Does Anti-dumping Enforcement Generate Threat?

Transcription

1 MPRA Munich Personal RePEc Archive Does Anti-duming Enforcement Generate Threat? Sagnik Bagchi and Surajit Bhattacharyya and Krishnan Narayanan Indian Institute of Technology Bombay. February 04 Online at htt://mra.ub.uni-muenchen.de/5367/ MPRA Paer No. 5367, osted. February 04 4:5 UTC

2 Does Anti-duming Enforcement Generate Threat? Sagnik Bagchi, Surajit Bhattacharyya and K. Narayanan Abstract The last two decades have witnessed that countries across the world are guided by the rules and regulations of multilateral trading institutions (for examle, World Trade organization [WTO], International Monetary Fund [IMF]) in order to romote free and fair trade through gradual reduction in trade barriers. The World economy has noticed significant reduction in tariffs, yet we find a rise in non-tariff barriers (NTBs). However, we still find duming and few other trade strategies of the exorting countries as a major hindrance to free and fair trade. Such behaviour has led to contingent rotection as a tool of new-rotectionism. Among the contingent rotection measures, anti-duming (AD) has evolved as the most oular choice of strategy for the trading nations. The AD olicy invokes a threat to the exorter and thereby can change its strategic behaviour. We describe the henomenon of duming through a rice-leadershi model and thereby comute the otimal level of antiduming duty that can offset duming. Using a seuential game, we conclude that the credible threat of an AD duty restricts duming and thereby leads to a win-win situation for both the foreign and domestic firms. JEL: C0, C7, D43, F3, L40 Keywords: Price-leadershi, Duming, Anti-duming duty, Seuential game This aer has been ublished in Foreign Trade Review, February (04), Vol. 49 (): The authors are associated with the Deartment of Humanities and Social Sciences, Indian Institute of Technology Bombay. Corresonding author: bagchi.sagnik@gmail.com

3 . Introduction The ost-war eriod has witnessed significant growth of international trade which was mainly guided by the multifarious statues of the multilateral trading institutions (for examle, The General Agreement on Tariffs and Trade [GATT], IMF, WTO, etc.). The major objectives of these organizations were to romote free and fair trade. One of such means was by gradually bringing down the tariff rates, abolishing uota and other trade restrictions. Initially, the emerging economies (e.g., India, Argentina, Brazil and South Africa, among a few others) were not in favour of acceting the new trade olicies framed by these institutions. These economies were scetical because exosing the domestic economy to international cometition can create distortions in the form of material injury to the firms in the resective home country. In resonse to the divergence in trade olicies of develoed and develoing countries, WTO [GATT (948) and in subseuent rounds of deliberations] rovided a latform for the harmonization of the same and thereby romoting world trade. One such olicy measure was roviding rotection through contingent rotection. Contingent rotection comrises of AD measures, Countervailing and Safeguard measures. AD duty is used by the imorting country to offset the rice advantage exloited by the foreign firms on their low riced exorts. An AD initiation has two comonents: (i) a duming comonent based on rice discrimination (i.e., exort rice below the normal value) and (ii) the material injury. Under the WTO rules [Uruguay Round (994)], AD duty can only be alied if these two conditions are satisfied simultaneously. Countervailing measures on the other hand, hel the domestic firms in offsetting the advantages that the foreign firms ossess in terms of an exort subsidy given by their resective home countries. Lastly, the safeguard measures allow a grace eriod to those imorting countries whose domestic industries have lost comarative advantage due to the cometition faced from huge imorts. The intricate rocess of the safeguard and countervailing measures make them uite conservative in nature among the trading countries. On the contrary, AD rocedures are flexible and gets over within a eriod of 8 months; moreover, it is case secific which is unlike the tyical tariff olicy. It is argued that AD has a uniue combination of economic and olitical alicability (Prusa, 00). On the other hand, Finger (993), demonstrates that Material Injury to any industry means the actual or otential decline in sales, rofit, outut, market share, roductivity, return on investment, etc. See, Article 3 of the AD agreement of GATT (994) for details. AD initiation refers to the legal rocedural action taken by an imorting country in order to investigate an imorted roduct subseuently taken u by government authorities in both imorting and exorting countries.

