; it may have also been that at the time when these changes were introduced, individual member countries did not know what potentially powerful protectionist policy tools they were giving up. The United States was clearly unwilling to follow the EC precedent in dealing with Canada. In the Canada- U.S. negotiations, each party had fundamentally different perceptions of the problem. But in a multilateral forum, a small economy might have stronger negotiating possibilities since its own problems may be common to many participants. On the other hand, the regime now in place between Australia and New Zealand that has replaced antidumping measures with provisions of their respective competition laws suggests that the scope for desirable changes clearly also exists within bilateral contexts."5 34 IV. Conclusion Canada's recent experience with enforcement of "unfair" trade laws shows little evidence of influence from competition policies or an economywide perspective. The result has been high levels of protection for a small number of well-organized firms. Throughout the 1980s, the methodology for calculating dumping margins and assessing injury generally led to affirmative determinations in instances that would not be considered actionable under competition law. Serious predatory, anticompetitive concerns were probably not at issue in any of the cases in which antidumping duties were assessed. Canada's recent overhaul of its "unfair" trade legislation was not followed by any dramatic change in enforcement practice, as a comparison of practices before and after SIMA shows. In fact, a comparison of the actual levels of "effective" price increases suggests that the protectionist bias of Canadian enforcement has increased. Information on the broad regional pattern of "unfair" trade investigations suggests that the bias against exports from developing countries has also increased significantly in the years following the implementation of SIMA. This paper has argued that an approach based on competition policy principles or on an economywide perspective, by focusing on the broader impact of policies (beyond the direct benefits received by the few firms seeking preferential treatment), offers a more economically rational way of dealing with the issues currently addressed by "unfair" trade remedies. The concrete proposals Canada presented during the FTA negotiations to regulate Canada-U.S. cross-border pricing issues by competition principles demonstrate that the competition policy alternative is workable. However, as Canada's "unfair" trade laws are currently administered, the economywide perspective fits in more readily than competition policy. Relatively minor changes to existing laws -- requiring, for instance, mandatory public interest hearings to explicitly consider the economywide impact of proposed measures by making a modified section 45 an automatic structural component of each antidumping and 35 countervail case considered by the Tribunal -- are readily feasible from the legislative and administrative points of view. Unfortunately, international standards, as codified in the GATT and as practiced by the United States, the EC, and Australia, weigh against Canada's modifying its current standard. Canada, the first country to institute an antidumping system, is now constrained from adopting more sensible policies by the weight and momentum of the system it helped to develop. 36 References Australia, Attorney General's Department. 1990. "Review of Developments of International Trade: Recent Developments in Trans-Tasmar. Business Law." (Paper presented at the XVIIth International Trade Law Conference, September 1-2, 1990). Mimeo. Boner, R.A., and R. Krueger. 1991. "The Basics of Antitrust Policy: A Review of Ten Nations and the EEC." World Bank, Industry Series Paper No. 43, Washington, D.C. Buchanan, J.N. 1985. "Antidumping Law and the Special Import Measures Act." Canadian Business Law Journsl 11: 2-22. Canada, Department of Finance. 1980. "Proposals on Import Policy: A Discussion Paper Proposing Changes to Canadian Import Legislation" (the "White Paper"). Ottawa. Canada, Department of Finance. 1985. "Annex of Outstanding Antidumping Actions." In Semiannual Report of Antidumping Actions, July 1, 1984- December 31, 1984. Ottawa. Canada, Department of Industry, Science, and Technology. 1987. Industry Profile: Footwear. Ottawa. Canada. 1988. Competition Act. (Office Consolidation) R.S. 1985, c. C-34, various amendments, Ottawa. Canada. 1989a. Special Import Measures Act. (Office Consolidation) S.C. 1984, c. 25, amended by 1985, c. 14, 1986, c. 1, Ottawa. Canada, Department of Revenue, Customs and Excise. 1989b. "Special Import Measures Act Regulations." (FTA Revision, March 31, Memorandum D14-1-1). Ottawa. Canada, Department of Supply and Services (Statistics Canada). Various years. Canada Year Book. Ottawa: Minister of Supply and Services. Canadian Import Tribunal. 1987. "Public Interest: Grain Corn." Report of the Canadian Import Tribunal under Section 45 of the Special Import Measures Act, October 20. Ottawa. Canadian Import Tribunal. 