WPS4680


Policy ReseaRch WoRking PaPeR                   4680




        Safeguards and Antidumping in Latin
             American Trade Liberalization

                            J. Michael Finger
                             Julio J. Nogués




The World Bank
Development Research Group
Trade Team
July 2008

Policy ReseaRch WoRking PaPeR 4680


 Abstract

 The binding of tariff rates and adoption of the                                      on objective criteria. However, when Latin American
 General Agreement on Tariffs and Trade/World Trade                                   governments attempted to introduce economic sense
 Organization-sanctioned safeguards and antidumping                                   --such as base price comparisons on an economically
 mechanisms provided the basis to remove a multitude                                  sensible measure of long-run international price rather
 of instruments of protection in the Latin American                                   than the more generous constructed cost concept that
 countries discussed in this paper. At the same time,                                 is the core of WTO rules--protection-seekers used the
 they helped in maintaining centralized control over                                  rules against them. They pointed out that World Trade
 the management of pressures for protection in agencies                               Organization rules do not require the use of such criteria,
 with economy-wide accountabilities. The World Trade                                  nor do procedures in leading users (industrial countries)
 Organization's procedural requirements (for example, to                              include such criteria. In sum, the administrative content
 follow published criteria, or participation by interested                            of the rules supported liberalization; the economic
 parties) helped leaders to change the culture of decision-                           content did not.
 making from one based on relationships to one based



 This paper--a product of the Trade Team, Development Research Group--is part of a larger effort in the group to
 understand the forces behind trade policy reform in developing countries and how those forces interact with the policy
 guidance provided by international institutions. The authors acknowledge with gratitude the support provided by Axel
 van Trotsenburg, director of the World Bank office in Buenos Aires, also Bernard Hoekman, Beata S. Javorcik and
 Martin Ravallion of the Development Research Group. Financial support from the Bank-Netherlands Partnership is also
 acknowledged with gratitude. Policy Research Working Papers are also posted on the Web at http://econ.worldbank.org.
 The authors may be contacted at michael.finger@comcast.net and jnogues@fibertel.com.ar.




         The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development
         issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the
         names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those
         of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and
         its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent.


                                                         Produced by the Research Support Team

         Safeguards and Antidumping in Latin American Trade

                                       Liberalization

                             J. Michael Finger and Julio J. Nogués




The findings, interpretations and conclusions expressed in this paper are entirely those of
the authors. They do not necessarily represent the view of the World Bank, its Executive
Directors, or the countries they represent.



                                 ­­­­­­­­­­­­­­­­­­­­­­­­­­

Safeguards and Antidumping in Latin American Trade
Liberalization

J. Michael Finger and Julio J. Nogués




        This paper reports on success ­ success in the removal of trade barriers so as to integrate

Latin American economies into the international economy. More particularly, this paper is about

how several Latin American governments created and managed safeguards and antidumping

mechanisms as part of this liberalization.

        The paper summarizes the results of a project of studies that looked into how each of

seven Latin American countries--Argentina, Brazil, Chile, Colombia, Costa Rica, Mexico, and

Peru--have used these trade instruments. Each country study was conducted by analysts from

that country. Many of the analysts were high government officials during their country's

liberalization, so they have hands-on experience with the construction and the management of

these instruments.

        The study reversed the usually approach to examining application of safeguards and

antidumping. Rather than judge Latin American performance against WTO rules ­ examining if

Latin American officials followed the technicalities of measuring dumping, injury, and so forth ­

we examined the rules against Latin American performance. We accept that Latin American

leaders who led and took responsibility for liberalization were committed to it, then evaluate how

adoption of GATT/WTO sanctioned antidumping and safeguard mechanisms helped or hindered

their achievement of their objectives. We asked the authors of the country studies to examine

how each country's creation and use of these mechanisms as tools of policy management, for

example,

                                                 2

· What role policy managers saw for these instruments as part of a liberalization program

· How the use of these instruments has contributed to maintaining a dynamic toward openness

    to international trade

· Where these instruments have caused problems

· How policy managers have dealt with the problems.

        Because safeguards and antidumping are instruments that, by their nature, impose import

restrictions, identifying where policy managers have found discipline over application (for

instance, within the economic content of the rules or through the deterrent effect of the

information requirements imposed on petitioners and on the government), was of particular

interest in the analyses.

        In this summary, we provide first our reflections on each country study. (The studies

themselves are also available as Working Papers.) Next, we review the overall use of trade

defense measures and then go on to review several other cross-cutting issues. We compare the

policy-making institutions of the seven countries, focusing on how World Trade Organization

(WTO) rules for those mechanisms contributed to the liberalization effort and how those rules

might be made even more supportive of the liberalization effort.


I.REFLECTIONS ON THE COUNTRY STUDIES

Even though many elements that we highlight in these reflections are present in more than one

country's experiences, we chose the material in the reflections so that what we draw from one

study overlaps minimally with what we draw from another, thus bringing out the comparative

advantage lesson from each study. There is much more material in each country study than we

have included in our reflections; the reflections should serve as an invitation to read the studies,

not as a substitute for reading them.


Political Economy of Antidumping and Safeguards in Argentina, by Julio J.
Nogués and Elías Baracat (WPS3587)

In 1989, Argentina began a series of economic reforms that were revolutionary in speed, scope,

and depth. Major sectors of the economy that were previously dominated by public enterprises

were privatized, and foreign investment was put on an equal footing with domestic investment.

All import licensing requirements were removed. Tariff rates that had averaged 40 percent in the

                                                  3

mid-1980s fell to an average of 14 percent by the mid-1990s. On imports from within Mercosur

(Mercado Común del Sur, or Southern Cone Common Market), most tariffs fell to zero by the

mid-1990s. The adoption of GATT/WTO-sanctioned trade remedies ­ antidumping and

safeguard mechanisms ­ was a small but significant part of the reform package.

        In drafting the regulations and in creating the institutional mechanisms that would

administer them, Argentina made a serious attempt to incorporate good economics into their

structures and to establish discipline over their use. A presumption built into the mechanisms

was that adhering to WTO requirements would strengthen the government's position in resisting

domestic pressures for protection.

        That presumption turned out to be false.

        Imports increased sharply after the initial trade reforms, an increase resulting as much

from setting the exchange rate at a level that proved in time to be unsustainable as from the

removal of import restrictions. The business community subsequently applied pressures to use

the instruments of protection that remained, particularly antidumping.


Antidumping

        The government's strategy was to delay decisions on antidumping initiations while the

new regulatory and institutional frameworks were being completed. In a well-publicized speech

to the business community in April 1992, Minister of Economy Domingo Cavallo was emphatic

in announcing that the government would refrain from using antidumping measures until the

price level had stabilized. The main policy objective at the time was to tame inflation, and the

government considered import competition an important source of price discipline. Although the

government received 135 petitions between 1988 and 1994, it opened only 69 investigations.

        Complaints about the administration of the antidumping mechanism convinced the

government that it should update that mechanism. The first step was to modify Argentine law in

accordance with the Tokyo Round antidumping code. During the preparation of operational

regulations for the antidumping mechanism, the Unión Industrial Argentina submitted a draft

that more or less followed U.S. practice. The government, however, adopted a more liberal code

that included several provisions that could serve as the basis for limiting use. (Such provisions

were allowed, but not required, by the GATT code.) Among these provisions were:


                                                  4

· A lesser duty rule that allows the government to limit antidumping duties to the level

    necessary to prevent injury when that level is below the observed margin of dumping

· A method of imposing antidumping duties that makes it easy for exporters to know what

    price they must charge to avoid those duties

· A national interest clause that allows the government to deny antidumping measures, even

    when dumping and injury determinations are positive.

        The second step in reforming the antidumping mechanism was to create a technically

oriented commission to administer injury investigations--the Comisión Nacional de Comercio

Exterior (CNCE). A major piece of evidence influencing the government's assessment of the

need for a technically-oriented commission was the finding from analysis of existing

antidumping systems that the injury test provided more of an opportunity to apply discipline than

did the determination of dumping. The legal definition of dumping differs so much from its

intuitive or economic definition (pricing for export below home market price) that dumping in

the legal sense has been found in virtually every investigation in every country that had

antidumping regulations in place.

        The CNCE, created in 1994, functions under the Ministry of Economy. To ensure their

competence and their independence from the Commissioners, recruitment of the technical staff

was delegated to a private consulting company. Members of the technical staff prepare an injury

report, and then the commissioners decide if the evidence supports a positive injury

determination. The CNCE's decisions are made by a majority vote of its president and the four

commissioners. Following an affirmative determination, the CNCE assesses the need for and

recommends to the Minister of Economy the introduction of appropriate measures.

        Contrary to what the government hoped to achieve, macroeconomic circumstances in

Argentina overcame the capacity of WTO-sanctioned trade remedies and of the technical

analysis supplied by the CNCE to discipline the application of trade restrictions. The trade

liberalization policies and the overvaluation of the exchange rate led to a strong increase of

imports, from less than US$5 billion in the late 1980s to more than US$30 billion in the late

1990s. With this increase in imports came large-scale displacement of domestic production. For

the first time in several decades, the unemployment rate reached 10 percent in 1993, increasing

to 20 percent by 1995.


                                                  5

        After adopting the new antidumping regulations and creating the CNCE, Argentina

became one of the world's most intense users of antidumping measures. From 1995 to 1999,

Argentina applied 65 antidumping measures, exceeded only by the European Union (105), the

United States (104), and South Africa (75).

        Although the old antidumping rules had provided little discipline over when restrictions

could be imposed, they had likewise provided minimal restraint on the government's discretion

not to take action. Under the new rules with their deadlines and standards for content of a

petition, the government could not indefinitely postpone the initiation of investigations.

Moreover, in the distressed economic situation that existed in Argentina, many industries could

meet the standards for relief. Sound technical analysis of the facts found that many industries

were being injured by import competition and that, under the rules, were deserving of protection.

The new mechanism thus provided minimal political leverage for the government to resist

protectionist pressure. Particularly in 1999­2001, when the economy was in recession and the

peso was severely overvalued, the government was unable to set aside the technical

determinations. A high percentage of investigations ended with decisions to impose restrictions.


Safeguards

        In establishing a safeguard mechanism, the government attempted to follow the spirit as

well as the letter of the WTO Safeguards Agreement. An industry that petitions for protection

must present a plan for investment and structural reforms to demonstrate how it will overcome

its competitive disadvantage. Moreover, the CNCE is authorized to take into account the

requested actions' potential cost to import users.

        The CNCE turned down eight of the first nine safeguard petitions it received; through

2002 returned only 3 affirmative determinations. Consumer cost and the petitioner's not

providing a convincing plan for recovery were often the basis for not initiating an investigation

or for a negative determination. From the perspective of a protection seeker, the safeguard

instrument became a less-attractive alternative than the antidumping mechanism.

        The positive side of the story is that the hard-won trade liberalization, in large part, was

maintained. Since devaluation, Argentina initiated 14 antidumping investigations in 2002, 1 in

2003, and 7 in the first half of 2004.


                                                   6

Lessons

         Discipline has come, in part, from the widespread realization that isolation is not an

effective economic policy. As to Argentina's accepting the discipline of international rules, the

economic content of the rules has not been a source of discipline. In a situation of

macroeconomic balance, injury analysis separates the few from the many. In a situation of

macroeconomic imbalance, the many are judged deserving of protection.

