The Safeguards Clause

The Rational, Operations, and Prospects of GATT Article XIX

Copyright © 1998 Garret Wilson

University of London, School of Oriental and African Studies

MA International Studies and Diplomacy 1998/9

International Economics, Essay 1

December 8, 1998

Ever since Adam Smith published Wealth of Nations in 1776 (Irwin, 75), it has become increasingly evident that relaxing trade barriers generally raises the wealth of the nations involved. As Smith noted, "If a foreign country can supply us with a commodity cheaper than we ourselves can make, better buy it of them with some part of the produce of our own industry, employed in a way in which we have some advantage" (quoted by Irwin, 79). Others such as Torrens, Ricardo, and Mill further explained this comparative advantage, showing in detail how that, if a certain product is available at a lower cost from another country, the resources used locally can be transferred to other products for which the domestic country has an advantage in producing. Any restraints on imports or exports prevent resources from being used efficiently, thus hurting the country using such a trade barriers.

Since the time of these economic philosophers, the amount of trade protection between countries has risen and fallen for a variety of reasons. Although it has been evident for quite some time that multilateral trade liberalization would benefit all countries involved, it was not until 1947 that the General Agreement on Tariffs and Trade (GATT) was created to facilitate the reduction of trade barriers. Besides setting an upper limit on trade restrictions, it introduced the somewhat mis-named Most-Favored Nation (MFN) Clause (Article I), which states that countries may not play favorites; once they lower an import tariff for products from one country, they must lower the tariffs for all countries participating in GATT (Grimwade, 11).

A curious aspect of GATT is Article XIX, "Emergency Action on Imports of Particular Products," or what has become known as the, "Safeguards Clause." This provision essentially states that, if the GATT obligations cause a rise in imports to a certain country, and these imports harm domestic producers of competitive products, that country is allowed to temporarily protect its producers by enacting trade restrictions—the GATT obligations therefore no longer hold. However, the importing country is required to compensate the exporting country by making trade concessions in other areas.

At first glance Article XIX might seem quite peculiar; after all, the very essence of the concept of comparative advantage is that some countries can produce products at a lower price than others. The entire motivation of GATT is to lower trade restrictions so that this comparative advantage can be realized, prices will be lowered, and domestic resources will be transferred to areas in which they are the most efficient. It would seem, therefore, that GATT is all about lowering international prices and shifting resources domestically. Why would the agreement contain an article which seems to say that if the reasoning behind GATT works, the participating countries are free to abandon its principles? Why is there a clause that seems to effectively nullify the rational of the entire agreement?

Reasons for Article XIX

In order to understand the reasoning behind the existence of a Safeguards Clause, it would be wise to refer to the original text, however legalistic and complicated it may seem. The initial section of Article XIX, which provides somewhat of a summary of the entire article, reads as follows:

If, as a result of unforeseen developments and of the effect of the obligations incurred by a contracting party under this Agreement, including tariff concessions, any product is being imported into the territory of that contracting party in such increased quantities and under such conditions as to cause or threaten serious injury to domestic producers in that territory of like or directly competitive products, the contracting party shall be free, in respect of such product, and to the extent and for such time as may be necessary to prevent or remedy such injury, to suspend the obligation in whole or in part or to withdraw or modify the concession. (GATT, Article I:1:a, emphasis added).

The emphasized sections give more insight into the reasoning behind Article XIX: the Safeguards Clause cannot be invoked for any rise in imports that affect domestic producers, but only those imports that unexpectedly ("as a result of unforeseen developments") cause "serious injury" to the domestic producers. Furthermore, GATT obligations can only be suspended temporarily, "for such time as may be necessary" to fix the problem.

The Economical Explanation

The picture drawn by Smith, Torrens, and Ricardo was a very theoretical one. Its conclusions were based upon a few assumptions that might not be (or rather, were seldom) valid in the "real world." One such assumption was perfect factor mobility—the ability of resources, once they are being used inefficiently, to immediately and completely shift to other areas of production. In the "real world," the cold terms, "factors" and "resources" translate into actual individuals who seldom change as immediately and completely as economic theory would indicate.

By way of a hypothetical example, suppose that Japan could produce automobile transmissions much more efficiently than could the US, but this difference in cost was hidden by tariffs imposed by the US on Japanese imports. If GATT obligations caused these tariffs to be lowered substantially, the Japanese transmissions would suddenly become much less expensive than those from the US, resulting in most transmissions being purchased from Japan. This in turn would cause many workers in the US to lose their jobs. In the theoretical world of the classical economists, this would pose no problem: the auto workers would immediately become employed in sectors for which the US had a comparative advantage—designing equipment for orbiting satellites, for instance. To say that this shift in employment from building transmissions to designing satellites would not happen overnight is an understatement.

