Abstract

Since 1981, major changes have occurred in the pattern of trade of developing countries (Tables A15 and A16). Their share of world exports has declined, reflecting the substantial decline in the value of oil exports that offset the increase in their share of world non-oil exports, including world manufactured exports. The rapid growth of exports of the four Asian newly industrializing economies (NIEs)—Hong Kong, Korea, Singapore, and Taiwan Province of China—stands out in this trend. An increasing proportion of developing countries’ exports of manufactures was directed toward industrial countries, reflecting the continued importance of industrial countries as a market for the products of developing countries.

Trade Trends

Since 1981, major changes have occurred in the pattern of trade of developing countries (Tables A15 and A16). Their share of world exports has declined, reflecting the substantial decline in the value of oil exports that offset the increase in their share of world non-oil exports, including world manufactured exports. The rapid growth of exports of the four Asian newly industrializing economies (NIEs)—Hong Kong, Korea, Singapore, and Taiwan Province of China—stands out in this trend. An increasing proportion of developing countries’ exports of manufactures was directed toward industrial countries, reflecting the continued importance of industrial countries as a market for the products of developing countries.

The ratio of exports to GDP rose during 1981-85 for about half of the 48 developing countries surveyed, comparable to the increase in the 1973-81 period (Table A16). Declines since 1981 were largest among the oil exporting countries, while increases outnumbered falls in African and Western Hemisphere countries, reflecting their response to the debt crisis.

Among the developing countries, the combined share of world exports of the four Asian NIEs has risen steadily from 3 percent in 1973 to 4.3 percent in 1981 and to over 6 percent in 1986 (Table A1). During this period their exports grew at an annual average rate of 17 percent, and at a rate of 10 percent a year since 1981. Exports of other geographical groupings of developing countries declined during 1981-86, at annual average rates of 8 percent in Africa, 2 percent in other Asian developing countries, 5 percent in Latin America, and 17 percent in the Middle East, compared with a growth rate of world exports of about 2 percent.

In contrast to the developments in total exports, developing countries’ exports of manufactures grew in excess of the world rate during 1973-86. As a result, the share of developing countries in world exports of manufactures rose from 7 percent in 1973 to 12 percent in 1986 (Table A15). Most of this increase was attributable to the four Asian NIEs, whose share rose from 4 percent in 1973 to 8 percent in 1986; since 1981 the share of the other developing countries has fallen from 4.5 percent in 1981 to 3.9 percent in 1986 (Table A2).55 Between 1973 and 1985, slightly less than half of the increase in developing countries’ exports of manufactures came from engineering products (including machinery, transport equipment, office equipment, and electrical goods), into which the four Asian NIEs in particular have diversified (Table A17). Many developing countries’ exports of manufactures continued to be concentrated in traditional sectors like textiles and clothing. These sectors, together with other consumer goods, accounted for a further one third of incremental exports of manufactures, Overall, developing countries captured about 14 percent of the increase in world exports of manufactures, with increases above this average in clothing, textiles, other consumer goods, and other semimanufactures.

The decline in export earnings and the financial constraints arising from the debt crisis have led to a decline in the share of developing countries’ imports in world imports between 1981 and 1986 (Table A18). About three fourths of the countries included in Table A16 also experienced a decline in the ratio of imports to GDP during 1981-85, During 1973-81 this ratio had risen in about 70 percent of the cases.

Developing countries’ imports, after growing at an annual rate of 22 percent during 1973-81, declined by 4 percent a year in 1981-86. During the whole period their imports increased at a rate of about 11 percent a year, with imports of manufactures growing at a slightly higher rate than those of primary products. Since 1981, the imports of the African and Middle Eastern developing countries have declined at roughly the same annual rate of between 7 percent and 8 percent, while those of the Latin American countries, after falling by one third during 1981-85, recovered by almost 5 percent in 1986. The imports of the Asian developing countries fell marginally in 1982 but have since grown at a yearly rate of 2.5 percent.

Industrial countries remained by far the major suppliers of the developing countries, accounting for 64 percent of their imports in 1986. This percentage has risen since 1981, in line with the more rapid growth in industrial countries compared with developing countries during 1981-86. In contrast, the importance of developing countries as markets for industrial countries has declined; in 1986, 19 percent of industrial countries’ exports went to the developing countries, compared with 29 percent in 1981.