         The most important lesson we draw from this study is that, in spite of Argentina's

macroeconomic imbalance and its currency crisis, WTO rules did prevent the proliferation of

protection instruments that has characterized previous crises in Argentina--a proliferation that,

in the past, had made discipline impossible to maintain.


Antidumping and Safeguard Mechanisms: The Brazilian Experience, 1988­
2003, by Honorio Kume and Guida Piani (WPS3562)
         By the late 1980s, the import-substitution industrialization process and recurrent

exchange rate crises had led to the accumulation of many severe import controls in Brazil: such

as lists of forbidden items, enterprise-specific limits on foreign purchases, shipment-by-shipment

authorizations for imports of steel, and information technology products. Tariff rates were high,

many surcharges were applied, benchmark customs values could be applied if officials sensed

there were irregularities in import prices. Exemptions from duties were available in some

circumstances. Application of these restrictions and exceptions was done through processes that

allowed wide discretion to government officials, with decisions often made at the sectoral level.


Reforms

         Although other Latin American countries have opened their economies rapidly, usually

under the leadership of one administration, Brazil has conducted reform under several

presidencies. In 1988 and 1989, the Sarney government was able to achieve some tariff

reduction, and in March 1990, shortly after coming to office, the Collor de Melo administration

introduced a degree of flexibility to exchange-rate management, suspended the program for

enterprise-specific limits on foreign purchases, and eliminated a number of special import

regimes. The government also announced a tariff reform that by mid-1993 brought down the

average rate on industrial products to 20 percent.


                                                   7

        As in other Latin American countries, control of inflation was a major factor behind the

decision to reduce import barriers. The Plan Real, initiated in July 1994 by the Franco

administration and continued during most of the two terms of the Cardoso administration, had

inflation as its most prominent target. (From 1964 to 1994, prices in Brazil had risen by 1
quadrillion percent, with 1 quadrillion being 1015, or 1 million multiplied by 1 billion.) Fiscal and

monetary reforms were the plan's most prominent elements, but the concern to discipline prices

brought additional support for trade liberalization. Under the Plan Real, the Brazilian tariff

average (unweighted) that was more than 50 percent ad valorem in 1987 declined to 11 percent

by 1994.

        By the end of 1994, the inflow of foreign capital attracted by the plan's privatization

program brought substantial appreciation of the real exchange rate. This inflow of foreign

capital, along with the accumulation of import liberalizations, brought a more than threefold

increase of imports, and with this increase came intense pressure for protection from several

sectors. Mercosur arrangements limited possible increases of tariff rates; even so, the average

tariff rose by 1999 to almost 15 percent. The government applied safeguard actions on toys and

textiles and, in some sectors, used administrative measures such as import deposit requirements

or strict application of phytosanitary requirements to respond to protectionist pressures.

        In 1999, the Brazilian currency, the Real, was allowed to float freely; it quickly declined

by more than 40 percent. Since then, tariff reduction has been renewed, and by the end of 2003,

the average tariff was less than 11 percent.


Administering Institutions

        Policy formation in Brazil often requires an extended process of working toward

consensus among a number of government agencies and the sectoral interests each represents.

Administration of the antidumping, safeguards, and countervailing duties instruments reflects

this approach to governance. Currently, investigations are conducted by the Department of

Commercial Defense (DECOM) within the Ministry of Development, Industry, and Foreign

Trade.

        When DECOM's technical findings (dumping, injury, a causal link) are affirmative, the

Brazilian Chamber of Foreign Trade (CAMEX) reaches a decision on trade defense measures by

majority vote. CAMEX is an overview agency governed by a Council of Ministers. This council

                                                 8

is composed of the ministers of Finance; Development, Industry and Foreign Trade; Foreign

Affairs; Agriculture; and Planning and Budget, as well as the Civil Cabinet minister. The

minister of Development, Industry and Foreign Trade presides.

        Within CAMEX, protectionist and openness interests are balanced by the inclusion both

of ministries with sectoral interests and those with economy wide interests. Those with sectoral

interests are more inclined to view trade restrictions as support for jobs and production in Brazil.

The Ministry of Finance's economy wide responsibilities and its continuing concern about trade

openness as a means to control inflation make it a major force in favor of disciplined use.

Brazil's leadership position in international negotiations also comes into play. Brazil has a strong

record in opposition to the unjustified use of antidumping and to its inclusion in regional

arrangements.

        The content of Brazil's antidumping (and other trade defense) regulations also reflects the

Brazilian mode of conducting government affairs. In 1995, when Brazil modified its trade

defense regulations to comply with the Uruguay Round agreements, it added several provisions

that were allowed, but not mandated, by the agreements, for example,

    · A national interest provision that would permit CAMEX not to impose a restriction when

        the trade defense investigation reached an affirmative determination

    · A lesser duty rule that would allow an antidumping duty to be less than the dumping

        margin.


Experience

        Antidumping is by far the most often used commercial defense measure. From 1995

through June 2004, Brazil notified the WTO of 114 initiations and 59 measures. (Half of Brazil's

antidumping cases ended without restrictive action, compared with less than 40 percent across all

WTO members.) Steel, chemicals, and plastics are the most frequently affected sectors. During

the same time period, Brazil has taken safeguard action in only two cases, against toys and

against coconuts, and has not initiated a countervailing duty investigation since 1995. As in other

Latin American countries that have created trade defense mechanisms as part of a trade

liberalization program, Brazil has administered antidumping measures with attention to

separating the effect of exposure to normal international competition from the effect of exposure

to abnormal business practices. In half of Brazilian investigations, normal value--the value

                                                  9

against which export price is compared to determine whether there is dumping--was estimated

on the basis of international prices quoted in specialized publications or specialized information

services.

        Another outcome that the country study reports is a relatively high use of the lesser duty

rule--the application of an antidumping measure that is smaller than the observed dumping

margin. Overall, Brazilian antidumping duties have averaged about 60 percent of estimated

dumping margins. The balancing of interests in CAMEX is largely a matter of the degree of

protection provided. Of the 86 cases that the authors studied that did not lead to protective

measures, only 2 cases came to that result because of a determination of other, overriding

interests.

        There is, however, a worrying finding. The Brazil study looks into one question on which

the other studies did not concentrate: Have antidumping measures supported the monopoly

power of domestic enterprises? The authors found that 41 percent of antidumping restrictions

applied to a product for which there was only one domestic producer and 80 percent protected

industries in which there were five or fewer Brazilian producers.


Lessons

        Since it was first established in the late 1980s, the Ministry of Development, Industry,

and Foreign Trade has become more powerful in the administration of the antidumping

mechanism while the Ministry of Finance has become less influential. The consensual nature of

Brazil's decision-making process, however, mitigates against that change marking a significant

shift toward the imposition of import restrictions. The Brazilian process demands consideration

of many petitions, but it conditions petitioners to accept a relatively high rate of rejection. It also

conditions the acceptance of a modified outcome, hence the high rate of application of duties that

are lower than dumping margins.


Keeping Animal Spirits Asleep: The Case of Chile, by Sebastián Sáez (WPS3587)
        For a short period in the early days of its economic reform and liberalization program,

Chile applied a number of contingent protection measures. Since 1995, however, Chile has

imposed only six antidumping measures, six countervailing measures, and seven safeguard

measures, none more than 12 months in length. As of December 2003, none remained in effect.


                                                   10

        Among WTO members who use contingent protection instruments, Chile has perhaps the

most tightly disciplined management of these instruments. That discipline stems from a

prevailing politics that sees open trade policy as part of a reduction of government control over

the economy--a transformation that enjoys wide support--and as an institutional structure that

transforms this philosophy into a workable administrative system.


Chile's Economic Reforms

        Chile's trade opening was part of a broader macroeconomic stabilization process, and

economic reforms, in turn, were part of a large-scale institutional transformation that

encompassed the country's political regime and general economic framework as well as the role

of the state in the economy and in social policy.

        Chile's reforms began in 1973, more than a decade earlier than those in other Latin

American countries. By 1978, a tariff made up of widely varying rates that averaged more than

90 percent ad valorem had been replaced by a uniform 10 percent rate (with few exceptions).

Other trade restrictions such as quotas, import prohibitions, and advance deposit requirements

were eliminated.

        Even with tariff rates reduced to 10 percent, the government had a skeptical attitude

toward contingent protection and other mechanisms that would even temporarily suspend

adjustment to the new trade regime.


Initial Experience

        By the early 1980s, however, the Chilean real exchange rate had notably appreciated, and

many industries pressed for relief from import competition. The government viewed

antidumping measures as having little economic rationale and the antidumping mechanism as

being too easily captured by protectionist interests.

        Safeguards, likewise, were seen as having questionable economic justification. Moreover,

GATT rules required that any country that imposed a safeguard action must provide

compensation--in the form of equivalent reduction of restrictions on other products--to

principle-supplier exporting countries. If not, exporting countries would have the right to

retaliate against, in this case, Chilean exports.



                                                 11

        The government did, however, consider the correction of distortions a valid reason for

intervention in the economy. (A distortion exists when the price of a product does not equal its

private cost of production in a competitive environment.) A foreign subsidy might therefore be a

valid reason for trade protection.

        In 1981, Chile created a Subsidies Commission within its central bank (BCCH) and soon

became an active user of antisubsidy measures. On the basis of a finding of the commission, the

Minister of the Economy could impose tariff surcharges up to the 35 percent rate that was

Chile's ceiling binding under the GATT. The commission could also recommend, and the

minister could apply, minimum custom values for assessing duties.

        Chile had signed the Tokyo Round subsidies code, but its commitment to a GATT ceiling

binding of 35 percent allowed the latitude for tariff rate adjustments within that ceiling without

the need for a code-complying finding of subsidy and injury. Chile took approximately 60

actions in the early 1980s. Most of these were against four countries--Argentina, Brazil, Peru,

and Spain--and they led to considerable friction.

        During the early 1980s, trade and current account deficits increased considerably, and the

fixed exchange rate regime became unsustainable. In March 1983 Chile shifted to across-the-

board measures to manage its trade situation, temporarily doubling its tariff rate and, most

important, allowing the peso to float.

        Administration of these trade policy mechanisms at the time was a compromise among

the technicalities of the GATT subsidy code, the resources available to conduct investigations,

and the economic objectives of the government. The resource demands for establishing a subsidy

in accord with the demands of the GATT code meant that formal countervailing measures were

taken infrequently. The GATT required that tariff surcharges be applied to all imports from all

sources, independent of whether their prices were distorted. Hence, when the source of

distortion-bearing imports provided a small share of Chile's imports, the remedy applied was

usually a minimum customs value.


Current Instruments and Administration

        In 1989, the National Commission on Distortions (CND) succeeded the National

Commission for Price Distortions through Imports, which had succeeded the original Subsidies


                                                12

Commission. The CND is responsible for investigations related to all aspects of safeguard,

antidumping, and countervailing measures. The commission's members include

· One from the National Economic Prosecutor's Office, who chairs

· Two from the BCCH

· One each from the Ministries of Finance, Agriculture, Foreign Relations, and Economy

· One from the National Directorate of Customs.