The economic explanation for the Safeguards Clause, then, is that it provides a way for countries with high levels of protection to adjust to the liberal trade policies brought about by GATT. In the event that unexpected changes brought about by liberalization cause serious harm to domestic producers, these countries are allowed to implement trade liberalization slowly, allowing time for local resources to transfer to more productive areas. That's the standard economic explanation for the Safeguards Clause. But that's not the full story on why Article XIX exists.

The Political Explanation

To begin with, the economic explanation outlined above is dubious at places because the only aspect of production being examined is that of producers. If domestic producers are losing their jobs, the government can charge higher prices on imports (i.e. raise tariffs) so that consumers buy locally. This keeps the price of the product at a higher level than it would otherwise be, and we have so far ignored those who actually pay the higher prices: the consumers.

Import barriers protect domestic jobs by keeping the price of products artificially high. In effect, consumers are paying the wages of the effected employees. Temporary protection would not make sense if the adjustment costs to the society as a whole (consumers included) are more than the gains from trade expansion, and this is often the case (Grimwade, 83). For example, United States consumers have recently been paying $160,000 per year to protect automobile workers when the compensation per worker was less than $40,000 per year (Trebilcock, 169). Furthermore, the use of trade barriers to protect domestic jobs hardly provides an incentive for them to "adjust" unless some sort of definite time limit is established. If workers are still employed, there is little cause for them to seek employment in another area which provides a comparative advantage (Grimwade, 84).

The ultimate causes for the existence of Article XIX may therefore be more political than economical. Citizens are seldom well versed in the principles of comparative advantage and the benefits of free trade, and stories of trade-related layoffs often create images of foreign domination, as is illustrated by the promotion in the US of products "Made in America." Furthermore, producers are usually better organized and more influential than consumers, causing policy-makers to be more concerned about trade impacts on the former (Trebilcock, 167). Since the costs of protectionism are distributed widely, protectionism causes each consumer to lose a relatively small amount; on the other hand, while liberalization ultimately benefits consumers and it has a net benefit to society, its costs are concentrated in a relatively small number of producers, which means each producer has more of an incentive to request protection than a consumer has to request liberalization.

The GATT is an unprecedented agreement liberalizing trade, and at the time it was created many participants had concerns, for the reasons outlined above, about "selling" the concept of trade liberalization at home. If liberalization has the potential of drastically affecting some producers, this could cause political problems, not the least of which was the fact that the GATT itself might not be accepted by that country's constituents. United States policy went so far as to require an escape clause be present in any US trade agreement (Jackson, 179). Therefore, it follows that one of the most important reasons for Article XIX is that the participating parties thought it "better pragmatically to give in to the idea of temporary and limited import barriers for specific (and hopefully not too significant) cases as a way not only to alleviate some of the burdens of adjustment, but also to diminish the pressures for a more drastic departure" from liberalization—effectively using the Safeguards Clause to "buy votes" to ensure GATT's existence (Jackson, 177).

Problems with Article XIX

If Article XIX is therefore to be considered as an effective method for ensuring the participation of countries that have fears of liberalization, offering a way for them to ease trade liberalization pains if the need arises, the picture would be quite rosy indeed. In hindsight, however, it is evident that the Safeguards Clause has some serious problems, perhaps not with the spirit in which it was written, but with the letter in which it was made manifest. Paradoxically, its wording is in many places too ambiguous, inviting exploitation, yet at the same time is structured in such a way as to discourage its use.

Defining a Basis for Use

In examining the particulars of GATT safeguards provisions, perhaps it is best to start with the definition given by Article XIX for the circumstances under which it should be used. The article lists several stipulations which should be met before safeguards are invoked. The most basic stipulation is that a product be "imported... in... increased quantities..."—there must be a rise in imports. While this may seem straightforward enough, Article XIX does not specify whether the increase in imports should be relative or absolute, and consequently it has usually been interpreted to mean a relative increase (Jackson, 188). This means that a decrease in domestic production of a product could cause a relative increase in imports even if the absolute quantity of imports from another country stays the same, or even in some cases decreases!

There is furthermore no time frame indicated for an increase in imports. Since the purpose of GATT is to promote free trade, one would assume that over a long enough time period imports in some sectors would rise. These ambiguities of relative import increases over time could theoretically result in a case where import quantities have remained absolutely static for decades, but a country could suddenly invoke the Safeguards Clause if domestic production were to decline (Jackson, 185).