By region, the share of trade conducted with industrial countries has increased most in the developing countries of Europe and the Western Hemisphere since 1981 (Table A19). Classified by major export, oil exporting developing countries maintained their share of exports directed to industrial countries despite the decline in oil prices since 1981, although the trends in individual countries diverged. However, their share of world exports declined by three fifths since 1981 and their share of world imports by one half (Table A20). Exporters of manufactures, particularly Korea, increased their share of world trade as well as the proportion of their trade conducted with industrial countries in both exports and imports. By financial criteria, the share of the 15 heavily indebted countries in world exports has declined by one third since 1981, partly reflecting terms of trade losses. In imports, their share fell by one half to 3.6 percent in 1987, well below their share in 1973. The proportion of their exports directed to industrial countries increased considerably more than the developing country average, to 72 percent in 1987, Although a rising proportion of their imports originated in industrial countries, the dollar value of their imports from industrial countries has fallen by one third since 1981. Similar trends were apparent in the small low-income countries, which accounted for about 1 percent of world trade in 1986-87.

Trade Policies

The diverse historical and economic backgrounds and recent economic performance of developing economies complicate an overall assessment of their trade policies.56 Some developing economies (e.g., Hong Kong, Malaysia, Singapore, and many African countries) inherited relatively liberal trade regimes at independence; others (e.g., Argentina, Brazil, and many other Latin American countries) have historically maintained highly protective trade regimes. Their growth and development strategies have also varied: some have adopted inward-looking growth strategies, while others have adopted more outward-oriented growth strategies under which they have continued to liberalize their trade regimes.

An assessment of trade policies in developing countries is also complicated by other factors. First, countries undertaking trade liberalization programs usually as a first step replace quantitative restrictions with tariffs; this normally involves an initial increase in tariffs followed by a reduction. Second, customs duties have historically been an important source of government revenue in the early stages of economic development because they are easier to collect than domestic income or consumption taxes when tax administration is weak and tax handles are limited; smaller economies and Asian and African countries depend more heavily on tariffs as a source of revenue than other developing countries (Table A21). Finally, macroeconomic imbalances may result in an increase in trade protection as an alternative to remedial policies to correct the savings-investment balance; in such a situation, an increase in import duties may be a means to reduce the fiscal deficit.57

Tariffs

Statutory tariffs are generally higher in developing countries than in industrial countries, typically ranging from zero to very high maximum rates. A recent study of 50 developing countries, which account for about 15 percent of world trade (average of exports and imports), provides results based on 1985 data.58 It found that the unweighted average rate of tariffs for all products was 26 percent, or 34 percent if other import charges were included.59 The corresponding weighted averages (based on country imports) were 24 percent and 30 percent, respectively. The latter can be compared with less than 5 percent on average for OECD countries. While variations existed among regional groupings, the study reported an inverse relationship between per capita income and tariff levels (Table A22). This inverse relationship is consistent with other studies that indicate that customs revenues become less important as a source of government revenues as the income level increases,60 It is also consistent with other studies that indicate the superiority of outward-oriented over inward-oriented trade strategies in raising income levels.

The structure of tariffs in developing countries is broadly similar to that in industrial countries. Products such as tobacco, beverages, textiles, clothing, manufactures, and certain foodstuffs are subject to above-average duties, while fuels, chemicals, metal and metal products, and minerals and mineral products are subject to below-average tariffs.

Statutory rates tend to be substantially higher than average rates of duties collected (Table A21). The difference between statutory and average levels reflects (i) “duty drawback” schemes that some countries (e.g., Brazil and Colombia) allow on imports of raw materials and intermediate inputs; (ii) similar privileges that some countries (e.g., Brazil and Mexico) offer to attract foreign investment or to promote investments in specific projects or regions; (iii) preferential tariff reductions that a number of developing countries grant each other under preferential trade arrangements;61 and (iv) temporary tariff reductions on a continuous basis on a wide range of products (e.g., Brazil).

In developing countries the proportions of “bound” tariffs are much lower than in industrial countries.62 Only Mexico and Chile have bound 100 percent of their tariff schedules at maximum rates of 50 percent and 35 percent, respectively. For 18 other developing contracting parties for which information is available, the proportion ranges from zero percent to 39 percent with most falling in the 20-25 percent range.