        The CND is served by a Technical Secretariat, which is lodged in and under the budget of

the central bank. If the commission determines that the statutory requirements for action are met,

it must adopt a resolution recommending the appropriate tariff surcharges. The resolution, along

with the background information and conclusions of the investigation, are transmitted to the

president of Chile, who makes a final decision through a decree of the Ministry of Finance. The

discipline in the system comes from its structure, the statutory limits under which it operates,

and, most important, strong public support in favor of an open economy.


Structural Sources of Discipline

        There are several structural sources of discipline over the use of trade defense measures,

including the following:

·   General interest agencies of government dominate the contingent protection process. The

    final decision is by the president. The only sectoral interest ministry with a seat on the CND

    is the Ministry of Agriculture. Agriculture is one of Chile's most trade-sensitive industries,

    and exceptional protection instruments have been designed for agricultural products.

·   The CND is chaired by a representative of the National Economic Prosecutor's Office. That

    office is independent from any other governmental organization or service, particularly from

    the authorities and courts before which it fulfills its functions. It is responsible not only to see

    that competition law is enforced but also to actively promote competition in the Chilean

    economy.

·   The Technical Secretariat of the CND is lodged in and under the budget of the BCCH, an

    institution that has little attachment to sectoral interests.

·   A floating exchange rate regime has consistently avoided overvaluation of the peso since the

    early 1980s.



                                                   13

·  A low uniform tariff (with few exceptions) implies that contingent protection measures create

   strong reactions from final producers and consumers.


Statutory Discipline

·  There is statutory discipline over the use of trade defense measures as follows: Chilean law

   allows only tariff surcharges. Neither import quotas nor price undertakings can be used.

·  The president cannot impose an action more restrictive than that recommended by the CND.

·  The maximum length for which a measure may be imposed is one year. Extension requires

   an entirely new investigation, resolution, and decision.

·  CND decisions are by majority of the votes cast. However, if the recommended surcharge

   will increase the tariff above the bound rate, then the approval of three-quarters of the

   members of the commission is required.

·  An antidumping or countervailing duty must not exceed the margin of distortion calculated

   by comparing the prices of dumped with nondumped (or subsidized with unsubsidized)

   imports. The objective is to remove the distortion to competition rather than to prevent

   displacement of domestic production.

·  Consumers are recognized as interested parties. Regulations allow full participation during

   the course of an investigation by any person who may feel affected.


Widespread Support

       The structure of the contingent protection system described here is not one that carried

over from the military regime. The laws that govern the use of contingent protection all date

from the 1990s after the country returned to democratic government. The law that introduced

safeguards is from 1999 when the Chilean Congress demanded it in exchange for supporting the

reduction of the uniform tariff to 6 percent.


Lesson

       The trade regime has become a basic part of the new institutional structure of Chile's

economy, and its stability and consistency toward openness have supported and been supported

by the transition to democratic government. This support is the strongest source of discipline.

The key lesson we draw from Chile's example is that, even in international trade, good

economics, if skillfully managed, can be good politics.

                                                 14

Application of Safeguards and Antidumping Duties in Colombia, by Mauricio
Reina and Sandra Zuluaga (WPS3615)
        During the second half of the last century, Colombia applied an industrialization strategy

based on import substitution, as did most Latin American countries. While this strategy

promoted some diversification of production, it also fostered concentrated property structures,

high prices, low product quality, and few incentives for modernization. The high cost of imports

made production based on foreign raw materials unsustainable.

        Faced with this situation, the government implemented a trade liberalization policy that

began at the end of the 1980s and was consolidated during the early 1990s. An import licensing

arrangement that based import quotas largely on the volume of domestic production was

abandoned, and the percentage of tariff positions subject to quantitative restrictions fell from 73

percent to 1 percent. The economy's nominal average tariff fell from near 100 percent to 11 percent,

but escalation maintained high rates of effective protection for some processing activities. Colombia

deepened its integration in the Andean Community, where most tariff rates have been reduced to

zero, and concluded trade agreements with several countries, including Chile, Mexico, and the

República Bolivariana de Venezuela. These trade reforms were accompanied by strong currency

devaluations, and at the present time, the real rate of exchange is deemed to be close to its

equilibrium level.

        In the agricultural sector, the Andean Community countries maintain a system of variable

levies that protect approximately 150 tariff positions for which the implicit average tariff is in the

order of 60 percent. Reina and Zuluaga, authors of the study, report that a good share of

Colombia's agricultural sector has managed to remain outside the liberalization trends that began

in 1990.

        Part of the politics of liberalization was the establishment of legal antidumping and

safeguard instruments to deal with the foreign competition previously neutralized by the high

overall level of protection. Table 1 lists instruments and reports their frequencies of use.

(Colombia has undertaken no countervailing duty investigations.)

        The normal safeguard mechanism follows the WTO Safeguards Agreement and is usually

referred to as a WTO safeguard. A petitioner for a WTO safeguard action is subject to extensive

WTO-based requirements to provide the information necessary to demonstrate injury or threat of

injury. This information must be certified by a public accountant. In addition, the applicant must
                                                   15

submit an adjustment program that lists the steps it will take to modernize and adapt to the new

competitive conditions. Antidumping regulations were revised in 1995 to incorporate the

discipline of the Uruguay Round agreements.

        Larger enterprises in Colombia have adopted use of the trade defense instruments, much

as similar enterprises have done in other countries. Of the 37 antidumping cases that have been

completed, 26 have been from the steel, chemicals, and petrochemicals industries. Throughout

the world, these industries have been frequent petitioners for antidumping actions.

        Other sectors, however, have not been satisfied with the usability of the safeguard and

antidumping instruments. The experience of the textiles and apparel sector is illustrative. Even

though tariffs on textile and apparel imports are relatively high (average 18 percent), trade

liberalization produced a substantial increase of imports, and with the increase in imports came

an increase in complaints about evasion of import duties (smuggling) and imports from Asia at

very low prices. The sector soon became a user of the trade defense mechanisms: Three

antidumping investigations and six safeguard investigations were undertaken on its behalf.

        Preliminary measures as a result of these investigations brought a request from several

Asian exporters for WTO consultations, but the matter was dropped when the measures expired

and were not renewed. In 1997, the government, concerned that protection for the textile industry

would disadvantage apparel producers, worked out an industry development plan with the

industry that included temporary tariff reductions on capital goods and raw materials, stronger

enforcement of smuggling controls, and modification of safeguard and antidumping

procedures--within WTO Safeguards Agreement parameters--to expedite investigations and

introduce flexibility into the criteria for application.

        Experiences such as these brought forward a perceived need for an instrument that

imposes less stringent administrative and substantive standards. In response, the government in

1999 established a second safeguard mechanism, called a special safeguard, or safeguard by

reason of disruption. This instrument allows an increase of the tariff rate up to the level of

Colombia's WTO ceiling binding, an action that does not violate Columbia's obligations under

the GATT and the WTO agreements. Thus, the investigation and determination do not have to

meet the requirements of the WTO safeguards agreement for the increased import duty to be

legal by WTO's standards. The special safeguard's standard--disruption--is interpreted as the


                                                   16

existence of imports in unfair conditions such as low prices or an important increase of

quantities. It does not require an adjustment plan, but this restriction can be in place for a

maximum of two years and is not extendable. The study's authors, Reina and Zuluaga, found in

their interviews that the business community considers this instrument to be user-friendly.

        The largest number of safeguard actions by Colombia has come under another form of

safeguard--the provision in the Andean Community that allows for suspension of the benefits

from the free trade area, imposing the tariff rate applicable to imports from outside the

community. On a provisional basis, actions such as these may be taken unilaterally but cannot be

retained without the approval of the Community Secretariat. The study's authors learned from

interviews that the secretariat review requires approximately four months to complete. The rules

for this safeguard impose no limit on how long a measure--if approved--may be in place.

Approval may, however, impose such a limit. Table 1 documents the popularity of the less

stringent instruments.

        Colombia has undertaken four investigations for Andean Community safeguards on

agricultural products, with restrictions in place intermittently from 1996 through 2003. These

measures became an issue within the Andean Community and led to the creation of a special

arrangement in the Ministry of Agriculture to administer contingent duties.

        While the worldwide statistics show one safeguard case for every 55 antidumping cases,

the ratio for Colombia is nearly one to one. Moreover, 13 of 20 safeguard applications were not

dependent on the Safeguards Agreement for their WTO legality.




                                                  17

Lesson

         In Colombia's case, we find that the substantive content of the WTO rules on trade

defense instruments has not been sufficiently valuable to overcome the administrative demands

that these instruments put on the government and the business community. Thus, Colombia's

experience has involved a unique element--the development and use of safeguard instruments

that impose lesser administrative demands. As to the use of these instruments to restrict trade, the

number of restrictions is in line with the numbers observed for other countries in the study.

Moreover, the instruments have limits on the level of protection that can be applied. A special

safeguard cannot increase the tariff above Colombia's WTO bound rate, and an Andean

Community safeguard cannot increase the level of protection above the rate applied to imports

from outside the community.


Antidumping Policies and Safeguard Measures in the Context of Costa Rica's
Economic Liberalization, by Ricardo Monge-González and Francisco Monge-
Ariño (WPS3591)
         The Costa Rican economic crisis in the early 1980s sparked an extensive review of the

country's situation. From this review came a decision to move away from the import-substitution

policies then in place and, instead, to exploit the economy's comparative advantages in the

international economy. Initially, the strategy was to provide higher fiscal incentives, intended to

reduce the antiexport bias in its trade restrictions. Soon after this decision, the government

introduced an important reduction of import barriers along with other structural reforms that

increased the openness of the economy to foreign trade and capital flows. The government

devalued the colón and adopted a flexible exchange-rate regime. Since the mid-1980s, the real

exchange rate has been stable. The policy changes attracted considerable foreign investment.

         Reforms in Costa Rica were complemented by the United States' Caribbean Basin

Initiative of 1984. Under this initiative, a significant share of Costa Rican exports enjoy zero

tariffs in the United States.

         Costa Rica's average most-favored nation (MFN) tariff by 2001 had fallen below 7

percent, though rates are considerably higher for agricultural products than for industrial

products. As part of its market access commitments under the WTO Agreement on Agriculture,

Costa Rica established tariff quotas for various agricultural products (for example, certain dairy

and poultry products), and in almost all cases, the fill levels have been low. Moreover, Costa
                                                   18

Rica has introduced several special safeguard measures based on the WTO Agreement on

Agriculture.

        Costa Rican exports, approximately US$1 billion in 1984, had expanded to US$6 billion

by 2003. This export growth was accompanied by an important diversification away from coffee,

bananas, sugar, and beef. Other products accounted for 39 percent of total exports in 1982, but

grew to 87 percent of total exports by 2003. Meanwhile, the services sector, particularly tourism,

became a significant earner of foreign exchange. At the macroeconomic level, increased trade

contributed to overall gross domestic product (GDP) growth; with growth came reduced

unemployment and a substantial reduction in the share of the population below the poverty line.