4 anti-duming as ractised today comrises largely of bad economics and ower olitics. 3 (See, Bekker, 006, for similar arguments). AD law justifies the trade restriction it creates. Moreover, this law allows meagre firms to initiate AD cases, bring imorts under scrutiny and thereby rotect themselves from cometition. Otherwise, the low riced exorts can be countered with orthodox trade rotectionist measures. The main theme of the aer is to exlore whether the (credible) threat of an AD enforcement by the imort cometing domestic firm would alter the decision of foreign exorting firm with regard to selling below the normal value.. Anti-duming Its Poularity and Concerns Market distortions in the form of duming signal rice changes that over time lead to various market adjustments. If these disturbances are temorary, then there would be few adjustments. One such temoral adjustment is imosition of an anti-duming duty. The emergence of new rotectionism in the form of anti-duming across the develoing nations seems to be mostly concerned with safeguarding the domestic industries against these market distortions. Sometimes, nations use AD initiations as a threat to counter attack low riced exorts or to rotect their own sick industries. The dawn of the st century marked the centennial year (004) of the anti-duming measures and also by this time it had become the (trade) olicy choice for both develoing and develoed nations. Particularly, in the last two decades new AD users (comrising India, Argentina, Brazil, Mexico and South Africa, among a few others) have outaced the traditional AD users (U.S., Canada, Australia, New Zealand and the Euroean Community) in terms of AD filings. In the immediate aftermath of the Uruguay Round, these new AD users had initiated AD comlaints at an unrecedented rate. Comared to 980s when only a handful of nations had anti-duming codes, over 40 nations had joined the AD club in the 990s. The significant surt in AD activities across the world over the eriod has led to the dilemma of whether anti-duming is actually a form of rotection from genuine harmful 3 Practitioners of anti-duming laws have mechanized ways to convert initiations into measures. Turning down an anti-duming initiation by the government allows the ractisers to find another means to make the case into an affirmative one. 3

5 ractices or is it just a modified form of a rotectionist measure? 4 Recent available data reveal that AD initiations constitute around 89 er cent of the total contingent rotection measures all over the world and are significantly greater when average tariff rates across the globe have been declining (Figure ). Figure : Distribution of Various Contingent Protections, Data Source: WTO Reorts on Anti-duming, Safeguard and Countervailing Initiations Since the Uruguay Round till date, the WTO data secify India as the to users of AD. The first AD initiation of India was in 99 against the U.S., Jaan and Brazil for the roduct PVC resin. In articular during the eriod 995-0, India has filed 6 er cent of all global anti-duming cases, uite disroortionate to its share in global imorts (.04% in 00-). 5 As the economy rogressively oened u under the statutes of multilateral trading institutions, India s share in global AD activities has also increased. 4 Prior to the 980s there were less than 00 AD cases across the world; whereas, in the eriod between 995 and 0 there were more than 50 cases on an average every year. 5 Authors calculation based on 00- data available in the WTO Statistics data base. 4

6 Figure : Anti-duming Initiations and Degree of Oenness in India Data Source: WTO Reorts on AD Initiations; WTO Statistics Database and The World Develoment Indicators (World Bank). Note: India s Share in AD Initiations is defined as a ratio of India s AD Initiations to World AD Initiations. Degree of oenness is calculated as a ratio of trade to GDP (in current year rices). Most of the roducts which face AD initiations and measures are roducts of chemical and allied industries, lastic and rubber articles, textiles, base metals, machinery and mechanical aliances. These industries alone initiated around 9 er cent of AD cases during the eriod Out of these initiations, around 6 er cent have been converted into measures. With so many initiations across India and other nations, there is a concern among scholars of whether anti-duming has moved away from the original intentions that created it (i.e., rotection from redatory duming). We do not answer such uestions in this aer. However, we believe that credible threat of AD enforcement would alter trade ractices of the foreign (exorting) firm with resect to selling below the normal value and move towards free and fair trade. The itinerary of rest of the aer is as follows. Section focuses on the main theme of the aer. With the backdro of a rice leadershi model, we argue whether threat of an AD duty on the foreign firm, can effectively eliminate exorts below the normal value. Alongside, we suort the argument through a game theoretic exercise. Section 3 concludes the aer. 5