The combination of high statutory tariffs with substantially lower actual average tariffs and a tow level of tariff bindings has implications for the transparency of the trade system and the certainty of trading partners’ access to developing country markets. Average tariffs may be increased substantially through changes in duty remissions and other schemes without amendments to the tariff schedule. Moreover, where tariffs are not bound, statutory tariffs can be increased without legal implications in GATT.

Nontariff Measures63

Developing countries frequently use nontariff measures (NTMs) as a major form of protection. A study of 50 developing countries found that 40 percent of all tariff lines (weighted by economic size) were subject to some form of NTM.64 Excluding NTMs that were applied to all imports, the ratio was 27 percent. Import licensing was found to be the most common form of NTM, although foreign exchange restrictions were the most prevalent in Latin America and the second most frequent use in sub-Saharan Africa. As with tariffs, an inverse relationship was found between per capita income and the frequency of use of NTMs (Table A23).

A significant feature of NTMs of developing countries is that not only are they widespread but also they are stacked, that is, a given product is subject to more than one restriction.65 While foodstuffs are the most affected sector, it is notable that all categories have higher frequency of use of NTMs than most industrial countries. In particular, textiles, clothing and footwear, and iron and steel all had high frequency of use of NTMs (and higher tariffs) despite the apparent comparative advantage of developing countries in these products. In contrast to industrial countries in which there is increasing resort to discriminatory measures, NTMs in developing countries are normally applied on a nondiscriminatory basis.

The GATT provisions on balance of payments restrictions are the most frequently invoked justification for restrictions by developing countries that are contracting parties to GATT. Some 85 percent of quantitative restrictions that have been notified to GATT by 24 developing countries have been justified for balance of payments reasons.

Recent Developments

A trend toward more liberal trade policies is evident in a number of developing countries. This reform is part of wider structural reform efforts taking place, and indicates a growing awareness by these countries of the benefits of outward-oriented policies. In some Latin American countries (notably Bolivia and Mexico), trade liberalization has additionally occurred in the context of anti-inflation programs. Some countries have been able to roll back restrictive measures introduced at the outset of the debt crisis, while the strong external positions of Korea and Taiwan Province of China have permitted these countries to continue liberalizing their trade regimes.

Despite these positive developments, trade liberalization for many countries continues slowly because of inward-looking development policies or has regressed partly because of financial difficulties arising from a high debt-service burden or because of failure to implement domestic policies necessary to improve the trade balance.

Information collected on 31 developing countries for 1985-88 indicates that tariffs were lowered in 13 countries; for the rest, changes were mixed or no information was available (Table A24). In a number of these countries trade reform involving initially the substitution of tariffs for quantitative restrictions, and subsequently a reduction in tariffs, was under way.66 In some countries (such as Indonesia and Thailand) temporary surcharges or temporary increases in tariffs were used as supplementary measures to counter surges in imports; some of these countries continued to rely on quantitative restrictions as the basic mechanism for protection and defense against chronic balance of payments problems. In others (such as Brazil) domestic shortages were countered through temporary reductions in tariffs or temporary surcharges were eliminated when emergency situations no longer prevailed. Trade liberalization measures have also been taken to ease domestic inflationary pressures (such as in Mexico), and tariff reductions on certain products have been used to impose the discipline of world prices on domestic producers.

With regard to nontariff measures, 18 countries moved toward liberalization while 6 moved in the opposite direction. For some countries (Egypt, Korea, Mexico, Morocco, and Taiwan Province of China), the liberalization of quantitative restrictions has been accompanied by a general reduction in tariffs and import-related taxes.67 However, for other liberalizing countries, including those that have been substituting tariffs for nontariff barriers, the liberalization was sometimes accompanied by higher tariffs (Argentina and Bangladesh).

In April 1988, a group of developing countries agreed to set up their own trade preference system—the Global System of Trade Preferences (GSTP)—at a ministerial meeting in Belgrade. The agreement was adopted by 48 countries, including Argentina, Brazil, Egypt, India, Mexico, Nigeria, and Pakistan. The arrangement explicitly excludes large industrial nations, and aims to promote trade between developing countries. The initial impact of the system is not expected to be large; UNCTAD estimates that the Global System of Trade Preferences will cover less than $10 billion of imports.