Regulations and Organization

        Costa Rican regulations on unfair trade practices have been reviewed and adapted to the

agreements that were signed within the framework of the Uruguay Round. Costa Rica has also

adopted the Regulations on Unfair Business Practices of the Central American Common Market,

or Mercado Común Centroamericano, which provide for antidumping action by one member

against exports from another member and for common action against exports from an outside

country. Bilateral arrangements with Canada, Chile, the Dominican Republic, and Mexico

recognize the rights of the parties to take antidumping action under the GATT and WTO

agreements.

        In 1995, to administer safeguard and unfair trade instruments, the government created an

Office of Unfair Trade Practices and Safeguard Measures, reporting to the Ministry of Economy,

Industry, and Trade (MEIC). Because of relatively high operating costs in relation to the number

of petitions, the government closed this office in 2000 and assigned its responsibilities to the

Legal Office of the MEIC. Technical investigations are done by the Economics Department of

the MEIC. Unlike the other countries discussed here, Costa Rica has no specialized agency to

administer antidumping and safeguard investigations.


Experience

        Costa Rica appears to be a country where import-substitution industries, particularly in

the manufacturing sector, have accepted the challenges of adopting an open economy model of



                                                  19

development. Costa Rica's experience comprises eleven contingent protection cases: six

antidumping petitions and five safeguard petitions.

        Of the six antidumping complaints, two were dropped because the complainant did not

submit the information required to support an investigation. No dumping or injury was found in

two cases, and in one case, injury was found but attributed to causes other than imports. In the

one case that reached an affirmative final determination, by the time the investigation was

completed imports had dropped virtually to zero. An antidumping duty of 0 percent was

imposed.

        The outcome of one of these cases demonstrates the government's concern to promote

competition. When a petitioner requested that an investigation be terminated because it had

reached an agreement with the exporter, MEIC continued the investigation under its own

initiative. MEIC in time concluded that no import restriction was warranted, and then notified the

Commission to Promote Competition so that it could investigate the arrangement between the

parties involved. In another instance, the investigation revealed that the problem was under-

invoicing and smuggling; the problem was handled through the appropriate legal institutions.

        Of the five applications for safeguard action, only one has resulted in a restrictive

measure being imposed. The National Cabuya Board applied in September 1995 for an

investigation against imports of coarse fiber bags, used primarily to package coffee for export.

The subsequent investigation did verify a substantial increase of imports of bags. It found also a

substantial increase of coffee exports, but not one sufficient to explain the increase of imports of

bags without a substantial displacement of Costa Rican production. The MEIC imposed a

provisional measure, a duty increase to 140 percent, but asked the producers of bags for a

commitment to improve and expand their capacity. When after six months domestic producers

had not initiated that action, the provisional measure was withdrawn.

        The few requests for protection that the government has been called on to manage are an

obvious indication that the business community is an active partner in the new economic

strategy. Though many Costa Rican enterprises began under the umbrella of the previous import-

substitution model, many have succeeded in transforming into exporters, including two

enterprises that applied for antidumping protection a decade ago. Another indication is that,

although the MEIC has issued a decree to establish a new office to specialize in unfair trade and


                                                 20

safeguard matters, there has not been sufficient pressure from the business community for it to be

created. Administration of matters such as these remains with the MEIC's Legal Office and

Economics Department.


Lessons

        In Costa Rica as in Chile, the major element in the explanation of why antidumping and

safeguard action has been minimal is that the business community is an active partner in the new

economic strategy. (The fact that sensitive agricultural products remain highly protected should

not be overlooked.) In this environment, the government is able to provide a strictly disciplined

industrial policy. An industry actively seeking to achieve international levels of productivity and

product quality will be supported whereas an industry seeking protection from international

competition will not. Also important is the record of good macroeconomic management by the

government. The real exchange rate has been stable, and the rate of inflation has been low. There

has been no challenge in Costa Rica to take on macroeconomic problems with microeconomic

tools.



Antidumping and Safeguard Measures in the Political Economy of
Liberalization: The Mexican Case, by Luz Elena Reyes de la Torre and Jorge G.
González (WPS3684)
        Creating instruments of commercial defense was part of the bargain that the Mexican

government struck with industry to win its support for opening the economy to international

competition. At the same time, the government recognized that care must be taken to ensure that

the commercial defense "monster" remained chained to the service of the trade liberalization

program and did not develop a momentum of its own.

        There was some risk that it would. Antidumping, at least in rhetoric, was intended to sort

fair from unfair international competition. The government was aware that the instrument was

tainted by its protectionist use in other countries, hence its reluctance to unquestioningly follow

procedures as they had evolved in traditional users.

        The Mexican case explains how the government developed operational techniques to sort

unwanted competition from the competition that would serve their policy objectives: benefit

consumers and stimulate the efficiency of the economy. The WTO rules' tolerance for import

                                                  21

restrictions is more generous than is the concept that the Mexican government concluded would

serve the national economic interest.

        The Mexican case demonstrates the need for the technical capacity to operate commercial

defenses correctly. It also demonstrates the need for skilled political management to take an

emergency action that would save political support for trade liberalization without allowing that

action to shift the central tendency of the system--in other words, to ensure that an exception

remain an exception.


International Competition--Normal versus Abnormal

        Two examples illustrate how the Mexican government has made operational the concept

of separating normal international competition from unfair or distorted competition. In the first

example, diiodohydroxyquinoline is an ingredient in a medicine important in Mexico. It is

produced only in Germany, India, and Mexico, and the Mexican producer who brought an

antidumping case against the Indian producer is an affiliate of a German producer. The Indian

market was at the time highly protected. Therefore, the home market price was considerably

above the normal international price--the price that the Mexican government considered

appropriate to distinguish normal international competition from dumping. The Mexican

government's decision was to impose a duty of US$4.09 per kilo rather than the full dumping

margin of US$15.56 per kilo. In its investigation, the government concluded that a duty equal to

the full dumping margin would drive the Indian supplier out of the Mexican market and

eliminate competition there. In such situations, Mexico applies the lesser duty rule.

        In the second example, after conducting an investigation of imports of bond paper, the

government concluded that the suppliers accused of dumping were doing so to meet the price set

by more efficient suppliers. The government imposed no antidumping duty, the conclusion being

that injury suffered by Mexican producers resulted from their having to compete with normal

international competition rather than with dumping.

        In the schema in table 2, the diiodohydroxyquinoline case would fall in the second row;

the bond paper case, in the third. Only in the first situation described in the first row of the table

would the Mexican government apply an antidumping duty.




                                                  22

Emergency Measures

        In mid-1993, the Mexican government was in a critical stage of its liberalization

program. North American Free Trade Agreement (NAFTA) negotiations--ostensibly completed

in 1992--had been reopened in early 1993 by the insistence of the new U.S. administration of

President Bill Clinton that parallel agreements on labor and on the environment be included. The

Mexican government had to maintain the national consensus needed to complete negotiation of

the treaty and to gain its approval by its Congress

        By this time, Mexican trade barriers had been significantly reduced, and inflation had

brought approximately 30 percent appreciation of the real exchange rate. Imports by 1992 were

five times their level in 1987, an increase of more than 30 percent per year. Industry pressed hard

for the government to keep its promise to apply antidumping measures to provide relief in this

emergency situation. Were the government not to respond, it would risk loss of credibility for the

system of contingent protection and, with it, the ongoing trade liberalization program.

        The most focused pressure came from the Mexican steel industry. It was familiar with

antidumping procedures, having been the target of U.S. and European measures several times.

The industry presented a package of petitions that covered nearly all basic products and

supplying countries. The Mexican government, however, was reluctant to act on these petitions.

Use of antidumping measures by the U.S. government to protect its steel industry had played a

major role in capture there of the instrument for protectionist purposes. Moreover, an increase of

steel prices would be a burden on user industries and exporters.

        At the same time, producers of a broad range of final goods--including clothing, toys,

shoes, and electrical appliances--came forward with intense complaints about low-priced

imports, particularly from China. These producers, however, were minimally organized and

unfamiliar with antidumping procedures.

        The government's strategy included immediate action on a range of final goods from

China. As soon as the government self-initiated investigations on more than 3,000 tariff lines, it

installed preliminary duties that ranged from 129 percent to more than 1,000 percent. (Because

part of the discipline the government had built into its antidumping rules included strict standards

on the information a petitioner must supply, it would have been time-consuming, perhaps

impossible, for these producers to pull together a package of petitions.) Product coverage was

                                                 23

carefully selected to exclude duties on any products used as inputs by Mexican producers.

Moreover, China was not at the time a GATT contracting party; hence, Mexico's duties would

violate its international commitments.

        In contrast with their immediate action on the package of final goods, the government

delayed action on petitions from the steel industry. Although most investigations were opened in

1993, the measures that were applied came near the end of 1995. In the interval, the government

examined carefully where Mexican production could be expanded with minimal increase of cost

to users. It also created a system of certificates of final use to exempt critical inputs from duties

but, at the same time, ensure that these products did not substitute for products that Mexican

producers could supply. The duties imposed were much lower than those imposed on final

goods, mostly in the range of 30 to 50 percent ad valorem.

        Mexico's 83 antidumping initiations in 1993 were more than the number for any other

country. However, after devaluation of the peso in December 1994 relieved pressure for use of

commercial defense measures, Mexico averaged 8 initiations per year from 1995 through June

2004, as compared with 40 by India, 37 by the United States, 30 by the European Union. The

emergency action was indeed an exception.


Lessons

        Creating and using contingent protection instruments has been a fundamental part of

establishing the new open economy of Mexico. The political promise that the government made

to industry at the beginning of Mexican liberalization was to provide relief from unfair

competition while otherwise expecting Mexican industry to compete with the best in the world.

The administrative unit and procedures established in Mexico show how the good sense of this

promise can be translated into sound and impartial professional action. Mexican experience

demonstrates that good policy management requires more than technical skill. The political

acumen to judge when to emphasize the technical dimensions and when not to, was critical.


Antidumping Mechanisms and Safeguards in Peru, by Richard Webb, Josefina
Camminati, and Raúl León Thorne (WPS3658)
        Peru began its recent economic reforms in the face of a particularly severe economic and

political crisis. Inflation raged in triple digits, and domestic production had fallen drastically. A

collapse of public revenues brought severe reduction of public services, including courts,

                                                     24

schools, and police. Increasing terrorism from Sendero Luminoso and Túpac Amaru groups

added to the decline of public order. The multiple exchange rate system, along with high and

distorted tariff rates, created effective rates of protection of more than 250 percent for some

sectors and negative rates for others. Import prohibitions applied to 540 tariff lines that covered

almost one-fourth of domestic production; another 534 tariff lines were subject to quantitative

restrictions.

        The trade reforms that began after Alberto Fujimori became president differed from

previous reforms in several ways. In the election campaign of 1990, the liberal message--

principally from unsuccessful candidate Mario Vargas Llosa--was no longer dismissed as a front

for business interests. Liberalization was widely accepted as a way to advance the public interest,

not simply as a desperate means to take on an economic crisis. (Fujimori had not campaigned on

a liberal platform, but his practical perception of the emergency situation led him toward liberal

policies.)

        The government wanted to do more than just remove existing restrictions. It wanted to

institutionalize trade policy decisions as well as install technical criteria and a unified evaluation

system for protection requests. Previously, decisions on tariff rates, levels of quantitative

restrictions, and import bans as well as decisions about which enterprises would receive tariff

exemptions had been made by sector officials who operated with almost complete discretion.