1988a. "Preliminary Submission of the Director of Investigation and Research appointed under the Competition Act." (January 6). Ottawa. Canadian Import Tribunal. 1988b. "Additional Submission of the Director of Investigation and Research appointed under the Competition Act." (February 8). Ottawa. Canadian Import Tribunal. 1989. Annual Report 1988. Ottawa. Chen, T. 1987. "Are the anti-dumping laws of Canada and other Western countries keeping pace with China's economic structural reform?" Law and Policy in International Business 19: 717-64. Feltham, I.R., S.A. Salen, R.F. Mathieson, and R. Wonnacott. 1990. Competition (Antitrust) and Antidumping Laws in the Context of the Canada-U.S. Free Trade Agreement. (A Study for the Committee on Canada-United States Relations of the Canadian Chamber of Commerce and the Chamber of Commerce of the United States). Drart. General Agreement on Tariffs and Trade (GATT). 1980. "Agreement on Implementation of Article VI of the General Agreement on Tarilfs and Trade." In Basic Instruments and Selected Documents. Twanty-Sixth Supplement. Geneva. Giese, J. 1987. "The Special Imports Measures Act: Balancing the Interests of Foreign Exporters and Canadian Industries." Journal of World Trade Law 21: 9-25. Gillen, M.R., L.A.W. Hunter, D.E. Rosenthal, and W.T. Miller. 1987. "Canadian and U.S. Antitrust Law: Areas of Overlap Between Antitruct and Import Relief Laws." Canada-United States Law Journal 12: 39-73. Goldman, C.S. 1987. "Competition, Antidumping, and the Canada-U.S. Trade Negotiations." Canada-United States Law Journal 12: 95-106. Green, C. 1990. Canadian Industrial Organization and Policy. Toronto: McGraw- Hill Ryerson. 37 Grover, W., and R. Kwinter. 1987. "The New Competition Act." The Canadian Bar Review 66: 267-304. Hart, M. 1989a. "Dumping and Free Trade Areas." In J.H. Jackson and E.A. Vermulst, eds., Antidumping Law and Practice. Ann Arbor: University of Michigan Press. Hart, M. 1989b. "The Canada-United States Working Group on Subsidies: Problem, Opportunity or Solution?" (Octobet draft). Center for Trade Policy and Law, Carleton University, Ottawa, Canada. Herman, L.L. 1987. "Injury Findings by the Canadian Import Tribunal: The Decisive Elements." Review of International Business Law 1: 373-418. Hunter, L.A.W., and J.F. Blakney. 1987. "The Changing Canadian Competition Law Environment: Implications for International Business." Review of Internstional Business Law 1: 235-57. Lexenomics Inc. 1990. "The Relationship between Competition Policy and Antidumping Law: The Canadian Experience." (January). Ottawa. Magnus, P.A. 1989. "The Canadian Antidumping System." In J.H. Jackson and E.A. Vermulst, ed., Antidumping Law and Practice. Ann Arbor: The University of Michigan Press. Martin, R.J. 1984. "Background and Main Elements of the Special Import Measures Act." In K. Stegemann, rapporteur, Report of the Policy For-m on Special Import Measures Legislation. Kingston, Ontario: John Deutsch Institute for the Study of Economic Policy. Maule, C.J., and T.W. Ross. 1990. "Canada's New Competition Po'icy." George Washington Journal of International Law and Economics 23:59-109. McDonald, F. 1989. "The Relationship between Antidumping and Competition Policies." in P. Slayton and S. Cass, eds., Trade Law Topics. Toronto: Butterworths. McFetridge, D.G., and S. Wong. 1985. "Predatory Pricing in Canada: The Law and The Economics." The Canadian Bar Review. 63:685-733. Metzger, S. 1979. "The Amended Antidumping Code and the Trade Agreements Act of 1979." In J. Quinn and P. Slayton, eds., NonTariff Barriers after the Tokyo Round. Montreal: Institute for Research on Public Policy. Moroz, A.R., and G.E. Salembier. 1985. "A Quantitative Assessment of the Costs and Benefits of the Footwear Import Quota." Discussion Paper in International Economics. Ottawa: Inetitute for Research on Public Policy. Paterson, R.K. 1986. Canadian Regulation of International Trade and Investment. Toronto: Carswell. Porteous, S.D., and A.M. Rugman. 1989. "Canadian Unfair Trade Laws and Corporate Strategy." Review of International Buslness Law 3: 237-70. Rugman, A.M., and S.D. Porteous. 1989. "Canadian and U.S. Unfair Trade Laws: A Comparison of Their Legal and Administrative Structures." The Canadian Business Law Journal 16: 1-20. Stegemann, K. 1982a. "The Efficiency Rationale of Antidumping Policy and Other Measures of Contingency Protection." In J. Quinn and P. Slayton, eds., NonTariff Barriers After the Tokyo Round. Montreal: Institute for Research on Public Policy. Stegemann, R., rapporteur. 1984. Report of the Policy Forum on Special Import Measures Legislation. Kingston, Ontario: John Deutsch Institute for the Study of Economic Policy. Stegemann, K. 1990. "Price Ledertakings to Settle Antidumping Cases." (February draft). Department of Economics, Queen's University, Ontario, Canada. Stegemann, K. 1991. "Settlement of anti-dumping cases by price undertaking: Is the E.C. more liberal than Canada?" In P.K.M. Tharakan, ed., Policy Implications of Antidumping Measures. Amsterdam: Elsevier Science Pu lishers B.V., North-Holland. Varian, H.R. 1989. "Price Discrimination." In R. Schmalensee and R.D. Willig, eds., Handbook of Industrial Organization. Vol.I. Amsterdam: Elsevier Science Publishers B.V., North-Holland. 