Trade Policies of Newly Industrializing Economies

Some common features of the four Asian newly industrializing economies are their outward-oriented growth strategies, their relatively poor natural resource bases, and their recent high annual average growth rates. There are, however, also many differences among them. One is that Hong Kong and Singapore are basically free trade ports and have few or no trade or exchange restrictions. Korea and Taiwan Province of China are more complex, and developments need to be reviewed individually.

Korea has made significant progress in liberalizing its import system since 1980, when over 30 percent of tariff code items were listed as restricted imports. By 1983, the share of restricted items had been reduced to 19 percent and, starting in 1984, a major new five-year program of liberalization was launched. As a result, the ratio of restricted items was reduced to less than 5 percent by April 1, 1988.68 Agriculture remains the most heavily protected sector, accounting for over three fourths of the remaining restrictions. To safeguard against import surges, newly liberalized imports may be placed on an import surveillance list, or subjected to adjustment tariffs; however, the use of both procedures has been limited.69 The surveillance list is scheduled to be eliminated by the end of 1988.

The Tariff Act was amended with effect from January 1, 1984. Revisions in the Act aimed to improve the competitiveness of Korean industry and provided for a lowering of tariff rates and a narrowing of their dispersion. As a result, the average unweighted tariff rate was reduced from 23.7 percent in 1983 to 18.1 percent in 1988.

These measures may have been partly offset by the operation of 39 special laws for both agricultural and nonagricultural products, which permit government agencies to regulate their imports. Under the market diversification plan, items can be subject to special import approval if imported from a country that has a substantial trade surplus with Korea. Under the tariff quota plan, tariffs on quota items are increased when a specified quota ceiling is reached. The special laws are being reviewed, and steps are being taken to streamline their application and reduce the extent to which they serve as unnecessary barriers.

In Taiwan Province of China, high tariffs have been the main barrier to imports. Since the early 1980s tariffs have been cut and the proportion of imports subject to import licensing has been reduced. The average nominal tariff rate fell to 23 percent in 1986. In 1987 further tariff cuts were implemented affecting 40 percent of items and reducing the average nominal tariff to 20 percent. In 1988 tariff cuts averaging 50 percent were applied to about 3,500 items. At end-1986, about 20 percent of Taiwan Province of China’s imports were subject to non-automatic licenses; other nontariff measures included embargoes.

Countertrade70

The use of countertrade increased in importance in developing countries in the early 1980s but appears to have stagnated since 1984 and to have declined in 1987.71 Countertrade has been utilized as an export promotion tool, and as a way of overcoming shortages in foreign exchange and protectionist barriers in industrial countries. For instance, countertrade can be used to gain a larger share of a global quota, but it cannot gain greater access to markets protected by voluntary export restraints. Some developing countries may also have used it to counter the effects of overvalued exchange rates, in which it functions as an export subsidy. It may have also been used to secure the transfer of technology, to increase the domestic levels of production and employment, or to export commodities for which international cartel-type agreements exist.72 Latin American and African countries have also explored countertrade as a mechanism for intraregional economic cooperation.73

There are no international laws or agreements dealing directly with countertrade, but GATT provisions cover certain aspects of countertrade practices like subsidization, dumping, or discrimination.74 While, the Fund has no jurisdiction over countertrade unless exchange restrictions are involved (such as in bilateral clearing arrangements), it is generally concerned with the increased use of countertrade because it undermines the multilateral character of the trade and payments system and imposes additional costs on the participants. The complexity of countertrade arrangements and the involvement of intermediaries increase transaction costs. Furthermore, such arrangements lack price transparency, raise arbitrage possibilities, and reduce flexibility, all of which contribute to increased welfare costs. Countertrade practices may entail many of the restrictive and discriminatory practices traditionally associated with bilateralism.

The extent of countertrade is difficult to gauge because trade data are not differentiated according to the source of financing, and because countertrade often involves military purchases for which data are not always available.75 The OECD has estimated that a maximum of some $80 billion or 5 percent of world trade occurred through countertrade arrangements in 1983. This estimate excludes trade under bilateral payments arrangements and trade among Eastern European countries.76 Including these, the total would rise by 9 percent to 14 percent of world trade.77 The share of trade that occurs under documented countertrade agreements is highest between Eastern European countries and both developing and industrial countries, and among developing countries.