The lack of structure generated business uncertainty and permitted a high level of political

interference.


Institutional Reform and Indecopi

        The centerpiece of institutional reform was the Indecopi (Instituto Nacional de Defensa

de la Competencia y de la Protección de la Propiedad Intelectual, or National Institute for the

Defense of Competition and the Protection of Intellectual Property). Indecopi was created in

1993 to administer the new economic rules and actively defend the market system against the

lobby-based political erosions that had undermined previous reforms.

        Indecopi is headed by a president and an Executive Board who are appointed by the

president of Peru. The work of Indecopi is done by seven commissions, independent of one

another, whose members are appointed by the Executive Board. The establishing regulation


                                                    25

requires that each member be an experienced technical expert in the field that his or her

commission will regulate. Each commission is served by a Technical Secretariat.


Role of the Indecopi Commissions

        The Indecopi commissions are the major public organs for maintaining a competitive

market economy in Peru. The Commission on Technical Standards, for example, is responsible

not only for maintaining systems of standards but also for ensuring that businesses do not use

them to retard competition.

        The Commission for Control of Dumping and Subsidies (CDS; La Comisión de

Fiscalización de Dumping y Subsidios), created in 1991, was made a part of Indecopi when it

was established. When the CDS was created it incorporated the requirements of the relevant

GATT codes; it has since incorporated the Uruguay Round agreements into its criteria and

procedures. The CDS is the investigating and deciding agency on matters of antidumping and

countervailing duties. On safeguards, however, the CDS conducts the investigation, but the

decision rests with the Multisector Commission, which is made up of representatives of the

Ministries of Economy and Finance, Industry, and Production for the sector on behalf of which

the investigation was conducted.

        The antidumping-safeguard procedural difference reflects a difference in the economics

that lies between the two instruments. Dumping and subsidies are considered to be practices that

create distortions to be evaluated against technical criteria. Under the economic philosophy of

the system, a distortion merits correction; it creates an inequality between private versus social

costs and benefits. A safeguard action responds, however, to what is considered a political

situation. Technical analysis is needed to gauge the extent of injury and to assure that it results

from import competition. Nevertheless, the decision must balance the importance of the various

effects that will follow from the proposed action on consumers, production, and employment, not

only on the sector that petitioned for protection but also on other sectors and on other dimensions

of the national interest such as the viability of the politics of openness. An industry that petitions

for safeguard relief not only must document an increase of imports and consequent injury but

also must submit (a) an adjustment plan and (b) an economic report that quantifies the impact of

the requested measure on final and intermediate consumers and on the public interest.



                                                   26

Outcome

        The CDS and the other Indecopi commissions have provided an example of how public

decisions can be made in a transparent manner, in accord with objective, published criteria. They

have won the confidence of the public as different from the traditional bureaucracy, both in the

quality of their work and in the public's perception of whom they serve. The public media played

an important role, publicizing commission decisions that favored consumers and commission

efforts to facilitate the operations of small businesses.

        As to the CDS in particular, in its early years, it established a pattern of disciplined

procedure. As noted in table 3, from 1993 to 2000, the CDS imposed an average of only 1.4

antidumping measures per year. Moreover, standards for the content of a petition kept the

number of investigations relatively low.

        With the change in 2001 to the government of Alejandro Toledo (and already during the

temporary government of Valentín Paniagua in 2000), the industrial lobby strove to identify the

removal of import restrictions as being among the excesses of the Fujimori government and

discredit the hard-line technical and independent approach to determining trade policy that the

Indecopi and the CDS represent.

        A safeguard petition on behalf of the textiles and clothing sector became an important

vehicle for the pressure. The protection seekers opted for a safeguard petition because it offered

greater possibility to minimize the technical input and bring trade union and other political

pressures to bear. In the debate over whether to protect textiles and clothing producers, the major

argument against restriction has been the damage to the Peruvian economy that retaliation from

China would bring. This argument is being advanced by export industries, their trade unions, and

the Embassy of China in Peru. As this country study was completed in August 2004, the option

toward which the government was inclined was to shift from transitional safeguards against

China (allowed under China's terms of accession to the WTO) to generalized safeguards.

        The change of government also initiated a struggle to replace Indecopi's commissioners.

The new commissioners have closer ties to industry, and they have focused on enterprise

insolvency resulting from the 1997 banking crisis and have redirected resources toward issues of

enterprise restructuring. Matters of market structure and free competition have received less

attention.


                                                  27

        As to decisions on trade issues, there has been an increase in the share and in the number

of cases with restrictive outcomes. The numbers, however, are still low. Authors Webb,

Camminati, and Thorne report that, although decisions are now more influenced by the industrial

sector, the technical quality of investigations has not been lost.


Lessons

        The positive lesson we draw from Peruvian experience is the example it provides of a

unified, independent, and technically oriented decision structure. Peru's experience demonstrates

that such an institution is not impervious to political pressure; however, the institution has

maintained its technical integrity. Equally important, the idea of a unified decision structure

rather than a sector specific approach has prevailed.

        Not all the political considerations that have entered into trade policy determinations have

pushed in the protectionist direction. In the debate over whether to protect textiles and clothing

producers, the major argument against restriction is the damage to the Peruvian economy that

retaliation from exporters would bring. This argument is being advanced by export industries,

their trade unions, and the embassies of exporting countries.


II. SCOPEAND BACKGROUND OFTHE REFORMS

        Each of the reform experiences began in periods of macroeconomic crisis; hence, a major

part of the dynamic behind them was to control inflation and restore macroeconomic balance.

Microeconomic reforms included much more than removal of trade restrictions. Major sectors of

the economy previously dominated by public enterprises were privatized; foreign investment

rules were liberalized. In many countries, commercial legal codes were extensively revised, with

new emphasis on creating and maintaining competition in domestic markets. With respect to

trade policy, although the application of extensive import restrictions during the second half of

the last century had promoted some diversification of production, it also had fostered

concentrated property structures, high prices, low product quality, and few incentives for

modernization.

        Trade liberalization was novel in speed, depth, and scope. Import licensing requirements

and quantitative restrictions were almost entirely removed. Applied tariff rates that had averaged

50 percent or more in the 1970s and 1980s have fallen by three-fourths and, in Chile and Costa

                                                  28

Rica, to well below 10 percent. Regional arrangements such as Mercosur, the Andean

Community, the Central American Common Market, and NAFTA have been mechanisms for

additional liberalization. (For example, two-thirds of Mexico's imports are from the United

States and are thereby duty-free under NAFTA.) Separate protection regimes were maintained

for agriculture in several countries, but import-sensitive manufactured goods were not excluded

from the liberalization. There was no Multifiber Arrangement, no special arrangement for

textiles and clothing among these countries.

         In their economic histories, the seven countries had passed through several cycles of

protection and openness. Although it is common to describe the periods of protection as

application of an import-substitution strategy, swings toward protection are more accurately

described as accumulations of political victories by protection- and subsidy-seeking interests ­
not as a coherent applications of any particular economic strategy1. Many different import

control instruments were in use: tariffs, surcharges, benchmark customs values, enterprise-

specific limits on foreign purchases, and import prohibitions. Application of restrictions was

done through processes that allowed wide discretion to government officials, with decisions

often made at the sectoral level. Safeguards and antidumping--as GATT and WTO rules define

them--were rarely used. With almost all tariff lines unbound through the GATT, WTO, or other

international agreements, the international sanction that use of such instruments would have

provided was not needed.

         The reforms of the 1980s and 1990s took place as these countries increasingly saw the

advantages of active participation in the GATT and WTO systems. At the Uruguay Round, they

accepted ceilings on almost all tariff lines; hence, for any duty increase beyond these bindings,

they would need the clearance that use of a GATT- or WTO-sanctioned instrument would

provide. Moreover, there was widespread realization that, to prevent the eventual erosion of this

round of liberalization, new institutions for managing trade policy would be needed--institutions


1. Thirty years ago, Bela Balassa began the preface to his pathbreaking study, The Structure of Protection in
Developing Countries (1971), by noting that an import-substitution policy was the rationale for the trade policies of
many developing countries. He went on to point out that
whatever the intrinsic merits of this policy, its application has rarely been based on a consistent program of action.
Rather, the existing system of protection in many developing countries can be described as the historical result of
actions taken at different times and for different reasons. These actions have been in response to the particular
circumstances of the situation, and have often been conditioned by the demands of special interest groups. The
authorities have generally assumed a permissive attitude toward requests for protection and failed to inquire into the
impact of the measures applied on other industries and on the allocation of resources in the national economy. (xv)
                                                            29

that would help to control the emergence of many instruments and unify the application of the

instruments into a policy process that would bring forward the economic logic of how openness

would support economic development. The two sides of the issue reinforced each other: while

the need for reform suggested the value of active WTO membership, active WTO membership

required the use of approved instruments.

        Thus, these liberalizations differed from past ones in more than their scope and depth. An

important element was to institutionalize the changes that were made through the acceptance of

GATT and WTO bindings and the adoption of WTO-sanctioned trade defense or contingent

protection instruments.

        Creating trade defense mechanisms was often part of the bargain to gain industry

acceptance of liberalization. Industries find trade defense mechanisms attractive because a

safeguard mechanism can slow the pace of liberalization where its impact turns out to be

particularly severe. In principle, an antidumping mechanism will allow domestic industries to be

exposed to normal international competition but, at the same time, be protected from unfair

competition.

        From the perspective of maintaining the momentum of liberalization, these mechanisms,

once in place, should serve as a means to accommodate and isolate pressures that might

otherwise grow into large-scale threats; they should allow the possibility of one step back to

preserve two steps forward. In each country, trade liberalization did generate substantial

increases of imports, hence the need for mechanisms to manage these pressures. Thus,

maintaining an economically sensible trade policy was, in significant part, a matter of managing

pressures for exceptions--for protection for a particular industry--so that the exceptions

remained exceptions and the decision process reinforced, rather than undermined, the politics of

liberalization.

        An additional virtue of these mechanisms is that they would contribute to a shift of the

culture of policy management that reform leaders wanted to advance: a shift to a culture of

policy management based on facts of economic potential--unified, objective, and transparent

mechanisms--from one based on relationships. Finally, trade regulations would form a part of

the new regulatory structure intended to create a more competitive environment than existed

before.


                                                 30

III. TRADE DEFENSE USE--THE OVERALLRECORD

         Trade defense mechanisms are instruments for imposing import restrictions rather than

for removing them. Their creation raises the possibility of overuse: Have they been used so much

as to substantially offset the liberalization that they were intended to promote?

         With regard to applied tariff rates, table 4 presents the tariffs on industrial goods for

seven countries: Argentina, Brazil, Chile, Colombia, Costa Rica, Mexico and Peru.

         The record of use of antidumping, countervailing duties, and safeguard investigations and

measures since 1995 (as notified to the WTO) are presented in tables 5 through 8. Tables 5 and 6

list the numbers of antidumping investigations and measures by the seven countries, as notified

to the WTO. Countervailing duties have been minimally used; since 1996, only nine measures in

total by the seven countries in the study (table 7). Safeguard actions have also been infrequent.