38 Table 1 Determination of normal value in Canadian 'unfair" trade cases: frequency of methods applied, 1980-84 and 1985-89 (percentages) Before SIMA Under SIMA Method (1980-84)a (1985-89)b Home-country selling price 19 19 (exclusively) Home-country selling price 14 1 (together with ministerial specification') Third-country selling price 4 12 (exclusively) Third-country selling price 1 0 (together with ministerial specification0) Combination of home-country and 1 0 third-country selling price Constructed-value 12 21 (exclusively) Combination of constructed-value 33 39 and other' methods Ministerial specification' 16 8 (exclusively) Note: Data for each investigation were aggregated by country of export rather than by exporter, so more than one method was often reported for a given industry- country pair. While methods used in combination may be indicative of general trends, the information is less precise since the underlying data does not indicate for each exporter involved in an investigation what percentage of its goods had normal values established on each basis. a. Sample of 141 antidumping industry-country pairs. b. Sample of 98 antidumping industry-country pairs. c. "Ministerial specification" was also at times referred to as "best information available." d. "Other" includes simultaneously one or more of home-country selling price, home-country export price, third-country selling price, third-country constructed- value and ministerial discretion. Source: Author's calculations based on data from the International Economic Relations Division, Department of Finance. 39 Table 2 Determination of material injury in Canadian "unfair" trade cases: frequency of criteria applied, 1980-84 and 1985-89 (percentages) Before SiMA Under SIMA Criterion (1980-84)" (1985_89)b Price suppression 29 25 Decline in market share 22 23 Profitability 17 18 Output 10 6 Employment 5 5 Growth 3.5 3 Productivity 3.5 0 Otherc 10(11) 20(16) Sales loss/revenue loss (6) (6) Future injury/threat of injury (5) (5) Bankruptcies (0) (1) Bounded injury (Aug. 15-Apr. 1) (0) (1) Crop disposal (0) (1) Financial losses (0) (1) Injury in a previous case (0) (1) Land-value decline (0) (1) Note: Statistics are based on the subset of Tribunal investigations that resulted in positive findings of material injury for which information is available. For most cases, more than one criterion is listed; when the "other" category is broken down, more than one criterion is also sometimes listed. a. Sample of forty-two cases. b. Sample of twenty-five cases. c. Number in parentheses is number of times the criterion was used. Source: Author's calculations based on data from the International Economic Relations Division, Department of Finance. 40 Table 3 Disposition of Canadian "unfair" trade cases during the 1980s, 1980-84 and 1985.89 (number of cases) Total Terminated before Calendar number of primary Price Terminated Duties year cases initiated determination undertakings at Tribunal assessed Before SIMA 1980 14 2 0 3 9 1981 12 1 0 3 8 1982 25 [11 3 [1] 0 7 15 1983 16 [1] 1 0 5 [1] 10 1984 15 [31 0 1 4 10 [31 1980-84 82 [5] 7 [11 1 22 [1] 52 [3] Under SIMA 1985 20 [21 3 2 3 12 [21 1986 14 (3] 4 2 2 (2] 6 (1] 1987 13 [1] 0 2 6 5 [1] 1988 8 (11 2 1 2 [1] 3 1989 8 [2] 2 2 1 3 [2] 1985-89 63 M9] 11 9 14 [3] 29 [6] Total 145[14] 18 [1] 10 36 [4] 81 9M Note: Numbers in brackets refer to cases that included subsidy investigations. Source: Author's calculations, from the database of the International Economic Relations Division, Department of Finance. 41 Table 4 Canadian "unfair trade cases: flows and stocks, 1980-84 and 1985- 89 (number of cases) Initiated in Stock year ending New remedies Old remedies outstanding 31 December to be imposed; revokedb at year endc Before SIMA 1980 9 5 44 1981 8 4 48 1982 15 3 60 1983 10 3 67 1984 11 9 69 During SIMA 1985 14 14 69 1986 8 5 72 1987 7 6 73 1988 4 10 67 1989 5 5 67 a. All cases initiated in a given year that resulted in "effective" price increases (antidumping or countervailing duties or undertakings to raise export prices). b. Includes only cases in which injury findings were rescinded; does not include alterations. c. Data for 1984 are based on Canada (1985), with the addition of three countervailing duty cases that resulted in duties and one case initiated in 1984 that resulted in an undertaking. "Stock" for other years are approximate and are calculated as the difference between imposed and revoked remedies based on initiation dates and not on dates of duty effectiveness. Source: Author's calculations based on data from the International Economic Relations Division, Departmen. of Finance; Canadian Import Tribunal (1989) for data on "old remedies revoked"; and Canada (1985) for "stock outstanding at year end." 42 Table S Canadian "unair" trade cases Initiated during the 1980s by Industry, 1980-84 and 198S49 (number of cases) Ind' or P 1980 1981 1982 1983 1984 1980-84 1985 1986 1987 1988 1989 1985-89 Food and beverages 0 0 2 3 2 8 3 3 0 2 0 13 Textile Industries 0 1 3 1 0 6 0 1 0 0 1 3 Clothing and footwear 0 1 4 0 0 6 0 0 0 0 1 1.5 Wood and paperproducts 1 1 0 0 2 5 0 0 0 0 1 1.5 Chemical and petroleum 3 1 2 1 0 8 3 1 3 0 1 13 Nonmetallic minerals 0 0 0 2 1 4 0 2 1 0 0 5 Primarymetal industries 2 1 11 2 5 26 2 3 3 2 0 16 Furniture and fixtures 1 0 1 0 2 5 0 2 0 1 0 S Metal fabricating 1 3 1 1 0 7 3 1 1 1 1 11 Machinery (nonel-ctric) 1 2 0 0 0 4 1 1 1 0 1 6 Transport equipment 1 0 0 0 0 1 1 0 2 0 0 S Electrical products 3 2 0 2 1 10 3 0 0 2 1 9 Miscellaneous manufacturing 1 0 1 4 2 10 4 0 2 0 1 11 Total 14 12 25 16 15 100 20 14 13 8 8 100 Source: Author's calculations, from the database of the International Economic Relations Division, Department of Finance, Canada. 43 Table 6 Canadian "unfaire trade cases initiated during the 1980s: shares in cases and imports by developed and developing country groups (percentages) "Unfair trade investigations share Import share Group/country 1980-84 1985-89 1982 1987 Developed countries 74.0 56.3 86.9 88.7 United States 14.5 22.7 70.5 68.0 Othere 59.5 33.6 16.4 20.7 Developing countries 26.0 43.7 13.1 11.3 Asian Tigersb 10.4 16.0 3.1 4.5 Korea (7.5) (7.6) (0.9) (1.6) Taiwan (0.6) (5.9) (1.0) (1.7) Eastern Europe' 5.8 11.8 0.4 0.3 Brazil 3.5 5.9 0.7 0.7 China 3.5 2.5 0.3 0.7 Mexico 0.6 2.5 1.5 1.0 Other 2.2 5.0 7.1 4.1 a. Includes all Western European countries, Japan, Australia, and New Zealand. b. Includes Hong Kong, Korea, Taiwan, and Singapore. c. Includes the Soviet Union. Source: Author's calculations based on "unfair" trade inves.igations data from the International Economic Relations Division, Department of Finance, and import shares data from Statistics Canada, Canada Year Book. 44 Table 7 Canadian 'unfair" trade cases and 'effective" price increases initiated during the 1980s, by country, 1980-84 and 1985-89 (in descending order by number of investigations initiated) Investigations "Effective" price increases Country 1980-84 1985-89 1980-84 1985-89 United States 2s 27 [1] 16 (1) 14 (5) Japan 12 11 8 7 (2) West Germany 17 5 9 4 (1) Korea 13 9 10 5 Italy 16 [21 3 5 2 France 12 4 [11 6 2 United Kingdom 9 5 5 4 Brazil 6 7 [5] 6 5 (1) Spain 8[1] 2 4 1 Belgium 6 3 4 2 China 6 3 5 3 Taiwan 1 7 0 5 Poland 2 5 2 3 Sweden 3 3 3 2 Netherlands 5 [1) 0 4 0 Czechoslovakia 3 1 2 1 European Community 2 [2] 2 [2] 2 1 East Germany 1 3 1 2 (1) Luxembourg 3 1 1 1 (1) Mexico 1 3 0 1 Romania 2 2 1 0 Singapore 2 2 2 2 Hong Kong 2 1 2 1 Malaysia 1 2 1 2 Yugoslavia 1 2 1 2 (1) Finland 3 0 0 0 Argentina 0 2 0 2 Austria 1 1 1 1 Portugal 2 0 1 0 South Africa 2 0 1 0 USSR 1 1 1 0 Chile 0 i 0 0 Denmark 1 [1] 0 1 0 Norway 1 0 0 0 India 1 0 1 0 Ireland 1 0 0 0 Switzerland 1 0 0 0 Turkey 0 1 0 0 Total 173 [71 119 [9] 106 (1) 76(12) Note. Numbers in brackets are subsidy investigalions; numbers in parentheses are price undertakings. Source: Author's calculations based on "unfaie" trade remedies data from the International Economic Relations Division, Department of Finance and import shares data from Statistics Canada, Canada Year Book. 45 Figure 1 Canadian "unfair trade cases: levels of protection, 1980-84 and 1985-89 Number of exporting countrles 40* SO - 20 - --- 10 - I 1 0 1-10 11-20 21-30 S1-40 41-60 81-60 et-76 Prie Incresse. (% of normxl value) M 1980-1984 _ 1986-1989 Inoludes ail availgble Information on casee where dutle were asseessed or undertakings were ofttred Source: Author's calculations from the database of the International Economic Relations Division, Department of Finance. 46 Notes I am grateful for helpful comments from J. Michael Finger, Klaus Stegemann, and participants at a Word Bank workshop on antidumping. I also appreciate discussions with officials at the Canadian International Trade Tribunal; the Department of Consumer and Corporate Affairs, Bureau of Competition Policy; Department of Finance, International Economic Relations Division; and the Department of Revenue, Customs, and Excise, Assessments Program, Customs Programs Branch; and with individuals at the Institute for Research on Public Policy and the Center for Trade Policy and Law. 1. Canada's first antitrust or competition law was passed in 1889 (Act for the Prevention and Suppression of Combinations Formed in Restraint of Trade) and precedes the U.S. Sherman Act (1890). Price discrimination and predatory pricing laws were added in 1935, when it became clear that certain pricing practices helped powerful rather than efficient firms. Canada invented the first antidumping system in 1904 (An Act to Amend the Customs Tariff). A provision for countervailing duties (against perceived "unfair" trade practices originating in the public rather than private sector) was not added until 1955 (section 7), however, and regulations to make it effective were not introduced until 1977. For an overview of the history of Canada's competition laws, see Maule and Ross (1990, 63-74); on antidumping law, see Magnus (1989, 174-8); and on antisubsidy law, see Hart (1989b, especially 40-2). 