For the latest thinking about the international financial system, monetary policy, economic development, poverty reduction, and other critical issues, subscribe to Finance & Development (F&D). This lively quarterly magazine brings you in-depth analyses of these and other subjects by the IMF’s own staff as well as by prominent international experts. Articles are written for lay readers who want to enrich their understanding of the workings of the global economy and the policies and activities of the IMF.


For the latest thinking about the international financial system, monetary policy, economic development, poverty reduction, and other critical issues, subscribe to Finance & Development (F&D). This lively quarterly magazine brings you in-depth analyses of these and other subjects by the IMF’s own staff as well as by prominent international experts. Articles are written for lay readers who want to enrich their understanding of the workings of the global economy and the policies and activities of the IMF.

ARE LOW, uniform tariffs part of the solution to improving competitiveness? Policymakers in many developing countries seem to think so. They are beginning to put in place reforms aimed at simplifying tariff structures and making them more uniform, as well as at bringing tariff levels more in line with those of industrial economies.

For years policymakers in many developing countries felt that balance of payments problems could be fixed by imposing tariffs that were far above those prevailing in industrial countries and that had a wide variety of rates. In theory, high and dispersed tariffs can fulfill certain government objectives—increased revenue, income distribution, and protection of domestic industries—if better, trade-neutral domestic taxes or subsidies are difficult to use. In practice, however, this type of tariff structure tends to distort the allocation of resources, creates an anti-export bias, inhibits the use of foreign technology, and reduces the comparative advantage that many labor-intensive developing countries could exploit to improve income distribution. Other, indirect costs—such as preferential treatment for certain products and industries, and smuggling—may only exacerbate the situation. By the late 1970s, many of these disadvantages surfaced and it became clear that high and widely dispersed tariffs were themselves part of the problem.

Case for uniformity

To be most effective, taxes should be broadly based, that is, cover as many taxpayers as possible, and be set at relatively low rates. Since tariffs are essentially a tax on foreign trade, the case for them is no different. Thus, policy prescriptions on tariff reform made under World Bank- and IMF-supported programs have emphasized reductions in both the average tariff level and the range of tariff rates. Indeed, the eventual goal is to reach what is called a uniform effective rate—that is, a structure with one or very few tariff rates. Ultimately, the objective is to achieve uniform effective protection, or equal protection to value added in all domestic industries, which would take into account tariffs on both outputs and inputs. This goal, however, is seldom achieved, and a uniform nominal tariff structure serves the purpose. It generally approximates the theoretical target and has other practical advantages as well.

A uniform structure responds to many of the problems developing countries in general experienced under earlier trade regimes. Resources get misallocated under a structure that has very high tariffs on consumer goods and low tariffs on raw materials and intermediates. This arrangement protects domestically produced consumer goods, encouraging domestic resources to move away from exports toward producing inefficient import substitutes. Thus, the import tax effectively acts as an export tax. A low, uniform tariff with few exemptions minimizes this anti-export bias and, by improving the allocation of resources, increases a country’s competitiveness. It also reduces the possibility of misclassifying products, thereby discouraging importers from shifting goods from a high- to a low-tariff category.

A uniform structure offers developing countries other important advantages. It eliminates cumbersome paperwork and is easier to administer than more complex systems, especially if customs departments function poorly or are constrained by a lack of resources. It dissuades interest groups from lobbying to secure greater protection for themselves, because any increase one group secures is automatically available to others.

Finally, uniformity responds to a growing skepticism about the merits or feasibility of governments picking winners and losers. For example, the infant industry argument—often invoked to justify selective protection of industries—requires governments to identify industries that could become competitive in the long run but require some initial protection. For this to work, governments must maintain appropriate levels of protection, remain immune to lobbying from special interest groups, and be able to reverse the protection once the infants have “grown up.” However, experience has shown that this process can seldom be carried out successfully. If some protection is believed to be necessary, the easiest way to do it is to cover all threatened industries with a few standard tariffs.

A sample of reformers

Since the mid-1980s, a number of developing countries have undertaken far-reaching reforms of their tariff structures, in the hope that tariff reform would lead to greater international competitiveness and improved export performance. The list includes several Latin American countries (Bolivia, Chile, Colombia, Mexico, and Uruguay) and Ghana. Other countries in Asia (Bangladesh, India, Pakistan, and Sri Lanka) and Africa (Egypt, Kenya, and Uganda) are beginning to do the same. For all of these countries, reducing or eliminating tariffs has been a key component of trade reform, which in turn has been an important element of overall structural reform.