Over the same period, Chile has notified the WTO of seven safeguard measures, three

preliminary and four definitive measures. All have been tariff measures at rates below 15 percent

and in place for no longer than 12 months. No other country in the group has applied more than

three safeguard measures.

         Tables 5 and 6 list the numbers of antidumping investigations and measures notified by

the seven countries. For the countries in this study, as for other WTO members, antidumping has

been by far the most frequently used mechanism for managing pressures for exceptional
protection,2 Colombia's use of within-bindings safeguards being the only exception. Several of

the countries studied were, for short periods, major users of trade defense measures. Chile, in the

early 1980s, initiated more countervailing duty cases than any other GATT member; Mexico, in

1993, initiated more antidumping investigations than any other GATT contracting party. The

country studies show, however, that these cases were truly emergency actions, coming at critical

times in the trade liberalization process. Through the use of trade defense measures and other

actions, governments were able to complete critical steps, and the rate of use dropped radically

afterward.

         Although trade defense measures are, in theory, intended to deal with other matters, these

country studies show that an overvalued exchange rate often creates the conditions in which a


2. Together, WTO members have initiated more than 2,500 antidumping investigations since 1995, compared with
174 countervailing duty investigations and 142 safeguard investigations.
                                                       31

safeguard or antidumping investigation will return an affirmative determination. The Argentine
study provides an explicit demonstration3. Exchange rate overvaluation was part of the situations

in Chile and Mexico mentioned in the above reflections, and the Brazil study also describes an

incident in which overvaluation brought intense pressure for import protection. The record

shows, however, that when the exchange rate was appropriately adjusted, use of trade defenses

quickly receded.


IV. RATIONALE FORANTIDUMPING MECHANISMS

         The rationale for antidumping enforcement in Latin American countries tends to focus on

the economic concept of distortion. A distortion exists when the price of a product does not equal

its cost of production in a competitive environment. In modern economics, as opposed to the

mercantilist economics from which the concept of dumping is drawn, the national economic

interest, or economic welfare, would be higher if the distortion did not exist. A distortion thus

justifies government intervention in the market.

         When they established their procedures, Latin American governments were aware from

analysis of antidumping systems in traditional users such as the United States and the European

Union that dumping in its legal sense--even more than in its theoretical sense--did not identify

circumstances in which government intervention would advance the national economic interest.

Antidumping enforcement would use neither the exporter's home market price nor the
constructed cost4 as the measure of what price should be. Instead, the price that enforcement

attempted to reach would be the prevailing international price in competitive markets. Such

interpretation followed the promise that governments had made to industry when they created the

antidumping mechanisms; industry would be expected to adjust to normal international

competition but would be protected from extraordinary situations.

         The latitude provided by WTO rules allowed governments to accommodate at two levels

the mismatch between the economic theory on which WTO rules were based and the concept

that justified government intervention. At the level of antidumping investigation, WTO rules



3. Econometric studies based on information from other countries confirm that the business cycle influences the use
of antidumping and other trade defenses. For example, see Leidy (1997) and Takacs (1981).
4. The WTO rules sanction constructed cost formulas that allow generous overstatement of what normal business
accounting practices would determine.
                                                       32

allow, in certain circumstances, the use of third-market price as the measure of normal value. An

internationally competitive environment could be chosen as this third market. The more general

basis for latitude is that the WTO rules do not demand that an antidumping duty be equal to the

measured dumping margin. The duty cannot be larger, but it can be smaller, even zero. Latin

American governments adopted the lesser duty rule, which, in theory, provides that the

antidumping duty be set no higher than the level necessary to eliminate injury to domestic

producers.

        Although the lesser duty rule is sometimes described as being allowed by the WTO

antidumping agreement, given that WTO members did not give up their sovereign right to set a

lower duty by accepting the agreement, there is no need for a WTO rule to create that right.

Within the rules, a WTO member can set an antidumping duty at any level determined with

reference to any criterion the government considers appropriate--limited by the measured

dumping margin. Because the rules for measuring the dumping margin are generous, once an

affirmative determination is returned, the latitude is considerable.

        Similarly, these countries included in their antidumping processes a national interest

clause that allows the deciding agency to take no action or to modify action if the national

interest would thereby be better served. Again, this action is a right not given up by accepting to

abide by the WTO agreement; it is not a privilege granted by the agreement. The country studies

explain additional criteria and limits governments included in their mechanisms.


V. RATIONALE FOR SAFEGUARD MECHANISMS

        Though they have been infrequently used, safeguard regulations are in place in each of

the seven countries. The Peru study provides a sharp example of the difference in philosophy

behind safeguards and antidumping. Antidumping is a technical matter; a distortion compromises

economic welfare (the national economic interest) and therefore should be corrected. A

safeguard action responds to a political situation. Technical analysis is needed to gauge the

extent of injury and assure that it results from import competition. The decision should, however,

balance the importance of the various effects that will follow from the proposed action on

consumers, production, and employment, not only in the sector that petitioned for protection but

also in other sectors. The decision should also take into account other matters such as how action

or inaction would affect the viability of the liberalization process and the possibility of retaliation
                                                  33

by exporting countries. In Peru, an industry that petitions for safeguard relief not only must

document an increase of imports and consequent injury but also must submit an adjustment plan

and an economic report that quantifies the effect of the requested measure on final and

intermediate consumers and on the public interest.

        In practice, the most frequent users of safeguards have been Chile and Colombia. In

Chile, increased import duties have been used on isolated occasions to provide short-term

breathing space for domestic enterprises beset by competition from imports. Colombia's use of

safeguards differs in that the most frequently used safeguard instrument avoids the procedural

and substantive requirements of the WTO agreement by restraining restrictive action within the

ceiling bindings the government has accepted at the WTO.

        The Argentine experience is also illustrative. The trade commission there turned down

eight of the first nine safeguard requests it received; the response by protection seekers was to

apply for relief under antidumping regulations. In theory, antidumping applies only when an

economic distortion can be identified, but in practice, its standards overlap considerably with the

injury standard of safeguards. Moreover, when qualifying for antidumping relief, an industry

need not meet the safeguard requirements to provide a workable recovery plan nor subject itself

to review of how its protection will affect consumers or production in other sectors. Though

different in theory, antidumping and safeguards have proven in practice to be quite fungible. In

practice, the use of one versus the other is a matter of administrative or political convenience, not
of the economics of the underlying situation5.


VI. DECISION-MAKING BODIES

        Although each country's contingent protection processes have at their base a capacity to

conduct investigations and reach determinations in accordance with the WTO rules,

arrangements for deciding the restriction that will be imposed vary considerably. In antidumping

cases, Argentina and Peru, for example, have created primarily technical mechanisms. The

technical factors along with the independence and the analytical skills of the commission

members are paramount. In contrast, Brazil entrusts the decision--once a technical



5. Finger (2002) provides a general review of the point, drawing evidence from the evolution of use of GATT­WTO
pressure-valve instruments since the GATT was first adopted.
                                                         34

determination is reached--to a vote by an overview board whose members come from a number

of ministries that represent a diversity of interests. Colombia likewise entrusts the decision to an

interministerial body, presided over by the country's president.

         Chile has a system that controls application of contingent protection measures at several

levels. At the administrative level, general interest government agencies dominate the

commission that reviews technical determinations and prepares a recommendation for policy

action. The commission is chaired by an official from the agency of government responsible for

enforcing competition policy, and the final decision is made by the country's president. At the

statutory level, only tariff surcharges may be applied, for no more than one year. A restriction

that will increase the tariff above the bound rate requires approval of three fourths of the

members of the commission. Moreover, consumers are recognized as interested parties.

         No factor is more important to the disciplined use of contingent protection than the

widespread support for the economic philosophy of openness. This point is made in each country

study. The Chilean and Costa Rican studies particularly bring out the importance of deep

constituent support for the economic philosophy of openness and constituent appreciation of the

value of a competitive economy.

         Costa Rica has no specialized agency to conduct contingent protection investigations.

They are part of the normal workload of the legal and economics staffs of the trade ministry. The

positive side of this arrangement is that a staff with more general responsibilities might remain

more sensitive to general economic interests than a specialized staff. Because the technical work

in an antidumping or safeguard investigation amounts in substance to document foreign

unfairness and displacement of domestic production by imports, professionalism can take on a
pro-protection slant6.

         It should be noted that Chile and Costa Rica have maintained relatively high protection

for agriculture, and that they have a good record for macroeconomic stability, low inflation, and

stable exchange rates. There has been no challenge to take on macroeconomic problems with

trade instruments.




6. For example, Boltuck and Litan (1991) document the biases in the administration of U.S. regulations.
                                                      35

VII. SOURCES OF DISCIPLINE

        After reviewing the country studies, we can look again at a question posed at the

beginning of the paper. Because safeguard and antidumping are instruments that, by their nature,

impose import restrictions, where did policy managers find discipline over application? What

helped promote acceptance of negative decisions?

        The administrative content of WTO rules has made an invaluable contribution. A major

source of erosion of previous liberalizations had been the tendency for instruments to proliferate,

often controlled by different ministries and not unified by process or by criteria. Ministries most

closely associated with industrial sectors were not counterbalanced by ministries with

overreaching responsibilities or by consumer or import user interests. (In politics, concentrated

interests have the advantage over more­dispersed interests, with that advantage being multiplied

when decision making is at the sectoral level.) The WTO rules not only supported unification of

process and criteria but also provided a template for objectivity; public notice of criteria,

participation and access to information by interested parties, and publication of the legal and

factual basis for decisions. They provided guidelines for objectivity, transparency, and

accountability in public decision making, contributing not only to better trade policy but also to

change of the policy management culture.

        An important source of discipline has been the rigorous standard for the information that

a protection-seeking industry was required to supply, both for initiation of an investigation and

as evidence to support the impact of international competition on the industry. Peru also imposes

an application fee, currently about US$925.

        Effectiveness with respect to the continuance of liberalization has been a matter not only

of the judicious application of WTO requirements but also of the skilled use of the discretion

they allow. In Mexico, the steel industry was familiar with WTO-based instruments, having

been party to actions applied by the United States and the European Union against their exports.

Other sectors found the administrative demands more formidable. When the survival of the

liberalization program depended on such action, the Mexican government self-initiated

investigations that involved smaller-scale producers of toys, clothing, and other consumer

products. The combination of strict information requirements and the government's authority to

self-initiate investigations provided a valuable degree of discretion. At the same time, the

                                                 36

government used the discretion in its regulations to proceed more slowly with the investigation

of the complaints by the steel industry and work out specific mechanisms to bring user interests

into play.

        In Colombia, the steel, chemicals, and petrochemicals industries adapted quickly to the

information demands of the WTO-based system of trade defense. Sectors composed of smaller

enterprises, however, found the requirements burdensome and, in time, convinced the

government to created alternate trade defense mechanisms. Colombia's WTO bindings discipline

the use of these instruments. The adoption of a unified mechanism for alternate instruments is

another source of discipline.

        The economic content of the WTO standards is another matter. Apart from the cost of

demonstrating that the standards have been met, analysis of experience in other countries has

shown that the standards required by the WTO agreements on safeguards and antidumping are

generous in determining that petitioners are deserving of protection. Aware of this characteristic,

the Latin American countries adopted their own additional bases for not taking action (for

example, a national interest clause) or for limiting the degree of action that was taken (for

example, a lesser duty rule).