2. Although this ambitious proposal did not become part of the final Agreement, it is nevertheless indicative of Canada's position on the general principles underlying the proposed regime. 3. As discussed below, an injury test (that is, an "effect-on-competition" test where the pricing activity is measured by its effect, intended effect, or tendency of lessening competition or eliminating a competitor) is not a required element in the law on vertical price discrimination. 4. This point is made forcefully in a recent review of the competition policies of ten nations and the EC (Boner and Krueger 1991, 41 and 118). The Competition Act, for instance, provides for customs duty reduction as an alternative to asset divestiture to remedy mergers that violate the Act. This provision was applied successfully as part of the remedy in the 1989 Asea Brown Bovery - Westinghouse case; the merger was allowed to proceed without divestiture as long as existing import tariffs were removed and A.B.B. undertook not to initiate any antidumping investigation to protect its position for a period of five years. 5. The director already had authority to appear before federal boards, commissions, or other tribunals. The new Act, by authorizing appearances before provincial bodies, strengthened the director's capacity to intervene to promote competition and efficiency (Maule and Ross 1990, 61). 6. For a succinct discussion of the few actions that have been brought under the Competition Act's subsections 50(l)(a) (on vertical price discrimination) and (b) (on geographic price discrimination), see Green (1990, 379-80). For a more detailed discussion on price discrimination and predatory pricing and a 47 comparison with U.S. laws, see villen et al. (1987, 39-57) and Feltham et al. (1990, 45-102). 7. The principle of what constitutes an "unreasonably low" price is discussed at length in the Hoffmann-LaRoche case, as summarized in McFetridge and Wong (1985, 690-5). 8. The Competition Tribunal issued its first decision under the "abuse of dominant position" provision on October 5, 1990, in The Director of Investigation and Research v. The Nutra!veet Company. The case is relevant to the application of competition law to transborder business practices since it involves a U.S. enterprise's activities in Canada as well as imports from several sources. For a detailed discussion of the civil law provisions and this particular case, see Feltham et al. (1990, 62-75). 9. SIMA is the principal Canadian legislation for invoking all "trade remedy" actions, including GATT Article XIX type safeguard actions. The second principal statute concerning trade remedy actions is the Canadian International Trade Tribunal Act of 1988, which united the agencies responsible for international trade matters (the Canadian Import Tribunal, the Tariff Board of Canada, and the Textile and Clothing Board) into an amalgamated Canadian International Trade Tribunal; the procedures followed in antidumping and countervailing duty cases remained unchanged. 10. Minutes of the Proceedings and Evidence of the Subcommittee on Import Pollcy (1981-82), cited in Rugman and Porteous (1989, 3); see also Martin (1984) and Hart (1989b, 40). 11. Since the basic functions of this body have not changed substantially over the years (whereas its name has, from Antidumping Tribunal to Canadian Import Tribunal to Canadian International Trade Tribunal), it will be referred to consistently as the "Trade Tribunal." 12. An investigation can also be initiated on the Department's own initiative, although this option rarely occurs in practice. 13. Canadian producers can take as long as is necessary to complete a "properly documented" filing. The complainant must provide all supporting material, including detailed information regarding Canadian production of the goods, evidence of dumping or subsidizing, and of material injury. If the complaint is not properly documented, the Department will specify in detail what additional information or evidence is required; in this sense, Canadian producers are receiving invaluable assistance not available to foreign exporters or Canadian importers. 14. In the United States, the authorities assume that the complainant represents a majority of producers, and responsibility for disproving this assumption rests with the exporter or foreign government (Rugman and Porteous 1989, 5). Although the Department takes a more active role in this determination, several petitions covering less than half of Canadian production have been accepted, as is discussed later in this paper. 48 15. Since the Trade Tribunal is the specialist body on matters relating to injury, the Department's injury inquiry at the initiation stage is less formal and based on much less information than the Trade Tribunal's subsequent inquiry (Magnus 1989, 181-2). 