Estimates made by various bodies indicate a sharp growth of countertrade between 1980 and 1984, followed by stagnation and a decline in 1987. In 1987, the number of countertrade agreements signed decreased by about 45 percent. The trend toward more open export credit and cover policies since 1985 may have reduced countertrade transactions. The high transaction and opportunity costs of countertrade agreements may also have contributed to this decline.

Countertrade has normally involved raw materials, particularly oil, but also cereals, textiles, and clothing. The use of oil in countertrade continues, albeit at a reduced rate, despite a 1985 decision by the Organization of Petroleum Exporting Countries (OPEC) to phase out their use of countertrade agreements. Among OPEC countries, Indonesia, the Islamic Republic of Iran, Iraq, Libya, and Saudi Arabia have been involved in countertrade. Indonesia, which has legislation on countertrade, signed about 75 percent fewer agreements in 1987 than in 1983, its peak year for such agreements.

Other Asian countries that have used countertrade to varying degrees are China, India, Korea, Malaysia, Pakistan, the Philippines, Singapore, Taiwan Province of China, and Thailand. Much of their countertrade is with other developing countries, although Korea, Malaysia, Taiwan Province of China, and Thailand have also used it to increase their trade with centrally planned economies, and China has used it in trade with Western economies.

In Latin America, Argentina has used countertrade in its trade with centrally planned economies and in its purchases of natural gas from Bolivia. Brazil’s use of countertrade has declined in importance in recent years. Although Colombia, Ecuador, and Mexico have regulations relating to countertrade, such trade is not mandatory and has declined in recent years. In Africa, a number of countries, including Ghana, Nigeria, Uganda, and Zimbabwe, have engaged in countertrade in an attempt to save foreign exchange reserves.

Trade and Trade-Related Industrial Policies Affecting Developing Countries

The trade and industrial policies of industrial countries—the major markets for developing countries’ exports—affect the market access of developing countries and thus have an important impact on them. A number of developing countries argue that trade-restricting measures by industrial countries hinder their integration into the world economy and their resolution of debt problems. Some developing countries that have recently liberalized their trade regimes or continued their liberalization efforts either have faced increased barriers abroad or existing barriers have become binding as exports expanded. For instance, Mexico’s non-oil exports, which nearly doubled over the two years up to 1987, have been increasingly subjected to antidumping and countervailing duty investigations, and existing barriers have become a binding constraint on export expansion.78 Domestic political support for liberalization measures is thereby reduced. Some middle-income developing countries have also indicated that their exports to developing countries are adversely affected by their inability to match the grant element of mixed credits extended by industrial countries.

Generalized System of Preferences

Industrial countries grant tariff preferences to developing countries under the Generalized System of Preferences (GSP).79 Trade preferences are also granted to selected developing countries under various regional trading arrangements.80

Currently 20 OECD countries and a number of Eastern bloc countries operate GSP schemes, with more than 140 beneficiaries. In 1986 $36 billion of exports from developing countries received preferential treatment by OECD countries, compared with $25 billion in 1980 and $10 billion in 1976, when GSP schemes came into full operation (Table A25). The growth of imports under the GSP slowed down significantly during 1980-86, to an annual rate of 6 percent, compared with an annual average growth rate of 21 percent during 1976-80, largely reflecting the slowdown in the growth of total and dutiable exports from the beneficiary countries. The ratio of imports accorded GSP treatment to total imports of OECD countries from beneficiary countries continued to increase, and by 1986 the ratio reached 15 percent, compared with 8 percent in 1980 and 7 percent in 1976.

Under most GSP schemes, “sensitive” items such as textiles, clothing, and footwear are excluded from preferences, while others, such as certain petrochemicals, receive limited GSP coverage. For example, in the EC some 140 sensitive products, such as methanol, are subject to either GSP ceilings—where the most-favored-nation tariff can be reintroduced at the request of the domestic industry once the ceilings are reached—or GSP tariff quotas—where the most-favored-nation tariff is automatically reintroduced when the quota level is reached. In the case of EC petrochemical imports from Middle Eastern exporters, the quota level is typically reached early in the year. The textiles and clothing sectors, the products of which represent about 17 percent of all industrial tariff lines, account for about half of industrial products excluded from all GSP schemes taken together.81