        When Latin American governments included the straightforward economic sense of

taking into account the negative impact of an import restriction on domestic users, protection-

seekers used the rules against them ­ pointed out that WTO rules do not require consideration of

such elements, that procedures in `leading' users do not include them.

        Another attempt to improve on the economics of the WTO rules was to base price

comparisons on a measure of long-term international price. Mexico's experience is particularly

relevant here. Attempts to develop an operational measure of long term international price were

however overwhelmed by references by protection-seekers to the idea of `constructed cost,'

around which GATT/WTO discussion has swarmed and which at a technical level has been long
captured by protectionist interests.7 Lacking support from WTO rules or negotiations, these

attempts eventually waned.




7This capture has been the subject of a voluminous literature on the `administrative bias' in administration of
antidumping rules. Boltuck and Litan (1991) provide an earlier, Nelson Vandenbussche (2005) a later summing up
on the point.
                                                         37

        Latin American experiences confirm that the dumping and injury criteria required by

WTO rules are generous. To keep the number and the degree of restrictions in check, the
governments often had to depend on the discretionary authority they retained8. The economic

content of the rules did not create political capital that governments could use to explain a

negative response to a request for protection; governments had to use political capital otherwise

gained to explain why protection was not given to a petitioner that the rules found to be

deserving.


VIII.            RECOMMENDATIONS FOR WTO RULES

        The country studies reviewed here demonstrate that the WTO rules on safeguards and

antidumping have proven to be essential and that the courage and skill of policy managers have

proven to be indispensable in the sweeping trade liberalization that has taken place in Latin

America trade policy over the past quarter century.

        In this final section, we ask how the WTO rules could have been even more useful and

more supportive of the liberalization that these countries set out to achieve. To be effective, a

policy manager must master the technicalities, but we have found that the effective use of

safeguards and antidumping mechanisms is as much art as science. In areas of economic policy

in which the advice to "follow the rules" has perhaps been overemphasized, we will restrict

ourselves to general observations; we leave the crafting of negotiating proposals to people with

expertise in such work.

        One might identify two objectives with respect to how international rules might shape

domestic processes through which governments make decisions to impose protection. From

perhaps an ideal perspective, one might hope that they would guide a country to identify those

interventions that add more to the national economic interest than they take away. Eliminating

distortions, in theory, follows from this perspective.

        The experiences studied here show, however, that there are instances in which other

considerations make it impossible to avoid what might be an economically unsound trade

intervention. In those situations, good policy becomes a more pragmatic matter, a matter of


8. Although the discretionary authority that counties retained and the political review processes that countries
superimposed over technical determinations in a number of instances led to a reduced level of restriction, there were
few instances in which no restriction was imposed after an affirmative determination.
                                                          38

managing interventions to maintain the momentum of liberalization and strengthen the politics of

avoiding, rather than of imposing, such restrictions in the future.

         The first objective is about economically good interventions versus economically bad; the

second is--pragmatically speaking and over the long run--about less interventions rather than

more. It is also about the number and form of interventions in the short run that, over the long

run, will minimize opposition to liberalization.


The Rules Are Too Generous
         Although the administrative dimensions of the rules have been supportive, the economic

content has been less so. A key point that emerges from the country studies is that, from an

economics perspective, WTO rules on safeguards and antidumping are too generous with respect

to separating either good interventions from bad or less interventions from more. Although

antidumping, in theory, is about interventions that make economic sense, in reality, WTO

guidelines allow restrictions that amount to ordinary protection. Aware of this characteristic,

Latin American governments attempted to make operational the economic concept of distortion.

Qualitatively speaking, this concept would help them to separate trade interventions whose

economics is good from interventions whose economics is bad. Sticking to the concept of

distortion would also reduce the number of interventions that received sanction. At the same

time, these governments introduced lesser duty rules and national interest clauses--additional

bases not to restrict in instances in which the WTO rules would allow a restriction.

         The safeguard mechanisms that the governments set up also included disciplines not

mandated by WTO rules. WTO safeguard rules require only an investigation and determination

of injury from imports and of the cost that import competition imposes on competing domestic

producers. They do not require that benefits to industrial users and to consumers likewise be

taken into account. In the interest of reaching economically sensible safeguard decisions, Latin

American governments supplemented their safeguard processes with requirements for an

adjustment plan (and an economic report in the case of Peru) that quantifies the impact of the

requested restriction on final and intermediate consumers and the public interest. The logic of the

first of these is to assure that the protected industry will eventually be viable in an open economy

and will not be a long-term burden on the rest of the economy. The logic of the second is that the

imports that are an inconvenience for some domestic interests provide benefits for other

                                                   39

domestic interests--lower-priced and state-of-the-art inputs for domestic producers and final

goods for consumers.


Overcoming Conflicts Created by the Economic Content of the Rules
          The over-generosity of the economic content of the WTO rules leads to two points of

conflict between the rules and the objective of defending liberalization:

· The case for protection has the weight of international obligation behind it whereas the limits

     that governments included in their regulations do not.

· Technical analysis is expected--politically--to work in favor of liberalization, yet its

     economics is to make the case for protection, to document the costs that liberalization

     imposes on some domestic interests.

          Latin American experience suggests a straightforward way to overcome these conflicts:

mandate identification of the impact on users and consumers as well as the impact on competing

domestic producers. The policy process would ask: Who in the domestic economy would benefit

from the proposed import restriction, and who would lose? By how much?

          The technicalities would be simple: recognize domestic users and consumers as interested

parties, and require that the investigation determine the effect on them of the proposed restriction

in parallel with its determination of injury from trade to the protection seeker. The effect of the

restriction on users and consumers would be measured in the same dimensions as injury--for

example, jobs lost because of higher costs and lower profits--the standard metric of effect. The

existing rules of procedure would apply to the investigation of the costs of protection equally

with that of the benefits. All interests would be treated the same in process and in concept.

Even in those instances in which the decision is to restrict imports, the process would bring

forward the reasoning behind the liberalization, the benefits from openness. The limits that Latin

American governments imposed on their use of trade defense measures would have been more

effective against domestic pressures for protection if the reasons for minimizing use had had the
force of international obligation behind them9.



9. Another conflict, perhaps more semantic than substantive, is between the national interest and the protection-
seeking industry's interest. The protection seeker's interest is part of, not other than, the national interest. The effect
of a trade restriction on the national interest, in concept, should include the cost of protection and those who bear
this cost, along with the benefits and those who receive these benefits. A shortcoming of an injury investigation is
that it does not give equal standing to all parts of the national interest.
                                                             40

          Beyond making economic sense, such a safeguard process would be politically balanced.

It takes into account the effect of trade on all interests in the country: those that benefit from

liberalization and those who are burdened. This balance would not be a matter of the technical

versus the political. The benefits to trade for user interests would be measured in the same
business and economic terms as the costs to protection seekers10.

          A shortcoming of this suggestion is that it would increase the cost of an investigation,

would impose administrative costs on potential victims of the restriction as well as on the

potential beneficiaries. If however discipline over use were based on equal treatment of the costs

of protection and its benefits rather than on the complexity of determining the benefits, then

economies could be achieved without compromise to the viability of the information developed.

          The balanced cost-benefit process would have to be applied in antidumping

investigations as well as in safeguard investigations. The situations to which antidumping or

safeguards can be applied overlap considerably. Because antidumping's popularity is not based

on its unique identification of the existing business or economic situation, expanding the scope

of the safeguard process would have the effect only of further increasing the use of antidumping

in its place.

          Other than modifying the technical dimensions of investigations to include the gains from

trade, the alternative for reform is further manipulation of the technicalities of determining

dumping or injury. That manipulation--illustrated by the proposals submitted at the Doha

Negotiations--is not about matching the operational guidelines more closely with the economic

concepts of dumping (predatory pricing) or externality--in theory, situations in which a trade

intervention provides a net addition to the national economic interest. This struggle over

technicalities, we contend, will have little effect on the quality or the quantity of import
restrictions that are applied.11 There are sufficient technicalities that adding a few here, trimming



10. The process could also be balanced in that users or consumer groups could petition for removal of an import
restriction on the basis of the cost it imposed on them. History provides at least one example of a mechanism
balanced in this way. Section 336 of the U.S. Tariff Act of 1930 empowered the U.S. president to increase or to
decrease a tariff rate based on investigation by the U.S. Tariff Commission of the cost of producing the product in
the United States compared with that cost in exporting countries. (The U.S. tariff would be set equal to the measured
difference.) From 1931­41, the commission conducted investigations on 101 products. These investigations led to
29 tariff increases and 25 tariff reductions; in the other 47 instances, the rate was not changed.
11Borrowing a concept from mathematics, the rules on trade remedies provide an over-determined system, e.g., 15
equations to solve for 2 unknowns. Operating the system requires mastery of its technicalities, e.g., how to solve
simultaneous equations. Within the technicalities there remain however many degrees of freedom. Two equations
                                                            41

a few there, will have no effect. Thinking, even tinkering "within the box," is not likely to help.

To support liberalization, the WTO rules should support identification of the benefits from

liberalization.

         The conclusions in a paragraph: The binding of tariff rates and adoption of GATT/WTO-

sanctioned safeguards and antidumping mechanisms provided the basis to remove a multitude of

instruments of protection, and to maintain centralized control over the management of pressures

for protection in agencies with economy-wide accountabilities. WTO procedural requirements

(e.g., to follow published criteria, participation by interested parties) provided a basis for

changing the culture of decision-making from one based on relationships to one based on

objective criteria. However, when Latin American governments included the straightforward

economic sense of taking into account the negative impact of an import restriction on domestic

users, protection-seekers used the rules against them ­ pointed out that WTO rules do not require

consideration of such elements, that procedures in `leading' users do not include them.

Likewise, attempts to base price comparisons on an economically sensible measure of long run

international price brought protectionist criticism for not using the more generous `constructed

cost' concept applied in `leading' countries and around which WTO discussion swarms. Finally,

WTO rules fail to take into account the reality of resources available for policy management in

developing countries. Simpler procedures that made economic sense were found not in

compliance. In a sentence, the administrative content of the rules supported liberalization, the

economic content did not.




IX. REFERENCES

Balassa, Bela. 1971. The Structure of Protection in Developing Countries. Baltimore, MD: Johns
         Hopkins University Press.
Boltuck, Richard, and Robert Litan, eds. 1991. Down in the Dumps: Administration of the Unfair
         Trade Laws. Washington, DC : Brookings.
Finger, J. Michael. 2002. "Safeguards: Making Sense of GATT/WTO Provisions Allowing for
         Import Restrictions." In Development, Trade, and the WTO: A Handbook, ed. Bernard


are sufficient to solve for two unknowns, having 15 equations allows 105 possible pairs, each pair likely to provide a
different answer.