Of course, not only must imports increase but they must be because of "unforeseen developments" and "cause or threaten serious injury to domestic producers... of like or directly competitive products." The number of terms left undefined is quite large, providing still more ambiguity. What are "unforeseen developments?" What counts as "serious injury?" What does it mean to "threaten" as opposed to "cause" serious injury? How should "directly competitive products" be determined?

The definition of "directly competitive products" governs how easy it is for the Safeguards Clause to be invoked. The product being imported can be defined in broad or narrow terms, with quite different results. If an imported product is classified as, "ladies' party dresses," for example, then such imports could be interpreted to cause injury to a relatively few number of jobs domestically. If, on the other hand, the same imported product is classified simply as, "garments," a rise in quantity of such imports could be construed as harming the domestic production of even farmers' overalls, providing a basis for the invocation of Article XIX.

Not only does the Safeguards Clause apply to "serious injury," which is in itself ambiguous, the article also applies to the "threat of serious injury." If an import is to be considered a "threat of serious injury" to a domestic product, this would mean that the serious injury has not yet occurred, which necessitates, as Jackson notes, that those applying the article try to predict the future (Jackson, 190). A claim that in import will sometime in the future cause harm to a domestic product gives policy makers incredible leeway in applying safeguards.

Perhaps the most ambiguous criteria in Article XIX is that the increase in imports be a result of "unforeseen developments." Presumably, the original intent of this phrase is to restrict application of the article to those instances in which the increased imports were unexpectedly causing problems in the market and industries were having trouble adjusting to market liberalizations. In practice, however, this phrase has been interpreted in a variety of questionable ways, most notably in the so-called Hatter's Fur Case of 1951.

In this instance, the United States claimed that rising imports of hatter's fur from Czechoslovakia were harming domestic producers and proceeded to take action under Article XIX. The reason for the rise in imports was not in dispute: there had recently been a change in the preferred style of women's hats (Trebilcock, 164). The US declared that in this case it was sufficient enough to meet the "unforeseen developments" criteria. As Jackson notes, this broad interpretation could provide an excuse for using Article XIX as a result of any style changes—the width of men's neckties, for example—effectively making the phrase, "unforeseen developments" meaningless (Jackson, 187).

Application of Remedies

After the importing country has determined that it can invoke the Safeguards Clause, it may "suspend the obligation in whole or in part or... withdraw or modify the concession." The "obligation" which the country may suspend is not defined. The first interpretation that comes to mind is that this refers to the specific GATT obligation or concession (lowered tariffs, for example) that is directly causing harm to the domestic product, and this may very well be what the original authors had in mind. Even if this interpretation were spelled out, however, the question of which obligation is causing the problem would still need to be answered, and since, as we saw earlier, even defining the presence of a problem is ambiguous, pinpointing the GATT concession that is hurting the domestic import leaves a broad interpretation to the discretion of the importing country.

The simple reading of "obligations," however, allows even further leeway, in theory allowing the importing country to remove itself from GATT obligations of its choosing, even the Most-Favored Nation obligation of Article I (Jackson, 197). As mentioned earlier, this article captures the spirit of GATT in that, "any advantage, favour, privilege or immunity granted... shall be accorded immediately and unconditionally to the like product originating in or destined for the territories of all other contracting parties."

A broad reading of "obligations" would allow suspension of even this fundamental provision of GATT, resulting in tariffs or other protectionistic policies which target specific countries. Although some have argued that it is appropriate to only apply barriers to products from the country specifically causing the injury, it is counterintuitive to effectively punish countries that are reaping the benefits that liberal trade policy is designed to bring about, namely, an increase in efficiency and thereby an increase in production by free access to products internationally (Jackson, 196).

Furthermore, a broad reading of "obligations" would allow countries to choose its preferred method of protecting the domestic product, allowing the imposition not only of tariffs but of quantitative restrictions as well. Such restrictions, such as quotas, are frowned upon throughout the GATT language for a variety of reasons, including their discrimination against new entrants to the market. Their imposition requires decisions to be made concerning which countries are to receive what percentage of the import market. This in practice usually leads to discrimination among supplying countries (Jackson, 196). The Article XIX wording, by implicitly allowing quantitative restrictions, provides another method of circumventing the MFN ideal.

For providing a time limit on Article XIX actions, "for such time as may be necessary" could hardly be less specific. Once a country evokes safeguard actions, the language of GATT does not specify any time limit on those actions. This detracts from Article XIX as a temporary adjustment mechanism and instead allows for interpretations that can be used for more permanent protectionism.