7 . The Model In this section we argue that the threat of an otimal ad-valorem AD duty would alter the ricing decision of the foreign firm. For similar arguments see, (Reitzes, 993; Tivig and Walz, 000). With this roosition, we take into account the sustained material injury to the domestic firm and thereby calculate the otimal duty reuired to offset duming. We consider a duooly model involving the domestic firm and a foreign firm, both roducing like roduct. 6 The technologically suerior foreign firm is the rice leader. The foreign firm (Firm ) exorts amount of outut to the home country. It enjoys a cost advantage in roduction of and can sell in the home (exort) market below its normal value. On the other hand, Firm (the domestic firm) roduces amount of outut and is assumed to be the single roducer of the roduct in its home country. In the home market, both the firms face a linear market demand function a b a, b 0. 7 Suose that the cost functions for firm and are: c ( ) F ; 0 and ( ) F c F and F denote fixed costs for firm and, resectively. Total market outut is. The rofit-maximization roblem for the domestic firm (Firm ) in is: max ( ) C( ) F a b F Hence, the domestic firm s monooly euilibrium rice-outut combination will be: M a b, M ab ( ) bb ( ) The maximum rofit of the domestic firm is: a F bb ( ) auturky However, in the ost-trade scenario, the domestic firm can no longer charge the monooly rice as it faces cometition from the foreign firm. As a result, the domestic firms lobby to 6 Identical Products that are alike in all resects or in the absence of such a roduct, another roduct which although not alike in all resects, has characteristics closely resembling those of the roduct under consideration. For details see, Article.6 of the AD agreement of GATT (994). 7 In, the domestic firm also faces the same market demand function. We assume that firm does not face any caacity constraint. 6

8 their government authorities for rotection. One such olicy choice by the domestic government is AD initiation. Considering the lausible actions of both the foreign and domestic firm, we comute and comare the otimal rice, outut and rofit conditions under each of the following strategy combinations. Case I: Exorts Below Normal Value with No AD Initiation With trade, the domestic firm would accet the rice (as arameter) set by the foreign firm, and choose its outut level. 8 The ost-trade rofit maximization roblem for the domestic firm can be written as: max F The follower (Firm ) would choose an outut level where rice is eual to its marginal cost. With as the (arametric) revailing rice in the domestic market, the foreign firm would be serving the residual demand. The rofit maximization roblem of the foreign firm is: a max b b F The trade euilibrium rice-outut combinations for the two firms are: The rofit of the two concerned firms will be: a ( b ) ( ) ; a b ( b ) { a ( b )} { a ( b )} 4( b) 8( b ) F ; F In order to highlight the sustained material injury and argue in favour of the AD olicy we comare the re-trade otimal rice, outut and rofit level with the corresonding trade euilibrium rice, outut and rofit values for the domestic firm. See, Table in which we consider some hyothetical values for a, b and. 8 being the exort rice which is below normal value. 7

9 Table : Analysis of The Model Based on Hyothetical Values Values Autarky Post-Trade Material Injury a Firm suffers 3.9% and b % decline in outut 49.67, 45.75, share and rofit levels, , resectively. a 00 b a 00 b a 00 b a 00 b a 00 b a 00 b a 00 b , , , 4.87, 903., , 38.66, 745.7, , 36., 64.4, , 33.58, 00, , 5.5, , , 7.67, 484, 98 4, Firm suffers 7.87% and 9.8% decline in outut share and rofit levels, resectively. Firm suffers.06% and 89.68% decline in outut share and rofit levels, resectively. Firm suffers 3.34% and 87.48% decline in outut share and rofit levels, resectively. Firm suffers 6.05% and 85.9% decline in outut share and rofit levels, resectively. Firm suffers 3.49% and 80.49% decline in outut share and rofit levels, resectively. Firm suffers 9.3% and 75.0% decline in outut share and rofit levels, resectively. Firm suffers 30% and 70.59% decline in outut share and rofit levels, resectively. Had the domestic firm, acceted the revailing exort rice, it would not have been able to sustain the losses, as it can nowhere reach the rice level that the foreign firm charge. We argue that an AD duty would yield the best result for the domestic firm as the magnitude of material injury sustained to the domestic firm (Firm ) varies with the elasticity of demand and it is case secific. 8