                                                        42

       Hoekman, Aaditya Mattoo, and Philip English, Ch. 22, pp. 195-205, Washington, DC:
       World Bank.
Kume, Honorio and Guida Piani. 2005. Antidumping and Safeguard Mechanisms: The Brazilian
       Experience, 1988­2003, World Bank Policy Research Working Paper WPS3562.
Leidy, Michael. 1997. "Macroeconomic Conditions and Pressures for Protection under
       Antidumping and Countervailing Duty Laws: Empirical Evidence from the U.S." IMF
       Staff Papers 44 (1): 132­44.
Monge-González, Ricardo and Francisco Monge-Ariño. 2005. Antidumping Policies and
       Safeguard Measures in the Context of Costa Rica's Economic Liberalization, World
       Bank Policy Research Working Paper WPS3591.
Nelson, Douglas R. and Hylke Vandenbussche. 2005. The WTO and Antidumping. (2 Vols.)
       Cheltenham: Edward Elgar.
Nogués, Julio J. and Elías Baracat. 2005. Political Economy of Antidumping and Safeguards in
       Argentina, World Bank Policy Research Working Paper WPS3587.
Sáez, Sebastián. 2005. Keeping Animal Spirits Asleep: The Case of Chile, World Bank Policy
       Research Working Paper WPS3587.
Reina, Mauricio and Sandra Zuluaga. 2005. Application of Safeguards and Antidumping Duties
       in Colombia, World Bank Policy Research Working Paper WPS3615.
Reyes de la Torre, Luz Elena and Jorge G. González. 2005. Antidumping and Safeguard
       Measures in the Political Economy of Liberalization: The Mexican Case, World Bank
       Policy Research Working Paper WPS3684.
Webb, Richard, Josefina Camminati, and Raúl León Thorne. 2005. Antidumping Mechanisms
       and Safeguards in Peru, World Bank Policy Research Working Paper WPS3658.
Takacs, Wendy. 1981. "Pressures for Protectionism: An Empirical Analysis." Economic Inquiry
       27 (4): 687­93.
WTO (World Trade Organization). 2000-4. "Report [year] of the Committee on Anti-Dumping
       Practices" [Document G/L/409 for 2000, Document G/L/494 for 2001. Document
       G/L/583 for 2002, Document G/L/651 for 2003, Document G/L/707 for 2004]. Geneva:
       WTO.




                                                43

X. TABLES

Table 1 Colombia: Safeguard and Antidumping Investigations 1990­June 2004
                                     Restriction

                                          Percentage
                  Number                   of number
Category          completed    Number      completed

Antidumping       37             23             62

Safeguard         34             25             74

  WTO             11             10             91

  Special         10               5            50

  Andean          13             10             77

Source: Tabulated from date in Table 5.2 of WPS3615




                                            44

Table 2 Mexico: Antidumping Duty Levels in Different Situations
Price comparison                      Level of duty

Normal value = International price >  Duty = Dumping margin

Export price

Normal value > International price >  Duty = Difference between export price and

Export price                          international price

Export price = International price <  No antidumping duty ­ injury (if observed) is

Normal value                          from normal international competition

Source: WPS3684




                                           45

Table 3 Outcomes of Peru's Antidumping Investigations 1993­June 2004
                                             1993­2000    2001­June 2004

Number of investigations completed              32              25

Number of restrictive measures                  11              18

Number closed without restrictive measures      21              4

Restrictive measures as percentage of

investigations completed                        34              82

Average number of definitive measures per

year                                            1.4            4.5

Source: WPS3658




                                           46

Table 4 Applied Tariff Rates on Industrial Goods for Selected Latin American Countries
Most-favored nation rates, percentage ad valorem)

                           Simple   Minimum Maximum

Country          Year     Average      Rate      Rate

Argentina        2003        15           0       35

Brazil           2003        14           0       35

Chile            2002         7           0        7

Colombia         2002        13           0       35

Costa Rica       2002         6           0       15

Mexico           2003        18           0       35

Peru             2000        13          12       20

Source: UNCTAD (United Nations Conference on Trade and Development) TRAINS Data base.




                                              47

Table 5 Antidumping Initiations as Notified to WTO, 1995­June 2004: Countries included in the study and major users
                                                                                              Per      Per    Per

                                                                            Jan.­             year     year   year

                                                                            June              1995­    1995­ 2002­

Initiations    1995 1996 1997 1998 1999 2000 2001 2002 2003 2004                     Totals   J2004    2001   J2004

Argentina        27      22     14       8     23   45    26     14     1         7     187      19.7   23.6     8.8

Brazil            5      18     11     18      16   11    17      8     4         6     114      12.0   13.7     7.2

Chile             4       3      0       2      0    5     0      0     0         0       14      3.1    2.0     0.0

Colombia          4       1      1       6      2    3     6      0     0         0       23      2.4    3.3     0.0

Costa Rica        0       4      1       1      0    0     0      0     0         0        6      0.6    0.9     0.0

Mexico            4       4      6     12      11    7     5     10    14         3       76      8.0    7.0    10.8

Peru              2       7      2       3      8    1     8     12     5         3       51      5.4    4.4     8.0

Major users,

for

comparison

United States    14      22     15     36      47   47    76     35    37        21     350      36.8   36.7    37.2

European

Community        33      25     41     22      65   32    29     20     7        13     287      30.2   35.3    16.0

South Africa     16      33     23     41      16   21     6      4     8         4     172      18.1   22.3     6.4

India             6      21     13     27      65   41    79     81    46         4     383      40.3   36.0    52.4

Source: WTO Database




                                                           48

Table 6 Antidumping Measures as Notified to WTO, 1995­June 2004

                                                                                Per   Per   Per   Measures

                                                                    Jan.­       year  year  year  in place as

                                                                    June        1995­ 1995­ 2002­  of June

Measures          1995 1996 1997 1998 1999 2000 2001 2002 2003 2004      Totals J2004 2001  J2004  30, 2004

Argentina           13    20     11      12   9     15     15  24 19    1   139  14.6  13.6  17.6       76

Brazil               2     6      2      14    5     9     13   4  2    2    59   6.2   7.3   3.2       54

Chile                2     0      2       2    0     0      0   0  0    0      6  0.6   0.9   0.0        0

Colombia            1      1      1       0   6      2      0   0  0    0    11   1.2   1.6   0.0        0

Costa Rica           0     0      0       0    0     0      0   0  1    0      1  0.1   0.0   0.4        1

Mexico              16     4      7       7   7      7      3   4  7    3    65   6.8   7.3   5.6       58

Peru                 2     1      2       0    3     4      1   6  7    4    30   3.2   1.9   6.8       29

Major users, for

comparison

United States       33    11     20      16   24    32     33  24 12    6   211  22.2  24.1  16.8      293

European

Community           15    23     23      26   18    41     13  25  3    6   193  20.3  22.7  13.6      165

South Africa         0     8     18      14   35    13      5  15  1    2   111  11.7  13.3   7.2       84

India                7     2      8      22   22    57     38  64 53    6   279  29.4  22.3  49.2      216

Source: WTO Database




                                                        49

Table 7 Countervailing Duty Initiations and Measures 1995­June 2004
                                                                                           Per   Per   Measures

                                                                              Jan.­ Totals Year  Year  in place as

                                                                              June  1995­  1995­ 1996­ of June 30,

               1995     1996     1997     1998   1999     2000 2001 2002 2003 2004  J2004  J2004 J2004 2004

Initiations

Argentina      1        1        1        0      0        0     0   0   1    0      4      0.4   0.4   n.a.

Brazil         0        0        0        0      0        0     1   0   1    0      2      0.2   0.2   n.a.

Chile          0        0        0        0      4        0     0   0   0    0      4      0.4   0.5   n.a.

Costa Rica     0        0        0        0      0        0     0   0   1    0      1      0.1         n.a.

Mexico         0        0        1        0      0        0     0   0   1    0      2      0.2   0.2   n.a.

Peru           0        0        1        0      0        1     0   1   0    0      3      0.3   0.4   n.a.

Measures

Argentina      0        2        0        2      0        0     0   0    0   0      4      0.4   0.5   3

Brazil         5        0        0        0      0        0     0   0   0    0      5      0.5   0.0   0

Chile          0        0        0        0      0        2     0   0   0    0      2      0.2   0.2   0

Costa Rica     0        0        0        0      0        0     0   0   0    1      1      0.1   0.1   0

Mexico         7        0        0        0      0        0     0   0   0    0      7      0.7   0.0   1

Peru           1        0        0        0      0        0     1   0   1    0      3      0.3   0.2   1



                                                           50

Source: WTO Database Note: n.a. = not applicable. Colombia has initiated no countervailing duty investigation. Comparative
numbers for measures in place; United States 57, European Communities 18, Canada 10.




                                                             51

     Table 8 Numbers of Safeguard Initiations and Other Safeguard Actions Notified to the
     WTO by Selected Latin American Countries; 1996­October 18, 2004 (blank spaces imply an
     entry of 0, the 0's wee left out to reduce clutter)


                                                                                     Jan. 1 -­

                                                                                      Oct. 18,  1995­ 2

                                  1996    1997    1998   1999 2000 2001 2002 2003      2004     2000

Argentina

Initiations                                 1       1          1    1                    1        3

Preliminary measures                                                1

Definitive measures applied                 1                  1    1                             2

Negative decision on definitive

measures

Presumed negative                                   1                                            1

Ongoing                                                                                 1

Brazil

Initiations                         1                               1    1                        1

Preliminary measures                1                                    1                        1

Definitive measures applied         1                               1    1                       1

Negative decision on definitive

measures

Presumed negative

Ongoing

Chile

Initiations                                               1    4    2    3                        5

Preliminary measures                                      1    2                                  3

Definitive measures applied                               1    1    1    1                        2

Negative decision on definitive

measures                                                       2    1    2                        2

Presumed negative                                              1                                 1

Ongoing

Colombia

Initiations                                               1                             1        1

Preliminary measures

                                                       52

Definitive measures applied

Negative decision on definitive

measures                                  1                   1

Presumed negative

Ongoing                                                  1

Costa Rica

Initiations                                   1               1

Preliminary measures                          1               1

Definitive measures applied

Negative decision on definitive

measures

Presumed negative                             1               1

Ongoing

Mexico

Initiations                                         1

Preliminary measures

Definitive measures applied

Negative decision on definitive

measures

Presumed negative                                   1

Ongoing

Peru

Initiations                                              1

Preliminary measures

Definitive measures applied

Negative decision on definitive

measures

Presumed negative

Ongoing                                                  1

Seven Countries

Initiations                     1  1  1   2   6  4  5    3    11

Preliminary measures            1         1   3  1  1         5

Definitive measures applied     1  1      1   2  3  2         5

                                        53

Negative decision on definitive

measures                                                      1        2        1     2                         3

Presumed negative                                    1                2               1                         3

Ongoing                                                                                                  3

   Source: Tabulated from information in WTO "Report [one per year 2000-4]of the Committee on
   Safeguards to the Council for Trade in Goods" for 2000 2004
   Note: The actions are tabulated by the year in which each investigation was initiated, for example, if an
   Argentine investigation initiated in 1997 led to a definitive measure put in place in 1998, then that
   definitive measure is included in the tabulation for 1997, not for 1998.
   The countries in the table all notified zero actions in 1995, the first year the WTO notification
   requirement was in force.
   The WTO reports no notification of measure taken or of termination for several investigations that have
   past the WTO deadline for a final determination. We have labeled these "outcomes" as "presumed
   negative."




                                                         54