The GSP schemes of the EC, Japan, and the United States differ in their country and product coverage. For example, in fuels and petrochemicals, refined oil products (such as gasoline and jet fuel oil) enter the EC duty free, as do most first-stage processing products based on oil or natural gas. About 5 percent of the Community’s annual consumption of other petrochemicals is imported. Most of these imports enter duty free under free trade or preferential agreements with EFTA or the Mediterranean countries; some of these imports are from the Gulf Cooperation Council countries, either under GSP ceilings or tariff quotas or subject to MFN tariff rates. Japan, by contrast, does not include refined oil products or petroleum gases in its GSP scheme. However, for many petrochemical products, such as methanol, the Japanese scheme features a global amount by product of about 10 percent of imports, distributed on a first-come, first-served basis. The U.S. scheme excludes OPEC countries (except Ecuador, Indonesia, and Venezuela) and those above a certain per capita income. However, petrochemicals are generally included and are given duty-free access subject to certain product graduation requirements.

In recent years, efforts to improve GSP schemes have been offset to some extent by a reduction in benefits in some schemes, including those of the EC (see Appendix I) and the United States. Some schemes have introduced lower margins of preference, stricter limits on the amount of preferential imports, and differential application of preferential treatment among beneficiaries, including pro duct-specific graduation, and, more recently, country-specific graduation related to income levels. In the latter, countries can be graduated from schemes on the basis of “competitive needs” criteria. The major beneficiaries (Hong Kong, Korea, and Taiwan Province of China) have been the countries most affected by these changes. The least developed countries receive more preferential treatment than other developing countries in most GSP systems. Generally this treatment implies zero tariff rates for least developed countries where other developing countries pay duties at some non-zero preferential rates.

Tariff Peaks and Escalation

Tariff protection in industrial countries in the form of peaks and escalation tends to affect developing countries especially heavily because it is concentrated in traditional export sectors, such as textiles and clothing, where the share of developing countries in the imports of industrial countries tends to be high. Further, high tariff items often have lower GSP coverage than low tariff lines (Table A26) and tend to have an overall higher incidence of nontariff measures.82

The trade effects of tariff peaks appear to apply largely to labor-intensive and resource-intensive sectors, suggesting that the industrial policies of the developed countries in these areas could have significantly influenced resource allocation in the developing countries. It is estimated that, disregarding nontariff measures, imposing a 10 percent MFN ceiling on the tariff rates of the industrial countries would increase their total imports from developing countries by about 1.5 percent (equivalent to almost 16 percent of trade covered by liberalization); the elimination of all tariffs on an MFN basis by the industrial countries would lead to an overall trade expansion for developing countries of some 5 percent.83 Much of the expansion would come in the area of textiles and clothing, although food processing and miscellaneous manufactures would also benefit. These results are derived from a static model and could well be underestimates once the dynamic effects of improved resource allocations are taken into account.

Tariff escalation introduces trade biases, favoring the importation of raw materials rather than processed items. Owing to escalation, the effective protection afforded to processing industries in most developed countries through tariffs can be higher than the tariffs indicate. This is exacerbated by the fact that in many instances nontariff barriers also escalate from raw to processed materials; for example, the incidence of non-tariff barriers on the importation of raw cotton is low, if not zero, in most developed countries whereas cotton fabrics from low-cost suppliers are subject to the Multi-fiber Arrangement.

The trade bias effects of tariff escalation are most pronounced in areas such as tropical foodstuffs and fabrics and, to some extent, in certain petrochemicals, where developing countries are the major source of the raw supplies.84 This indicates that reduced tariff escalation could yield an increase in developing countries’ exports of processed products, particularly as studies show that import demand elasticities in the industrial countries increase with fabrication.85 If their costs remain competitive and they have access to the distribution chain, the initial expansion of exports would probably be centered on those developing countries that already have the relevant manufacturing skills, such as Brazil, and a number of Middle Eastern and Asian middle-income countries that have a high concentration of the required raw materials. Further, in a number of other countries heavily dependent on tropical exports and with a high population base, improved access to industrial countries could trigger investment plans for the establishment of processing facilities, with consequent positive employment effects.