10 Also, it is found that unlike orthodox rotectionist measures, the rate of an AD duty can be varied deending uon the magnitude of material injury. Further, one can say from fig. 3 that beyond the rofit maximizing rice of firm (the foreign firm), imosition of an AD duty would cause the exort rice to increase. This will lead to a decline in rofits of foreign firm. Figure 3: Profit Curve of the Foreign Firm Case II: Exorts Below Normal Value with Affirmative AD Case We consider an ad-valorem anti-duming duty on euilibrium rice below the normal value. Domestic firm would now accet the rice where t 0 is the rate of t AD duty. Such an AD duty will raise the exort rice to the level of normal value and ensure duming margin to be zero. 9 choose its outut level ( ). Accordingly, the domestic (imorting) firm would The rofit maximization roblem of the domestic firm now becomes: max ~ ~ ~~ ~ F Since the foreign firm would oerate in the domestic market based uon the residual demand, the rofit maximization roblem for the foreign firm would be a ( bt t ) max F b b 9 Duming margin is defined as the difference between normal value minus exort rice. 9

11 Trade euilibrium rice-outut combination with the AD duty revailing in the home market for the two firms would be: ~ a ( b ) and ( bt t ) ~ a ( b ) a ( b ) ( ~ t ) ( b ) ( bt t ) Comared to Case I, a commitment to an anti-duming olicy increases domestic outut, decreases foreign exorts, and increases the market rice. Case III: Exorts Above Normal Value with Counterfeit AD Case When exorts are made above normal value, the revailing rice in the home market would be ˆ and t. As the domestic firm does not get to know initially whether the imorts are above or below the normal value before filing an AD investigation, the AD case is likely to be a counterfeit one. Under such circumstances both the firms would bear an unnecessary cost of fighting out the AD case. In order to cature such unnecessary costs, we modify the initial cost functions to the followings: where, Fi Fi i, i, and i ( ) F and c( ) F c counterfeit AD case. 0 The rofit maximization roblem for the domestic firm is: includes the monetary and non-monetary cost of a ˆ ˆ ˆ ˆ ˆ max F As earlier, again the follower would choose an outut level where ˆ ˆ. The rofit maximization roblem of the foreign firm would be: a ˆ max ˆ ˆ ˆ F ˆ b b Then the trade euilibrium rice-outut combination for the two firms would be: 0 Examles of non-monetary cost are loss of goodwill, freuent visits to case hearings, time cost, among others. Also known as harassment effect. When exorts are made above normal value and there is no counterfeit AD case, the trade euilibrium riceoutut combinations for the foreign as well as the domestic firm will be similar to that of ˆ and ˆ, resectively. However, the rofit function for the two firms would not consider the monetary and nonmonetary cost of a counterfeit AD case. In other words, rofit of the two firms would be larger than what these resective firms earn (i.e. rofit) in Case III. 0