How have the recent reformers fared as they have moved from high and widely dispersed rates toward uniformity? To answer this question, a sample of six countries was studied; some of them have successfully completed the current phase of their reforms (Colombia and Ghana), and for others the process is ongoing (Bangladesh, Brazil, Egypt, and the Republic of Korea). The reform of quantitative restrictions, which were often the binding constraint in the pre-reform period, was undertaken prior to or at the same time as tariff reforms. The two early reformers (Colombia and Ghana) eliminated quantitative restrictions before reforming tariffs.

In all cases, the basic cascading structure, which involves higher tariffs on goods at more advanced stages of processing, was retained after the reform process. Thus, effective protection for final goods has remained greater than that for intermediate and capital goods. But because pre-reform maximum rates on final goods were substantially reduced, the overall effective level of protection—relative to intermediate and capital goods—has been reduced.

Pace of reform. There was a wide range in the pace of reform, with Colombia adopting the “big bang” approach, implementing its reforms in two years, whereas Ghana implemented its reforms gradually over a 9-year period. In the other sample countries, the pace of reform was much slower, although international considerations, including the Uruguay Round trade negotiations, did help quicken the pace for all.

Degree of differentiation. Governments in the sample considered here chose minimally differentiated tariff structures in order to reconcile the need to simplify and unify tariffs with the desire, in some cases, to continue to use tariffs to pursue multiple objectives. The four countries for which data are available (Bangladesh, Brazil, Colombia, and Egypt) had more than 20 different rates prior to the reforms; these were reduced to between 4 and 10 (Table 1). For the most successful reformers (Colombia and Ghana), the target number of bands was about 4—close to the number used in the value-added tax (VAT) structure favored by many industrialized European countries.

Table 1.

Simplifying the tariff maze

(percent, unless otherwise specified)

article image
Sources: Trade Policy Review Mechanism reports. General Agreement on Tariffs and Trade (GATT); World Bank; and IMF.Note: “Reform” data refer either to the current period or to the target period, n.a.: Data not available.

Although no explicit commitment has been announced, Brazil should eliminate all surcharges and fees in order to be compatible with MERCOSUR Common External Tariff.

Higher rates apply to selected products.

Simple average of nominal tariffs, except for pre-reform period in Egypt (weighted average).

Special surcharges. Prior to the reforms, the countries in the sample had levied a number of exclusive or discriminatory charges on imports, resulting in wide variations in protection; for example, Brazil had 11 and Egypt 5 extra charges. Often these charges were applied at different rates, and each could have its own separate list of exemptions. Bangladesh had three separate rates for its sales tax on imports, and Ghana levied its super sales tax on imports at rates ranging anywhere from 10 percent to 500 percent. Such charges on imports were either eliminated during the reforms or assimilated into the new tariff.

Transparency. The elimination of other charges on imports not only reduced the level of protection but also simplified the tariff structure and made it more transparent. Converting specific into ad valorem rates also helped transparency, as did the elimination of arbitrary exemptions.

Maximum rates The sample countries tended to reduce the maximum rates substantially, often from very high initial levels (400 percent in Bangladesh, 200 percent in Colombia, and 160 percent in Ghana) to as low as 10-25 percent. Reducing the maximum rate, especially from very high initial levels, is less harmful for fiscal revenue than is commonly believed. In some instances, the revenue effect can be positive because lower tariffs tend to increase import volumes and reduce smuggling, evasion, and lobbying for exemptions.

Minimum rates. Two countries in the sample, Egypt and Bangladesh, also raised their minimum rates. A minimum tariff usually raises fiscal revenues substantially and reduces the spread of effective protection, especially when it is levied on intermediate and capital goods used to produce final consumer goods. The most successful reformers—Colombia and Ghana—among the sample countries aimed for minimum tariff levels of 5-10 percent. Overall, average tariff levels dropped substantially, with Colombia’s, Ghana’s, and Korea’s falling to the lowest levels—10-17 percent But these averages understate the true extent of the reduction, as they do not take into account the decreases in or elimination of other charges on imports or nontariff barriers.

Announcements. Four countries in the sample, including Colombia and Ghana, announced in advance their plans for tariff reduction (Table 2). Preannouncement, especially when the reforms are implemented gradually, helps investors and producers make decisions based on the target tariff structure; it also reduces costs arising from uncertainty about future tariffs and hence about long-run price signals.

Table 2.