It would seem at this point that the language of Article XIX is so broad and ambiguous as to promote its use indiscriminately whenever a country would wish to bypass GATT obligations. The latter part of the article, however, specifies that if the contracting parties cannot come to an agreement on what actions the importing country may take, that country may nevertheless take action but the affected country "shall then be free... to suspend... the application to the trade of the contracting party taking such action... such substantially equivalent concessions... of which the CONTRACTING PARTIES do not disapprove."

It is this admittedly convoluted wording essentially says that exporting countries may retaliate with their own counter trade restriction if the importing country invokes Article XIX. To prevent this, it has over the years become customary for the country invoking safeguards to provide a concession to the country or countries affected. This has traditionally been in the form of the importing country lowering of tariffs in other areas (Jackson, 193). This compensation requirement of Article XIX has become one of the largest deterrents against its use.

In the first place, if a country is using Article XIX in the way it was originally intended (i.e. there is a true emergency and imports are unexpectedly hurting domestic producers), the country could reasonably question the need for any compensation. Not only does the requirement of compensation discourage the use of safeguards, the fact that Article XIX requires executive approval makes it difficult for a government to justify safeguard measures to its constituency when compensation is required (Trebilcock, 163).

Another problem appears in GATT's effectiveness: the GATT has over the years succeeded in drastically lowering the average tariff of all participating members. Since application of the Safeguards Clause requires compensation, countries often find it quite difficult to find tariffs in other areas that are high enough that by lowering them can provide adequate compensation to the affected parties (Jackson, 194). If many parties are involved, compensating each of them can require quite a number of concessions.

For these reasons, although the wording of Article XIX is quite ambiguous, the sheer number of criteria for its use coupled with its requirement of compensation has tended to discourage its use. Many countries have instead opted to invoke other protectionistic actions, such as antidumping, countervailing actions, and gray-area measures such as Voluntary Export Restraints (VERs). Between 1950 and 1986 there were only 132 actions taken under Article XIX. In contrast, from 1979 to 1988, 1,833 antidumping actions were taken, 429 countervailing actions were initiated, and there were 261 VERs known to be in existence (Trebilcock, 162).

Antidumping and countervailing actions are less desirable than safeguard actions for many reasons, including their wider allowable range of application and the absence of any sort of time limit on their use. Gray-area measures such as VERs are not only in most cases unregulated, they circumvent GATT by forming unilateral situations which are supposedly "voluntarily" initiated by the exporting country. Gray-area measures not only hurt the world economy as a whole but in many cases hurt the importing country, which initiated the VER by sending signals in some fashion to the exporting country.

Article XIX as it reads is therefore flawed in many respects. The prerequisites for its use are numerous, but they are uniformly vague and open to a variety of interpretations, making many of them useless as adequate restrictions. Its lack of a specified time limit does not provide for the temporary adjustments it was meant for. On the other hand, its compensation requirement has reduced the number of countries using the Safeguards Clause and has instead provided an incentive to seek other more harmful and unregulated forms of protection.

The Uruguay Round and the Safeguards Agreement

To address these shortcomings, the various GATT rounds have attempted to amend Article XIX, although until recently such attempts have been unsuccessful. While the Tokyo Round ended in failure in this respect (Jackson, 209), the recently-completed Uruguay Round, which established the World Trade Organization (WTO), succeeded in responding to many of the inadequacies of Article XIX by the creation of the Agreement on Safeguards.

Perhaps the most important contribution of the new safeguards agreement is the prohibition of all existing gray-area measures (including VERs and similar measures), which are required to be eliminated within four years of the establishment of the WTO. A time limit on the duration of safeguard measures was also introduced, set initially at four years with the option of extending the limit to a total of eight years (Grimwade, 90). The agreement also relaxes the compensation requirements in circumstances which result from absolute increases in imports, as opposed to relative increases (Jackson, 211). These modifications should help prevent the abuse of the Safeguards Clause, while at the same time discouraging the countries from resorting to unregulated forms of protection.

Other areas of ambiguity have been cleared up. It is now required that the importing country conduct a proper investigation before invoking safeguards, and "serious injury" is now defined more clearly (Grimwade, 89). A Committee on Safeguards is now assigned to monitoring countries to ensure that the agreement is carried out (Grimwade, 91). It is now specified that safeguard measures are to be carried out irrespective to the source, upholding the MFN spirit of GATT (Qureshi, 35).

The Uruguay Round's Agreement on Safeguards therefore solves many of the deficiencies of the original Article XIX. Although there are still some areas which should still be improved (notably the lack of restrictions of safeguards to tariffs, which still would allow quantitative restrictions), on the whole the new changes make large strides in removing the aspects of Article XIX which promoted its abuse and discouraged its use. In this regard, the theoretical framework of free trade set forth by Smith, Torrens, and Ricardo has been further realized.

Works Cited