12 ˆ a ( )( b ) and a ( )( b ) ˆ ( ˆ ) ( b ) Comared to Case I and II, the foreign firm would not exort above normal value, as its trade euilibrium outut falls. Thus, in order to encourage exorts above normal value in the home market, the home country must have the ad-valorem duty: t. a ( )( b ) In other words, t acts as a credible threat for the foreign firm and restrict them only to exort above normal value. In the ost-trade scenario the domestic firm enjoys a larger market share when the revailing rice in the home market (i.e., exort rice) is above the normal value. Thus, we have ˆ ~. On the contrary, when exorts are made below the normal value, the domestic firm enjoys an increment in rofit with an ad valorem AD duty levied. Given the non-monetary cost associated with a counterfeit AD case, the domestic firm would not like to engage itself in such a situation. Nonetheless, when the foreign firm exorts above normal value, and there is no counterfeit AD case filed by the domestic firm, both firms(s) enjoy a higher rofit share. In order to substantiate the general validity of the model, we erform the entire exercise with different cost secifications, such as: c ) F ;, 0 and ( c ( ) F ;, 0 With these new cost functions, the results do not change ualitatively. However, the new [ ( b )] otimal ad-valorem AD duty is: t new. [ a ( b )( )] Thus, a credible threat of an AD duty can otentially alter the ricing strategy of the exorting firm from below to above normal value and create a win-win situation for both the firms. t would ensure, foreign firm s outut in Case II to be less than that in Case III.

13 . Game Theoretic Exercise We continue with the assumtion that the foreign firm has technological sueriority (and therefore, cost advantage) in roducing the like roduct. As a result, the foreign firm has a choice to sell below the normal value. 3 In retaliation, the domestic (imorting) firm can either initiate an anti-duming investigation or can ot to set its rice as that of leader. When the foreign firm exorts below the normal value, it enjoys a higher ay-off given the domestic firm does not retaliate but simly ot to set the same rice as that of exorter. However, the domestic firm would not refer to set such a rice as it would be extremely difficult to reduce its selling rice anywhere close to it. There is a good reason to believe that the foreign firm would have to comromise with a relatively lower ay-off when the domestic firm selects the strategy of initiating an AD case. Under such a circumstance, it is highly robable that an AD initiation by the domestic firm would get converted into an AD measure. Moreover, as the foreign (exorting) firm sells below the normal value it may have to sustain short term losses aart from incurring a significant cost of fighting out the AD case. On the other hand, when the foreign firm decides to sell above the normal value, the domestic firm has a higher likelihood to set its rice somewhere close to the rice charged by the foreign firm (leader). The domestic firm knows that not only there is a high cost of being engaged in an counterfeit AD case; also the domestic firm can now better comete with the rice charged by the foreign firm. 4 In other words, when the exorting firm decides to sell above the normal value, it would be wise for the imort cometing domestic firm to set its rice around the rice charged by the exorting firm over an AD initiation. Given the common knowledge and costs associated with AD investigation for both the layers, it is rational for the exorting firm to sell above the normal value. We consider a seuential form game. The foreign firm makes the choice first and then the domestic firm retaliates. As the domestic (imorting) firm does not get to realize initially whether the imorts are above or below the normal value before filing an AD investigation, 3 The underlying assumtion here is that exorts made below the normal value causes material injury to the domestic firm(s). 4 Such a high cost romts the domestic firm not to suffer from moral hazard roblem.

14 we therefore, assign a singleton information set containing two decision nodes for the domestic firm. EF EF EF IF IF IF IF IF IF a c e g b d f h where Figure 3: Extensive Form Game EF : Price below Normal Value; EF : Price above Normal Value; and IF : Initiate AD; IF : Accet Leader s Price. We imose the following restrictions on the ay-offs: Exorting firm: a < c and e < g Imorting firm: b > d and f < h When the exorts are made below the normal value and subseuently the domestic (imorting) firm initiates an AD rocedure (IF ), the foreign (exorting) firm would earn a lower rofit than what it would earn if the domestic firm accets the rice set by the foreign firm (IF ); i.e., a < c. This is because the foreign firm has high robability to face an AD measure (which may be very costly) when selling below the normal value. On the same note, an AD initiation by the domestic (imorting) firm would yield it a higher ay-off (i.e., b > d) as there is a high robability of recovering the sustained material injury through the AD measure. When the foreign (exorting) firm sells above the normal value (EF ), an AD initiation by the imorting firm would yield a lower ay-off for both the firms (i.e., e < g and f < h). Desite selling above the normal value if the domestic (imorting) firm initiates an antiduming case against its foreign counterart, it will not only involve monetary cost of 3