16. For complex cases, this can be extended to 135 days. Also, the Department may terminate the investigation before this point if it finds that evidence is insufficient to justify proceeding, that the margin of dumping, the volume of dumped goods, or the amount of subsidy is negligible, or that new evidence suggests no reasonable indication of injury. 17. Although the Disclosure of Information provisions of SIMA (sections 82-88) permit access to confidential information on the discretion of the Deputy Minister, current administrative practices by Revenue Canada reportedly prevent such access (Magnus 1989, 180). 18. As discussed later, the public interest clause provides a caveat to this strict separation in cases where imposition of a duty in the full amount might not be in the public interest. 19. Interventions resulting in conceivably different price changes are only possible under two mezhanisms: price undertakings (which must occur before the beginning of the TrJbunal's injury investigation) and the public interest provision (which allows for lower duties at the discretion of the minister of finance); these are discussed later in the paper. 20. Among principal users of antidumping measures, only the EC has adopted this recommendation (Stegemann 1990, 54-5). 21. Section 15 of SIMA outlines the determination of normal value of goods, basically a rewording of paragraph 1 of Article VI of the GATT Antidumping Code. Note that "primary" here does not mean most common practice but rather the first method that should be used, if possible. 22. Before the Department accepts a series of sales as the basis for normal value determination, the series is examined to determine if it is profitable on average. If so, the series could be used but all sales within that series, even those at a loss, are then considered in the calculations. 23. For printed evidence until mid-1986, see Chen (1987, 728); state- controlled or non-market economies subject to "unfair" trade investigations during the 1980s include China, Czechos.ovakia, East Germany, Poland, Romania, USSR and Yugoslavia. Even in post-SIMA cases where undertakings were accepted, estimated margins of dumping were substantial, as reported in Table 7 by Stegemann (1990, 163). 24. In the case of hydraulic turbines exports from Japan with components shipped directly from China to Canada (ADT-9-84), a Japanese exporter involved argued that the surrogate country chosen for the preliminary assessment of normal value of Chinese components should be changed in view of .he different levels of development of China and Japan. The subsequent replacement of India for Japan as surrogate country resulted in a reduction of the assessed dumping margin from 61 to 34 percent (Chen 1987, 731-3). 49 25. For a listing of several otheL relevant characteristics of firms most likely to offend constructed-value rules, see Lexenomics (1990, 4.9-4.10). 26. According to SIMA (section 2[l]), "Subsidy includes any financial or other commercial benefit that has accrued or will accrue, directly or indirectly, to persons engaged in the production, manufacture, growth, processing, purchase, distribution, transportation, sale, export or import of goods, as a result of any scheme, program, practice or thing done, provided or implemented by the government of a country other than Canada, but does not include the amount of any duty or internal tax imposed on goods by the government of the country of origin or country of export from which the goods, because of their exportation from the country of origin, have been exempted or have been or will be relieved by means of refund or drawback." While the U.S. countervailing duty statute does not define what constitutes a countervailable subsidy, the Canadian definition is so broad as to be little different in practice. 27. These costs include legal expenses as well as the possible payment of duties and all the costs associated with uncertainties surrounding the investigation (at least four months of uncertainty regarding the outcome of the case are saved since the undertaking normally must be settled within the first 120 days following receipt of a properly documented complaint). 28. In addition to being in line with the GATT code, the use of measures that likely raise prices by less that an otherwise-assessed dumping margin is motivated by a practical concern: assessed dumping margins are often so high that they are prohibitive, completely eliminating exports (Stegemann 1990, 55). However, it is somewhat problematic to make such comparisons on a case- by-case basis since undertakings must be accepted before there has been any offical "assessment" of the margin of dumping/subsidization (that is, before a preliminary determination is made). 29. Since four of the the five cases concerned several countries with considerable variation in dumping margins, acceptance of undertakings eliminating full dumping margins would have resulted in large variations in prices. So pragmatism probably played as important a role as any concern for competition promotion. The determination of price increases sufficient only to eliminate injury is difficult in practice, especially since the Department's injury investigation is very limited: the Department generally asks the complainant for an estimate of the price required to eliminate injury, based on domestic full-cost prices plus a reasonable profit. The intended effect of the other five price undertakings was to eliminate dumping (Stegemann 1990, 57-9). 