Employment may not be such an important consideration for some Middle Eastern countries, which might elect to invest financial reserves in oil refining and petrochemical production and distribution systems in industrial countries. Some diversification into offshore refining and distribution facilities has already taken place. Thus, Saudi Arabia recently purchased, for some $1 billion, a share in a number of U.S. refineries and access to gasoline stations. Such diversification (including into petrochemical facilities) serves a number of functions. First, it provides the producers of the primary product with a secure outlet for all or part of their output. Second, it serves as a hedge against fluctuating prices: crude oil prices generally fall more rapidly than the prices of refined or petrochemical products and thus, when oil prices are low, and prices of petrochemical feedstocks (oil and natural gas) are low in consequence, the offshore production and distribution investments will be relatively profitable, particularly as the cost of transporting oil is low compared with the cost of conveying petrochemicals, Third, diversification into offshore facilities is a hedge against protectionist measures on refined and petrochemical products by importing countries.

Sugar is an important example of tariff escalation, with the average tariff for imports of sugar preparations into industrial countries being about 20 times higher than that for the raw product, giving considerable effective protection to processing plants in the developed countries. It is indicative in this respect that well over 90 percent of developing countries’ exports in the sugar category are in the form of the raw material. These exports are hindered, however, by high nontariff barriers in many industrial countries, particularly the EC, Japan, and the United States, all of which maintain some form of price-support mechanism to protect domestic growers. These supports have encouraged the production of nonsugar sweeteners, such as high fructose corn syrup, which would not be profitable at the current free market price of sugar. As such, developing countries could gain significantly from a reduction of nontariff barriers, to improve their market access to the industrial countries for the raw product.

Countervailing and Antidumping Duties

Industrial countries have had extensive resort to countervailing and antidumping investigations on manufactured products, often against those from developing countries. Since 1981, almost 50 percent and 30 percent of the countervailing and antidumping investigations, respectively, were on products from developing countries. Some 34-41 percent of investigations on developing countries’ exports resulted in a negative finding of no dumping or subsidization. Evidence suggests that investigations are sometimes used to harass foreign exporters and that such actions are used as substitutes for safeguard measures.86 Further, the initiation of such cases can lead to price undertakings (not to export below a certain price) and to voluntary export restraints, which have negative trade effects. For example, as a result of antidumping investigations, six exporting countries agreed in November 1987 to observe a minimum price on their urea exports to the EC; for Kuwait this undertaking covered 46 percent of its 1986 petrochemical exports to the EC. Saudi Arabia did not agree to such an undertaking, and the EC imposed an antidumping duty of 46 percent on EC imports of urea from Saudi Arabia, covering 11 percent of Saudi Arabia’s 1986 exports of petrochemicals to the EC.

The above factors, together with their negative trade effects, suggest the need for a restrained use of countervailing and antidumping mechanisms and for very clear international disciplines in the area; only if the adverse effects of subsidies and/or dumping are severe and sudden would temporary protection against them be justified on social welfare grounds.87

Nontariff Measures

The incidence of nontariff measures applied by industrial countries on their nonfuel imports appears to fall somewhat more heavily on the exports of developing than of industrial countries; in 1987 the coverage ratios were some 25 percent and 21 percent, respectively, for the nonfuel imports of industrial countries from developing and industrial economies.88 Exports of manufactures by developing countries are particularly affected by such measures, with the highest concentration in steel, textiles and clothing, and footwear. Voluntary export restraints are common in these sectors.

Including the bilateral export restraints concluded under the MFA, a considerable portion (up to 50 percent) of world trade in textiles and clothing is managed and thus not subject to the normal forces of international trade. This affects developing countries particularly, as the preponderance of intra-developed-country trade is not subject to export restraint arrangements. Since 1968, upward of 30 percent of world trade in steel has come under VERs, affecting exports from nearly all existing third-country suppliers to the United States, the EC, and, most recently, Australia, as well as exports from the EC to the United States. Exports of agricultural and food products are also restrained by VERs, mainly from the more efficient producers, such as Argentina, to the EC. In automobiles and transport equipment, as well as in electronic products, Japanese and Korean exporters limit their sales to both the EC and the United States, while in footwear a number of OECD markets are protected by VERs with Korean and other developing country exporters.