The reform process at work

(percent, unless otherwise specified)

article image
Sources: Trade Policy Review Mechanism reports. General Agreement on Tariffs and Trade (GATT); World Bank; and IMF.n.a.: Data not available.

In the case of Bangladesh, the classification is primary, semiprocessed, and processed products; figures are simple averages of applied tariffs; for Korea, figures are for industrial tariffs.

Higher rates apply to selected products.

Tariff band allocation

How did developing countries allocate goods among tariff bands? The record shows that allocation was generally guided by three objectives: income distribution, protection, and export promotion. For purposes of income distribution, governments assign consumer and “luxury” goods to the highest tariff band and essential goods (such as food and medicine) to the lowest. This is often rein forced by protectionist interests in consumer goods industries or by government policies that foster such industries.

Moreover, government objectives often determine how intermediate and capital goods are treated. If governments want to foster a domestic capital goods industry, they generally place high tariffs on capital goods imports. This action effectively reduces the level of protection for the industry using the protected capital or intermediate goods as inputs. This can penalize exports by raising the cost of imported inputs and by creating incentives for producers to shift out of exports into the protected import substitutes.

To minimize this anti-export bias, all six countries in the sample employ duty-drawback (or equivalent) schemes (Table 2), under which exporters do not have to pay the tariffs on imported inputs used in export production. In principle, these schemes can have positive effects, but they can also raise difficulties. Administrative costs (including leakages and fraudulent claims for drawback) make them expensive to implement, especially when import tariffs are high. In addition, exports covered by these schemes may expand at the expense of other exports, and the schemes themselves may become a surrogate for more wide-ranging import liberalization.

Supporting reform

The value of tariff reforms risks being sharply undercut in the short term if the reforms are not simultaneously accompanied by macroeconomic reforms to compensate for any loss of fiscal revenue and the pressures on domestic producers that may arise from increased imports. In this regard, all of the sample countries undertook the necessary fiscal and balance of payments adjustments either before or during tariff reform.

The sample countries faced serious fiscal constraints as a result of the changes in tariff structures and levels because they had relied heavily on tariffs for revenue and needed time to develop alternative domestic taxes. (In the pre-reform period, tariffs accounted for between 25 percent and 35 percent of total tax revenue in Bangladesh, Colombia, Egypt, and Ghana. These shares declined in the postreform period.) The most successful reformers—Colombia and Ghana—were able to overcome this, and the fiscal deficit-to-GDP ratio in these countries dropped significantly during the reform period, falling from 2.4 percent to 0.9 percent in Colombia and from 6.5 percent to 0.4 percent in Ghana. In Colombia, fiscal action consisted of complementary domestic tax reforms that expanded the tax base; in Ghana, currency depreciation improved the fiscal situation. Bangladesh has also been able to accelerate the pace of reform by implementing a broadly based value-added tax.

On the balance of payments front, in all the sample countries, high tariffs had been used explicitly in response to payments problems. However, corrective macroeconomic policies (including more flexible exchange rates) helped give some breathing space during the reforms. For instance, the depreciation of the real effective exchange rate helped the external accounts in Bangladesh, Colombia, and Ghana.

What next?

Moving away from the sample, if we look at developing countries in general, international considerations, including regional integration initiatives, not only have apparently influenced the pace and scope of reforms but also have continued to shape the further steps that these countries might take. In Latin America, for example, tariff reform has been determined in the context of regional arrangements, such as the Caribbean Community (CARICOM), the Andean Pact (made up of Bolivia, Colombia, Ecuador, Peru, and Venezuela), and the Southern Cone Common Market (MERCOSUR, consisting of Argentina, Brazil, Paraguay, and Uruguay). These arrangements play an important part in the negotiations between regional partners on setting the common external tariff.

To ensure that reforms cannot be reversed because of domestic pressures, a number of countries have chosen to join the General Agreement on Tariffs and Trade (GATT) and “bind” their tariff structures, which represents an international commitment not to raise tariffs without consulting and compensating their trading partners. As a result of the recently concluded Uruguay Round agreement, developing countries will have bound a high proportion of their tariffs, as well as reduced them in industrial and agricultural sectors. However, because current tariff levels—even Colombia’s and Ghana’s—are still higher than those prevailing in several industrial countries, rates will no doubt need to be reduced further, unilaterally or through international negotiations. Ultimately, if countries decide some assistance should be provided to domestic industries, tariffs could be replaced by subsidies, which are easily identified and have more clearly visible costs.