30. This focus on "uses" was absent from previous Canadian legislation. The U.S. statute contains a similar reference, but the 1979 GATT code does not. For a more detailed discussion of the "like goods" determination and the reference to the actual use of the cross-elasticity concept, see Magnus (1989, 202-8). 31. Out of a total of 145 cases initiated during the 1980-89 period, as will be reviewed below. 50 32. SIMA refers only to "material injury to the production in Canada of like goods" (section 2[1]) and provides no illustrative lists of factors, indices, or thresholds of injury. In contrast, the U.S. legislation explicitly sets out factors in line with the GATT codes. However, the Tribunal has issued a set of rules that includes guidance on the evidence that should be presented on this issue, in line with the GATT codes (see Herman 1987, 391-3, and the citations therein). 33. See Rugman and Porteous (1989, 9-10) for arguments that the U.S. International Trade Commission has not analyzed causality in any meaningful way and that the search for simple correlations predominates (for example, it is often deemed sufficient if imports are seen to increase at the same time that industrial performance declines). 34. Herman (1987, 395-98) concluded from an analysis of recent cases that the Trade Tribunal applies a higher standard of Injury than that suggested by the 1979 GATT codes. 35. The EC code served as a model for section 45; although it does not concern the "public interest", it does require gains or losses from dumping to be examined from the community point of view, that is, not just from the producer's point of view (Rugman and Porteous 1989, 10). However, the EC's motivation seems to be largely political (to promote greater cohesion) rather than efficiency-based, though some commentators have argued that Canada's motivation was also largely political. 35. Concerns and recommendations were expressed in deliberations of the Parliamentary subcommittee examining the proposed legislation. Canadian legislators apparently did not want to stray too far from GATT code criteria since they wanted to ensure that Canadian industry received the same level of protection as comparable industries in other countries. For reflections on this issue and explicit reference to consumer interests, see the "Minutes of the Proceedings and Evidence of the Subcommittee on Import Policy," as quoted in Rugman and Porteous 1989, 10. 37. Surglcal Adhesive Tapes and Plasters from Japan (initiated on May 8, 1985), Grain Corn from the U.S. (initiated on July 2, 1986), and Yellow Onions from the U.S. (initiated on October 14, 1986). 38. For example, two separate sets of dumping complaints from the Canadian small electric motors industry (polyphase induction motors, 1 to 200 HP inclusive) resulted in the initiation of two distinct investigations in 1985, the first on February 7 (against Brazil, Japan, Mexico, Poland, Taiwan, and the U.K., including a concurrent subsidy investigation against Brazil) and the second on July 5 (against Romania). These are included in the database as two separate cases; while the first case resulted in the imposition of antidumping duties (and countervailing duties against Brazil), the second case was terminated following a finding of np material injury. 39. Under SINA, allegations of dumping and subsidy can be investigated in the same case. Of the fourteen subsidy investigations, seven were of this type (one case in each of the years 1983, 1984, 1985, 1987, and 1988, and two cases 51 in 1989); this explains why the total number of trade remedies cases is less than the sum of dumping and subsidy investigations. 40. Wh n information on final measures imposed was unavailable, preliminary determination levels were used. No information at all was available for eleven country-specific observations in the pre-SIMA period, and for one observation in the SIMA period. An equivalent rate of duty as a percentage of export price can easily be computed from the average mandated price increase. Let the reported duty (dumping or subsidy margin) be denoted by DM, the assessed normal value by n, and the export price by x. Given that DM - (n-x)/n, then the equivalent rate of export duty is given by DC[1/(l-DH)] - (n-x)/x. 41. The other case, brought by the EC against petrochemical feedstock, was unusual since it did not arise from industry complaints or considerations of injury and was settled before any duties were imposed (Hart 1989b, 41). 