Voluntary export restraints divert trade as they can create additional opportunities for some existing suppliers and new entrants. Typically, VERs are “leaky” when first applied in that they do not cover all sources of supply and/or allow substantial growth rates in exports (as has occurred in some bilateral agreements negotiated in the past under the MFA) from restrained suppliers to the protected market. As the domestic price of the product subject to the VER is likely to move above the world price, “leakiness” will provide an incentive, inter alia, for (i) new entrants to invest in capacity to export to the affected market; and (ii) restrained exporters to circumvent their constraints by exporting via third countries, including by establishing capacity in those countries.

The evidence suggests that U.S. voluntary export restraints have been relatively “leaky,” in contrast to those of the EC. Thus, from 1980 to 1986 import penetration (imports as a percentage of domestic production plus imports minus exports) of the U.S. market increased from 4.6 percent to 7.9 percent for textiles, from 14.5 percent to 25.4 percent for clothing, and from 16.3 percent to 23.1 percent for steel.89 These higher ratios include increases in market shares especially by new entrants from developing countries. By contrast, in the EC(10), the import penetration ratio for steel rose only marginally from 1980 to 1986;90 and increases in import market shares since the late 1970s for clothing and footwear appear to have gone mainly to other industrial, particularly EC, country partners.91 In the absence of known voluntary export restraint protection prior to late 1985, “leakiness” cannot be tested for Japan.

Over time, the trade-diverting effects of VERs tend to create pressure to broaden their country coverage. Thus, the United States negotiated VERs in 1984 with a number of new entrants from developing countries into its steel market, and the MFA has progressively become more restrictive. The diversion effects also cause VERs to spread across importing countries, resulting in globally managed market-sharing arrangements, as is now largely the case for steel and textiles and clothing. The market share begins to depend as much, if not more, on negotiating strength as on comparative advantage, and investments can come to rely on the retention of managed trade for their profitability. These factors fundamentally change the nature of competition in an industry: voluntary export restraints provide some real advantages to early and established producers, particularly in the form of rents and consequent resources for product upgrading and diversification; late starters can be quickly drawn into a VER network, as has happened with the MFA, perhaps inhibiting development on the basis of comparative advantage and protecting thereby the position of the earlier entrants.92 All these factors make VERs difficult to dismantle as well.

With the proliferation of nontariff measures, liberalization of trade by OECD countries would most profoundly affect trade now covered by restrictive agreements. By one estimate, removal by industrial countries of all barriers on imports from developing countries would increase their exports of textile and clothing by some 125 percent, of steel by about 62 percent, and of footwear by approximately 85 percent.93 The benefits would accrue mainly to those developing countries, such as Brazil and Korea, that already have well-established manufacturing sectors. Other studies bear out these results.94 Other developing countries would also benefit, however, especially in areas such as clothing where many Asian countries hold a comparative advantage. These are estimates of static effects only. A dynamic estimate would be higher as it would include the effects of increased opportunities for economies of scale, product differentiation, and specialization.

The agricultural policies of the industrial countries also have an important impact on developing countries, including a number that are highly indebted. These policies have distorted trade through domestic and border measures (see Section V). Domestic measures have encouraged surplus production in industrial countries, reducing world prices and the markets for agricultural products, thus depressing the incomes of exporters of agricultural products, including many developing countries. These effects have been exacerbated by subsidized exports of surplus production, which in turn has led to trade disputes and the adoption by industrial economies of additional farm support measures to safeguard the interests of their own producers. Domestic measures have been accompanied by restrictions on market access through border measures, including quotas and variable levies, and other tariff and nontariff measures protecting the markets for diverse agricultural products in almost all industrial countries.

The agricultural policies of the industrial countries—together with the pricing policies of many developing countries—have contributed to a sharp decline in the share of developing countries’ exports in world agricultural exports since the early 1960s. The decline is particularly marked for food products that are produced by industrial countries (Table 5). A recent study demonstrates that a relatively modest liberalization of agricultural trade, resulting in a 10 percent increase in world agricultural prices, would increase the income of developing countries by $26 billion (in 1985 dollars),95 an amount exceeding bilateral grants and loans under official development assistance. Fully half of this increase would accrue to the highly indebted developing countries (see Section V).

Table 5.

Export Shares of Food Products

(In percent)

article image
Sources: World Bank, World Development Report, 1986; and International Monetary Fund, International Financial Statistics.

Relative to export unit value for all products.