15 fighting out the AD case but also a significant time cost as well as the oortunity cost of losing that articular exort market till the case is resolved. Moreover, the foreign firm will be relatively better off if the domestic firm decides to oerate uon the rice set by the foreign (exorting) firm. Further one can say that given the otimal rate of AD duty, the foreign firm would like to exort above normal value. When exorts are made above the normal value, the domestic firm would refer to accet the revailing rice than to initiate an AD case. 5 Filing a counterfeit AD case would imly unnecessary cost to the domestic (imorting) firm. Also, the domestic firm can now comete with revailing exort rice. When the foreign firm decides to sell below the normal value, it earns a lesser ay-off comared to what it would earn if it sold above the normal value (i.e., a < e). This is because of the ossible retaliation by the domestic firm in terms of initiating an AD case against the foreign firm and therefore, imosing an AD duty on exorts. As an illustration we reresent the restrictions on the basis of some hyothetical values in fig. 4. EF EF EF IF IF IF IF IF IF Figure 4: The Pay-off Structure of Domestic and Foreign Firm When the exorts are made below the normal value (EF ), then it is more aying for the domestic firm to initiate AD (IF ) as the ay-off is higher. On the other hand, when exorts are made above the normal value (EF ), the domestic firm would accet the revailing rice (IF ) as the ay-off is relatively higher. Therefore, the strategy combination (EF - IF ) is the uniue Nash euilibrium. 5 We assume here that exorts above normal value do not necessarily cause any material injury to the domestic firm. 4

16 No matter what the domestic firm lays, selling above the normal value is clearly a referred strategy for the foreign firm. As a result, the exorting firm would not sell below the normal value and the imorting firm would find IF to be its dominant strategy. Thus, the credible threat of an anti-duming enforcement alters the strategy choice of the exorting firm and allows for the conduct of free and fair trade. 3. Summing U How would a domestic (imorting) firm resond to the case of duming? This aer analyzed the magnitude of sustained material injury and a ossible retaliation strategy by the firm(s) in terms of a rice-leadershi model. The level of AD duty varies with material injury and ensures exort rice to rise. A strategically chosen AD duty by the government of an imorting country over the rofit-maximizing rice of the foreign (exorting) firm will decrease its rofit. The domestic firm faces a trade-off in choosing alternative strategies (i.e., anti-duming initiations vis-à-vis a rice war). The (credible) threat of the ad-valorem anti-duming duty avoids exorts below normal value by the foreign firm. An anti-duming duty acts as an exante threat to the exorters and thus changes the euilibrium outcome. The threat of an antiduming enforcement might not be a ermanent one but it would manage to alter the original intentions of the exorting firm (i.e. redatory ricing). For instance, data on AD initiations reveal that around 6 er cent of AD initiations get converted to AD measures for the new users. It may be argued that owing to a high success rate, unfair trade ractices ertaining to selling below the normal value have been reduced and thus a decline in AD activities in recent times. Nonetheless, we still find that antiduming initiations are used much more than any other rotectionist measures in lieu of declining orthodox trade rotectionist measures (e.g., tariff barrier or a binding constraint of imort uota). 5

17 References Bekker, D. (006). The Strategic Use of Anti-duming in International Trade, South Africa Journal of Economics, Vol. 74(3), Finger, M. J. (993). Reform. In J. M. Finger (Eds.), Antiduming: How It Works And Who Gets Hurt ( ). The University of Michigan Press: Ann Arbor. Prusa, J. T. (00). On the Sread and Imact of Anti-duming, The Canadian Journal of Economics, Vol. 34(3), Reitzes, D. J. (993). Antiduming Policy, International Economic Review, Vol. 34(4), Tivig, T. and U. Walz. (000). Market Share, Cost-Based Duming, and Anti-Duming Policy, The Canadian Journal of Economics, Vol. 33(),