42. According to the GATT codes, subsidies with an export-enhancement effect (a subsidy to firms th.,t export a large proportion of their production to one country might be viewed as having this effect) can be unilaterally countervailed by an importing country. Subsidies with an import-displacement effect can only be countered through multilateral procedures. For a more detailed presentation of the points raised in this context, see Hart (1989b). 43. The following discussion is based on an excellent summary of the issues involved, the negotiations and the solution, in Hart (1989a, 336-42). For a detailed examination of the feasibility of relying on each country's existing competition laws in dealing with antidumping complaints in the context of the FTA, see Feltham et al. (1990). 44. Under Chapter 19 of the FTA, the Ca.aadian and U.S. governments agreed to work toward the development of a substitute system of rules for antidumping and countervailing duties as applied to their bilateral trade; a bilateral working group was created with the development of such a substitute system within seven years as one of its main objectives. 45. Australia and New Zealand moved toward bilateral free trade through the 1983 Australia-New Zealand Closer Economic Relations Trade Agreement, with full free trade in goods between the two countries achieved in 1990. Article 4 of the Protocol on the Acceleration of Free Trade In Goods (signed in August 1988) required that competition remedies replace antidumping proceedures between the two countries. These obligations were fulfilled in Australia by the Trade Practlces (Misuse of Trans Tasman Market Power) Act 1990, and in New Zealand by the Commerce Law Amendment Act and the Judicature, Evidence and Reclprocal Enforcement of Judgments Amendment Acts, laws which came into operation on July 1, 1990. For a discussion of the associated new investigation and enforcement procedures for business regulation in the free trade area, see Australia (1990). 52 PRE Working Paper Series Contact AuTAtL hor for pLWr WPS758 Is rhere Excess Co-Movement of Theodosios B. Palaskas August 1991 D. Gustafson Primary Commodity Prices? A Panos N. Varangis 33714 Co-Integration Test WPS759 The Profamilia Family Planning Jesus Amadeo August 1991 0. Nadora Program, Columbia: An Economic Dov Chernichovsky 31091 Perspective Gabriel Ojeda WPS760 How Conflicting Definitions of Alexander J. Yeats September 1991 J. Jacobson "Manufactures' Distort Output and 33710 Trade Statistics WPS761 Uncertainty and the Discrepancy Gerhard Pohl September 1991 P. Lee between Rate-of-Return Estimates Dubravko Mihaljek 81950 at Project Appraisal and Project Completion WPS762 Debt, Debt Relief, and Growth: Daniel Cohen September 1991 S. King-Watson A Bargaining Approach Tnierry Verdier 31047 WPS763 A Valuation Formula for LDC Debt Daniel Cohen September 1991 S. King-Watson 31047 WPS764 African Financing Needs in the 1990s Jorge Culagovski Victor Gabor Maria Cristina Germany Charles P. Humphreys WPS765 Withholding Taxes and International Harry Huizinga September 1991 S. King-Watson Bank Credit Terms 31047 WPS766 Economic Crisis, Structural Francois Diop September 1991 0. Nadora Adjustment, and Health in Africa Kenneth Hill 31091 Ismail Sirageldin WPS767 Framework for Macroeconomic Colin A. Bruce September 1991 M. Lynch Analysis (Applied to Kenya) David Ndii 34046 WPS768 Going to Market: Privatization in Manuel Hinds September 1991 L. R. Hovsepian Central and Eastern Europe Gerhard Pohl 37297 WPS769 Entry-Exit, Learning, and Lili Liu September 1991 D. Ballantyne Productivity Change: Evidence from 37947 Chile WPS770 Privatization in Eastern and Central Farid Dhanji September 1991 CECSE Europe: Objectives, Constraints, Branko Milanovic 37188 and Models of Divestiture PRE Working Paper Series Contact In^ AdAhor for par WPS771 Macroeconomic Structure and Ibrahim A. Elbadawi September 1991 S. Jonnakuty Policy in Zimbabwe: Analysis and Klaus Schmidt-Hobbel 39074 Empirical Model (1965-88) WPS772 Macroeconomic Adjustment to Oil Ibrahim A. Elbadawi September 1991 S. Jonnakuty Shocks and Fiscal Reform: Klaus Schmidt-Hebbel 39074 Simulations for Zimbabwe, 1988-95 WPS773 Are Ghana's Roads Paying Their Reuben Gronau September 1991 J. Francis Way? Assessing Road Use Cost and 35205 User Charges in Ghana WPS774 Agricultural Pricing Systems and Mark Gersovitz October 1991 B. Gregory Transportation Policy in Africa 33744 WPS775 The Macroeconomics of Public William Easterly October 1991 R. Martin Sec.:or Deficits: A Synthesis Klaus Schmidt-Hebbel 39065 WPS776 Enforcement of Canadian "Unfair" Mark A. Dutz October 1991 N. Artis Trade Laws: The Case for Competition 37947 Policies as an Antidote for Protection WPS777 Do the Benefits of Fixed Exchange Shantayanan Devarajan October 1991 A. BhalIa Rates Outweigh Their Costs? The Dani Rodrik 37699 Franc Zone in Africa WPS778 A Dynamic Bargaining Model of Eduardo Fernandez-Arias October 1991 S. King-Watson Sovereign Debt 31047 WPS779 Special Programme of Research, Janet Nassim October 1991 0. Nadora Development and Research Training 31019 in Human Reproduction WPS780 Optimal User Charges and Cost Ian G. Heggie October 1991 P. Cook Recovery for Roads in Developing Vincy Fon 33462 Countries